Adjunctive Host-Directed Therapy With Statins Improves Tuberculosis-Related Outcomes in Mice

Abstract

Background

Tuberculosis (TB) treatment is lengthy and complicated and patients often develop chronic lung disease. Recent attention has focused on host-directed therapies aimed at optimizing immune responses to Mycobacterium tuberculosis (Mtb), as adjunctive treatment given with antitubercular drugs. In addition to their cholesterol-lowering properties, 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors (statins) have broad anti-inflammatory and immunomodulatory activities.

Methods

In the current study, we screened 8 commercially available statins for cytotoxic effect, anti-TB activity, synergy with first-line drugs in macrophages, pharmacokinetics and adjunctive bactericidal activity, and, in 2 different mouse models, as adjunctive therapy to first-line TB drugs.

Results

Pravastatin showed the least toxicity in THP-1 and Vero cells. At nontoxic doses, atorvastatin and mevastatin were unable to inhibit Mtb growth in THP-1 cells. Simvastatin, fluvastatin, and pravastatin showed the most favorable therapeutic index and enhanced the antitubercular activity of the first-line drugs isoniazid, rifampin, and pyrazinamide in THP-1 cells. Pravastatin modulated phagosomal maturation characteristics in macrophages, phenocopying macrophage activation, and exhibited potent adjunctive activity in the standard mouse model of TB chemotherapy and in a mouse model of human-like necrotic TB lung granulomas.

Conclusions

These data provide compelling evidence for clinical evaluation of pravastatin as adjunctive, host-directed therapy for TB.

Despite more than a century of research, tuberculosis (TB) kills millions of people every year [1]. There is an imminent but unmet medical need for novel therapeutics to complement the current antimicrobial arsenal to combat drug-susceptible and drug-resistant TB [2–4]. In particular, attention has focused on the potential repurposing of existing US Food and Drug Administration-approved agents, many of which have broad immunomodulatory properties, and whose safety and tolerability profiles have been well characterized, in an effort to obtain accelerated approval for use as adjunctive host-directed therapy (HDT) for TB [5–8]. Host-directed therapy offers a novel strategy with mechanisms that include activating immune defense mechanisms or ameliorating tissue damage [9].

There has recently been considerable interest in repurposing 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) as novel antimicrobials, including as antitubercular agents [10]. In addition to their cholesterol-lowering properties, statins have been shown to have broad anti-inflammatory and immunomodulatory properties [11]. More recently, Parihar et al [12] showed that peripheral blood mononuclear cells and monocyte-derived macrophages from patients with familial hypercholesterolemia receiving statin therapy were more effective in controlling Mycobacterium tuberculosis (Mtb) growth compared with those of healthy donors, and that statin therapy protected mice against TB-induced pathology. We found that simvastatin adjunctive therapy enhanced the first-line anti-TB regimen’s antimicrobial activity and shortened the time required to achieve cure in a standard mouse model of chronic TB infection [13, 14]. In addition, atorvastatin was found to be synergistic with the key sterilizing drug, rifampin [15]. Recent retrospective cohort studies found that statin use was associated with a reduced incidence of active TB disease [16, 17]. However, several important questions must be answered before statins can be used as adjunctive anti-TB agents in the clinical setting. For example, it remains to be determined whether the anti-TB activity of statins represents a class effect, or whether there is a hierarchy in antitubercular potency and toxicity among statins.

The objective of this study was to screen 8 commercially available statins (pravastatin, simvastatin, fluvastatin, atorvastatin, lovastatin, rosuvastatin, pitavastatin, and mevastatin) in various preclinical models to prioritize 1 or more statins for future prospective clinical studies.

MATERIALS AND METHODS

Assessment of Statin Cytotoxicity

To determine the 50% cytotoxic concentration (CC₅₀) of each statin, exposures were performed with the human monocytic THP-1 cell line (TIB-202; ATCC) and Vero cells (CCL-81; ATCC). Drug cytotoxicity was determined by the MTS method (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; Promega, Madison, WI). Cells were treated with a range of doses of each statin (with lowest dose corresponding to the inhibitory dose [IC₅₀] reported in the literature), and toxicity was measured at 24 hours posttreatment. After treatment with statins, the MTS solution was added to each well and incubated at 37°C for 30 minutes. The absorbance at 490 nm was recorded, and the CC₅₀ was determined using a nonlinear regression fitted with Prism 6 (GraphPad).

Assessment of Statin Efficacy in Macrophages

To determine the efficacy of statins against intracellular bacilli, a bioluminescent Mtb strain (H37Rv-lux), which stably expresses an integrated bioluminescent reporter (firefly luxABCDE full operon), including the luciferase enzyme and associated luciferin substrate, was used to infect THP-1 cells [14, 18]. Mycobacterial growth was monitored in real-time using a luminometer by measuring the relative light units (RLUs), which serve as a reliable surrogate for colony-forming units (CFUs) [19]. After differentiation, the cells were infected with a frozen stock of H37Rv-lux at a multiplicity of infection of 1:20 [14, 20]. Multiple concentrations of each statin (all causing <20% reduction of cell viability after 6 days of treatment) were tested to determine the concentration required to inhibit Mtb growth by 50% (half maximal effective concentration [EC₅₀]) for each compound.

Evaluation of Synergy of Statins With Antituberculosis Drugs in THP-1 Cells

To determine whether statins can enhance the activity of anti-TB drugs against intracellular bacilli, we used the RLU assay described above. We first determined the EC₅₀ of each of the 3 first-line drugs separately and then in combination. When each statin (simvastatin, pravastatin, or fluvastatin) was paired with the 3-drug combination of rifampin (0.0055 μM), isoniazid (0.006 μM), and pyrazinamide (81.23 μM) (collectively, RHZ) [14], each antibiotic in the combination was used at half the concentration required to reduce RLUs by 50% when dosed individually.

Phagosomal pH and Proteolysis Bead Assays

Carboxyfluorescein (pH readout) and DQ-BSA (proteolysis readout) beads were generated as previously described [21, 22]. Full details of the methods are provided in the Supplementary Materials.

Bone marrow-derived macrophages (BMDMs) were isolated from BALB/cJ mice (Jackson Laboratory, Bar Harbor, ME), and maintained in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY) containing 10% fetal bovine serum (FBS) (Gibco), 15% L-cell conditioned media, 2 mM L-glutamine (Sigma-Aldrich, St. Louis, MO), 1 mM sodium pyruvate (Gibco), and antibiotics (penicillin/streptomycin) (Gibco) at 37°C in a 5% CO₂ atmosphere. For the plate reader assay, 2 × 10⁵ macrophages/well were seeded into a 96-well clear bottom black plate (Corning Costar, Corning, NY). Where applicable, macrophages were activated by treatment with 100 U/mL interferon-γ (PeproTech, Rocky Hill, NJ) and 10 ng/mL lipopolysaccharide (List Biological Laboratories, Campbell, CA). Two days after seeding, the media were replaced with fresh media containing appropriate concentrations of pravastatin or water as a carrier control. Bead assays were carried out 2 days after treatment. Macrophages were washed 3 times with assay buffer (phosphate-buffered saline, pH 7.2, 5% FBS, 5 mM dextrose, 1 mM CaCl₂, 2.7 mM KCl, 0.5 mM MgCl₂), before addition of sensor beads at ~2–5 beads/macrophage in assay buffer [21, 22]. Data acquisition on a Biotek Synergy H1 microplate reader was started within 2–3 minutes of bead addition. Signal detection was from the bottom using the following excitation/emission wavelengths: carboxyfluorescein, 450 nm/520 nm and 490 nm/520 nm; DQ-BSA, 490 nm/520 nm; and Alexa Fluor (AF), 594–590 nm/617 nm. A total of 3–4 replicate wells/condition were used, with temperature maintained at 37°C throughout the assay. Reads were taken every 2 minutes for 2 hours for pH bead assays and every 2 minutes for 4 hours for proteolysis bead assays. Analyses of the results were performed as previously described [21, 22]. In brief, for the pH assay, the ratio of the carboxyfluorescein fluorescence signal at excitation 490 nm (pH-sensitive) versus 450 nm (pH-insensitive) provides a readout of relative pH. For the proteolysis assay, AF594 serves as an internal calibration control, allowing ratiometric measurement, with the DQ-BSA/AF594 ratio reflecting changes in proteolysis.

Ethics Statement

All procedures involving animals were performed according to protocols approved by the Institutional Animal Care and Use Committee at the Johns Hopkins University.

Tolerability and Pharmacokinetics of Statins in Mice

In steady-state studies, separate groups of female C3HeB/FeJ mice (aged 5 to 6 weeks; Jackson Laboratory) were given 1 of the 7 statins (mevastatin was not tested due to cytotoxicity) orally at a dose of 90 mg/kg together with isoniazid 10 mg/kg, rifampin 10 mg/kg, pyrazinamide 150 mg/kg, and ethambutol 100 mg/kg (collectively, RHZE) for 5 successive days. Rifampin doses were separated from the accompanying drugs by 1 hour to minimize drug-drug interactions during absorption [23, 24]. Whole lung tissue and serum at each time point were frozen at −20°C and later analyzed for concentrations of each drug by liquid chromatography-tandem mass spectrometry. Serum and lung drug concentration data were entered into a WinNonlin worksheet (WinNonlin version 4.0, 2002; Pharsight, Mountain View, CA) and analyzed using standard noncompartmental techniques [19, 25].

Antituberculosis Activity of Statins in C3HeB/FeJ Mice

A total of 110 female C3HeB/FeJ mice (aged 5 to 6 weeks; Jackson Laboratory) were aerosol-infected with Mtb H37Rv using an inhalation exposure system (Glas-Col, Terre Haute, IN) calibrated to deliver ~10² CFU/mouse lung. Six weeks after aerosol infection, mice were treated with human-equivalent doses of RHZE for 2 months and were randomized to receive one of the following adjuntive statin regimens designed to simulate human daily exposures (Supplementary Tables S2 and S3). The basic experimental scheme is shown in Supplementary Table S4. The treatment was administered once daily (5 days a week) for a total of 8 weeks. All statins were obtained from Toronto Research Chemicals (North York, ON, Canada), except pitavastatin, which was obtained from Kowa Company, Ltd. (Nagoya, Japan).

Dose-Ranging Activity of Pravastatin in the Standard Mouse Model of Tuberculosis Chemotherapy

Six-week-old female BALB/c mice (Charles River Laboratories, Wilmington, MA) were aerosol-infected with ~5 × 10³ CFU of wild-type Mtb H37Rv, as described above. Treatment was initiated 4 weeks later with RHZE or RHZE + pravastatin at doses ranging from 30 to 180 mg/kg for a total of 8 weeks. Animal total body, lung, and spleen weights were recorded, and the lungs and spleen were examined grossly for visible lesions and were photographed at the time of sacrifice. Endpoints included the bactericidal activity of each regimen at defined time points.

Data Evaluation and Statistical Analysis

Log-transformed CFUs were used to calculate means and standard deviations (SDs). Comparisons of CFU data among experimental groups were performed by Student t test (2-tailed). Pairwise comparisons of group mean values for log₁₀ counts were made by using one-way analysis of variance and Bonferroni’s multiple comparison test posttest with GraphPad Prism 7 (GraphPad, San Diego, CA), and P < .05 was considered significant.

RESULTS

Statins Show Differential Cytotoxicity

First, we tested the cytotoxicity of 8 commercially available statins in THP-1 cells and Vero cells. A range of concentrations of each statin was tested to determine the cytotoxic concentration leading to a 50% reduction in cell viability (CC₅₀). An example is provided in Supplementary Figure S1 demonstrating how the CC₅₀ of statin was determined in THP-1 cells. This approach was performed with the remaining 7 clinically available statins. Based on CC₅₀, pravastatin was determined to be the least cytotoxic of the statins tested (Supplementary Table S1).

Simvastatin, Fluvastatin, and Pravastatin Exhibit the Most Potent Activity Against Intracellular Mycobacterium tuberculosis

At nontoxic doses, atorvastatin and mevastatin were unable to inhibit Mtb growth (Supplementary Table S1), whereas rosuvastatin, pitavastatin, and lovastatin gave <50% growth inhibition in our assay. The EC₅₀ values (≥50% Mtb growth inhibition) could be obtained only for simvastatin, pravastatin, and fluvastatin (Figure 1A), and each of these drugs showed a favorable therapeutic index, as measured by CC₅₀/EC₅₀ (Supplementary Table S1). Thus, the majority of compounds inhibited intracellular Mtb growth, pointing to a class effect. Observed differences in antimicrobial activity may be due to differences in structure among the various statins.

Statin Exposure Differentially Enhances the Activity of the First-Line Drugs Against Intracellular Mycobacterium tuberculosis

The EC₅₀ values of isoniazid (H), rifampin (R), and pyrazinamide (Z) against Mtb in THP-1 cells were determined to be 0.006 μM, 0.0055 μM, and 81.23 μM, respectively. To test the synergistic activity of statins with standard antitubercular drugs, each statin was used at the respective EC₅₀ in combination with the 3-drug combination HRZ, with each anti-TB drug dosed at half its EC₅₀. We found that combination therapy with HRZ had more potent activity against intracellular Mtb than each statin alone (Figure 1B). Adjunctive therapy with pravastatin 7.8 μM significantly increased killing by HRZ (P  = 0.0024) (Figure 1B). A similar effect was observed with simvastatin 0.2 μM (P = .024) and fluvastatin 0.032 μM (P = .0004).

Pravastatin Treatment Phenocopies Phagosomal Acidification and Proteolytic Activity Profiles of Activated Macrophages

Previous reports have demonstrated that exposure of Mtb-infected macrophages to simvastatin promotes phagosome-lysosome fusion and autophagy [12]. To determine whether these activities also apply to pravastatin, we investigated the capacity of this drug to modulate phagosome acidification and proteolytic activity in mouse BMDMs. Treatment of resting BMDMs with pravastatin phenocopied the proteolytic activity and acidification profiles of phagosomes observed in interferon-γ- and lipopolysaccharide-activated BMDMs. This effect, which was concentration-dependent, suggests a contributing factor to intracellular Mtb killing by pravastatin (Figure 2).

In Vivo Pharmacokinetics Studies and Selection of Statin Dose for Efficacy Studies in Mice

Before undertaking the in vivo efficacy studies, we studied the safety and pharmacokinetics (PK) of each statin dosed daily at 90 mg/kg by gavage in C3Heb/FeJ mice, a strain of mice that develops human-like, necrotic TB lung granulomas [26–28]. To select the dose of each statin for the efficacy studies, we aimed to reproduce the human area under the plasma drug concentration-time curve (AUC) at (1) the highest clinically licensed dose and (2) twice that dose for selected compounds exhibiting the most attractive cytotoxicity profile and anti-TB activity in macrophages (simvastatin and pravastatin). It is important to note that for most statins, matching the AUC resulted in a C_max significantly higher than the corresponding value in humans (Supplementary Table S2).

Significant effects of rifampin coadministration on simvastatin or pravastatin exposures were not observed in mice (Supplementary Table S3), indicating that the standard mouse model does not recapitulate the known drug-drug interactions observed between rifampin and these statins in humans [29, 30]. No clinically apparent signs of toxicity were observed in any of the statin groups.

Statin Adjunctive Therapy Improves the Bactericidal Activity of the First-Line Antitubercular Regimen in C3HeB/FeJ Mice

After 8 weeks of treatment, Mtb-infected mice receiving statin adjunctive therapy had significantly reduced lung bacillary burdens relative to control mice receiving RHZE alone. Relative to the control regimen, adjunctive therapy with simvastatin 90 mg/kg, pravastatin 90 mg/kg, pravastatin 50 mg/kg, or fluvastatin 15 mg/kg reduced colony counts in the lungs by 1.28 log₁₀ (P < .0001), 1.16 log₁₀ (P < .01), 0.78 log₁₀ (P < .05), and 0.90 log₁₀ (P < .05), respectively (Figure 3A). There were no significant differences in total body weights, lung and spleen weights, gross lung pathology of mice lungs, or surface area of lung involved by inflammation in the treatment groups compared with control groups treated with RHZE alone (Figure 3B).

Pravastatin Shows Adjunctive Activity in a Standard Mouse Model of Tuberculosis Chemotherapy

Given its favorable therapeutic index and reduced drug interactions with rifampin relative to simvastatin in humans [29, 30], we next performed a dose-ranging study of pravastatin adjunctive therapy in the mouse model of TB chemotherapy [25, 31]. The Mtb-infected BALB/c mice were treated with the standard TB regimen (RHZE) with or without increasing doses of pravastatin ranging from 30 to 180 mg/kg. Treatment was given orally 5 times weekly for 8 weeks (continuous phase) beginning 6 weeks after infection. Adjunctive treatment with pravastatin increased the bactericidal activity of the first-line regimen, reducing lung bacillary counts by 0.2–0.6 log₁₀, 0.3–0.6 log₁₀, and 0.3–0.8 log₁₀ compared with RHZE alone after treatment for 2 weeks, 4 weeks, and 8 weeks, respectively (Figure 4).

We also tested the effect of each drug regimen on lung inflammation by histological assessment at 8 weeks posttreatment. We observed that, in general, the percentage of lung surface area involved by inflammation correlated with the bactericidal activity of each drug regimen (Figure 5). Specifically, the percentage of lung surface area occupied by inflammatory cells was reduced by 61.51% in the RHZE-treated group (20.88% ± 7.72%) relative to the untreated control (82.40% ± 3.60%). Treatment with pravastatin adjunctive therapy showed a statistically nonsignificant further reduction in mean inflamed lung surface area relative to the RHZE control group (13.76% for RHZE + pravastatin 90 mg/kg, P = .3 and 9.61% for RHZE + pravastatin 180 mg/kg, P = .62).

DISCUSSION

Because the TB drug development pipeline remains sparse, novel approaches are required to shorten the duration of treatment and improve microbiological and clinical outcomes. The following HDT agents are currently being evaluated in phase 2 clinical trials as adjuncts to rifabutin-modified standard therapy in adults with drug-sensitive, smear-positive pulmonary TB: (1) the mammalian target of rapamycin inhibitor, everolimus (0.5 mg), (2) auranofin (6 mg), (3) vitamin D3, and (4) the phosphodiesterase-4 (PDE4) inhibitor, CC-11050 (ClinicalTrials.gov Identifier: NCT02968927) [32–35]. In a recent study, an analysis of retrospective cohorts showed that metformin use was associated with improved control of Mtb infection and decreased disease severity in patients with pulmonary TB [36].

In addition to pravastatin’s antitubercular activity in mice described in the present study, additional considerations favor the clinical use of pravastatin over other members of the class, including simvastatin, which showed the most favorable CC₅₀/EC₅₀ in macrophages in vitro. First and foremost, unlike simvastatin, the metabolism of pravastatin is not significantly mediated by cytochrome P450 enzymes [37–39]. Thus, although the steady-state AUC of both simvastatin and its active metabolite, simvastatin acid, are reduced by ~90% during coadministration with rifampin in humans [30], pravastatin exposures are reduced only by 30%–50% by rifampin [29], and the latter drug interaction can be overcome by doubling the dose of pravastatin [30]. This is particularly important because rifampin is a key sterilizing drug, and its inclusion in the first-line regimen has contributed significantly to treatment shortening to the current 6-month regimen for drug-susceptible TB. In addition, whereas human immunodeficiency virus (HIV) protease inhibitors have been shown to increase the median AUC of simvastatin by as much as 30-fold [40], pravastatin is one of the preferred statins for use in individuals with HIV infection due to its favorable toxicity profile and relatively low potential for drug interactions with antiretroviral drugs, an important consideration in populations in which TB and HIV are coendemic. Our previous PK data indicate that simvastatin is almost entirely converted to its acid metabolite in mice [14], whereas in humans, metabolism to simvastatin acid is ~1:1 [41]. Because it is not known where the parent drug or its acid metabolite is responsible for the observed anti-TB activity, it is possible that the mouse data overestimate the adjunctive activity of simvastatin.

Statins inhibit HMG-CoA reductase, the enzyme that catalyzes the conversion of HMG-CoA to mevalonate. The mevalonate pathway is essential for cholesterol synthesis, but it also contributes to the production of isoprenoids such as farnesyl pyrophosphate and geranylgeranyl pyrophosphate that are important for cell signaling and structure. A recent study found that inhibiting cholesterol with statins decreased the burden of Mycobacterium leprae in macrophages, highlighting the importance of cholesterol in promoting bacterial growth [42] and enhancing macrophage killing ability [15]. The ability of statins to reduce cholesterol has also been linked to increased killing of Helicobacter pylori [43], Coxiella burnetii [44], and Rickettsia conorii [45]. However, the beneficial effects of statins are not limited to cholesterol reduction alone, and they may involve processes associated with isoprenylation. These molecules serve as lipid labels for the posttranslational modification of several proteins, including G protein gamma subunits and small GTP-binding proteins, such as Ras, Rho, Rab, Rac, Ral, or Rap [46, 47]. Given the central role of small GTP-binding proteins in controlling the immune response, it is not surprising that statins exerts immunomodulatory effects independent of cholesterol lowering.

Our data showing reduced phagosome acidification and proteolysis of macrophages after pravastatin exposure are consistent with the notion that pravastatin-treated macrophages display an activated phenotype. Decreased proteolysis can promote antigen presentation [48, 49], which is beneficial for bacterial clearance, whereas the observed delay in acidification reflects a balance in the production of reactive oxygen species, which consume protons, and phagosomal acidification upon immune cell activation [48, 50]. Our macrophage functional data add to the endosomal marker data of Parihar et al [12], and the 2 studies are consistent with the conclusion that statin exposure induces macrophage activation.

From our work and that of other groups, it appears that statins can exert their effect on various cell types, and their anti-TB activity is likely multifactorial. We acknowledge that the mechanistic studies in the current study showing altered phagosome acidification and proteolytic activity with statin exposure do not necessarily demonstrate causality; however, they pave the way for further mechanistic studies to determine the molecular basis of the anti-TB activity of statins.

CONCLUSIONS

The preponderance of evidence suggests that statins may be useful as adjunctive HDT for TB. However, a number of important questions remain to be addressed before introducing statins as adjunctive agents in TB therapy, including the precise dosing of pravastatin, taking into consideration known drug interactions with rifampin, and its potential utility in patients with HIV coinfection. A randomized clinical trial, Statins as Adjunctive Therapy for TB (StAT-TB), is currently underway to determine whether pravastatin adjunctive therapy shortens the median time to sputum-culture negativity and improves lung function outcomes among HIV-infected and uninfected patients with drug-susceptible pulmonary TB.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. We thankfully acknowledge Michael L. Pinn for excellent technical assistance.

Disclaimer. The funding sources had no role in the study design, data collection, data analysis, data interpretation or writing of the report. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Financial support. This work was funded by Grant R00 AI114952 (to S. T.), BMGF Bill & Melinda Gates Foundation Grant OPP 1066499 (to V. D.), Grant R01 HL106788 (to M. L. G.), and ACTG AIDS Clinical Trials Group Grant 110007 and CFAR Centers for AIDS Research Supplement P30AI094189, R01 HL106786, and UH2/3 AI122309 (to P. C. K.).

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

References