The Parkinson's drug benztropine possesses histamine receptor 1-dependent host-directed antimicrobial activity against Mycobacterium tuberculosis

Abstract

Intracellular pathogens such as Mycobacterium tuberculosis (Mtb) evade host defence mechanisms to infect and survive within host cells. Host-directed therapy (HDT) offers a promising alternative to antibiotics and may overcome antimicrobial resistance. Using high-content screening, we identified benztropine (BZT), an approved Parkinson's disease drug, as a potent inhibitor of intracellular Mtb. BZT is active in both human and murine macrophages but is inactive in broth. In an aerosol Mtb mouse infection model, oral administration of BZT reduced the burden of Mtb in the lungs by up to 70%. BZT was also active against Salmonella enterica serovar Typhimurium (STm) in an abscess model of infection, significantly reducing size and bacterial load. Chemical competition assays, CRISPR knockouts, and siRNA silencing assays revealed that BZT's activity against Mtb is mediated via macrophage histamine receptor 1 (HRH1). Our findings establish BZT as a promising repurposed candidate and a lead compound for developing HRH1-targeting antibacterial HDTs.

Fig. 1. Screening the COVID Box identified compounds active against intracellular Mtb.

a Scatter plot showing the distribution of inhibitory effects of compounds (50 µM each) from the COVID Box library on intracellular Mtb in THP-1 macrophages 72 h post-infection. Green circles represent hits that inhibit Mtb growth by more than 70% with less than 25% cytotoxicity. b Seven hits with >70% intracellular Mtb growth inhibition (blue bars) and minimal cytotoxicity (red bars) were identified. Two are established antibiotics (doxycycline and tetracycline), while the other five were newly identified. c Chemical structures of the five newly identified hits. d Dose-dependent activity of BZT (IC₅₀ = 15 µM) against intracellular Mtb. e CFU counts of Mtb after 72 h of infection in THP-1 macrophages. NT non-treated control. Data represent the mean ± SEM of at least two independent experiments. Statistical significance was determined using an unpaired T-test; ^**p < 0.005, ^*p < 0.05.

Fig. 2. BZT displayed host-dependent bacteriostatic activity against Mtb.

a Comparison of Mtb growth inhibition between intracellular infection (ex vivo) and direct antimicrobial activity in broth (in vitro) of the five identified hits. BZT showed no activity in broth against Mtb. b BZT did not inhibit Mtb growth in broth at concentrations up to 100 µM, whereas ethaverine and drotaverine exhibited concentration-dependent growth inhibition. c Pre-treatment of THP-1-derived macrophages with BZT (30 µM) for 24 h before Mtb infection did not result in significant inhibition of Mtb growth. Comparative analysis of the effect of BZT treatment on changes; d in intracellular Mtb growth, e percentage of infection THP-1 macrophages, and f the average number of Mtb per infected cells at the start of the infection (time 0) and 96 h post-infection. BZT exhibited bacteriostatic effects, leading to a gradual reduction of all three parameters, but still maintained higher levels than the basal level at time 0. In contrast, the TB drug bedaquiline (BDQ, 4 µM) showed bactericidal activity. NT non-treated controls. Data represent mean ± SEM of at least two independent experiments. Statistical significance was determined using a paired ratio T-test; ^*p < 0.05.

Fig. 3. Specificity and spectrum of BZT’s intracellular antimicrobial activity.

a BZT showed a dose-dependent intracellular activity against Mtb in RAW 264.7 cells. b hMDMs infected with GFP-expressing Mtb were monitored for two weeks using live-cell imaging and automated GFP fluorescence area analysis. Mtb-infected hMDMs were treated with two-fold serial dilutions of BZT. Data were normalised to cells treated with the vehicle control (1% DMSO), defined as 0% inhibition, and the positive control (1 µg/mL RIF + INH, R + I), defined as 100% inhibition. c BZT demonstrated a dose-dependent inhibition of Mtb growth in hMDMs. d Comparison of Mtb growth in hMDMs isolated from three healthy donors. The statistical analysis was conducted by pooling data across all donors. BZT showed dose-dependent activity against STm in e THP-1 macrophages and f RAW 264.7 cells. Data represent mean ± SEM of at least two independent experiments. NT non-treated control.

Fig. 4. Chemical competition assay reveals that BZT’s activity is mediated through blocking HRH1.

In Mtb-infected THP-1 macrophages; a addition of histamine (HIS, 10 µM), the natural ligand of the H1 receptor, blocked BZT (30 µM)-mediated inhibition of intracellular Mtb growth and b pyrilamine (PYR, 50 µM), an HRH1 antagonist, exhibited intracellular activity against Mtb. Graphs showing % intracellular Mtb growth inhibition of; c eight FDA-approved H1 antagonists (50 µM each) and; d BZT (25 µM) and seventeen other structural analogues of BZT (25 µM each) in Mtb infected THP-1 macrophages all the tested compounds were added to the media 3 h post infection and Mtb growth inhibition was determined 72 h post infection. e Chemical structure of two of the BZT analogues with higher anti-Mtb potency than BZT. NT non-treated control. Data represent the mean ± SEM of at least two independent experiments. Statistical significance was determined using an unpaired T-test; ^***p < 0.001, ^**p < 0.005, ^*p < 0.05; ns non-significant.

Fig. 5. Silencing/knocking out the HRH1 gene mimics phenotypes observed with BZT treatment of Mtb-infected THP-1 macrophages.

a siRNA-mediated silencing of hrh1 transcription in THP-1 macrophages inhibited intracellular Mtb growth. The addition of BZT (30 µM) did not significantly affect Mtb growth, while the addition of exogenous HIS (10 µM) partially reversed Mtb inhibition caused by hrh1 siRNA. b Western blot of HRH1 protein expression levels in parental and HRH1 K/O cells. Actin is used as a loading control. c HRH1 CRISPR knockout THP-1 macrophage clones (HRH1 K/O) showed reduced intracellular Mtb levels. d Inhibition of Mtb growth in parental THP-macrophages was not significantly altered by BZT (10 µM) in HRH1 K/O. Colocalization of acidic phagosomes and pHrodo-labelled Mtb MC²6206 auxotroph expressing GFP; e Heat-killed (HK) or live Mtb parental THP-1 macrophages that were either non-treated (NT) or treated with BZT (30 µM); f Heat-killed (HK) or live Mtb-infected parental THP-1 macrophages and live Mtb-infected HRH1 K/O; and g HK or live Mtb-infected HRH1 K/O that were either not treated (NT) or treated with BZT (30 µM). Data normalised to HK Mtb-infected cells, indicating 100% colocalization. Data represent the mean ± SEM of at least two independent experiments. Statistical significance was determined using unpaired T-test; ^****p < 0.0001, ^***p < 0.001, ^**p < 0.005, ^*p < 0.05; ns non-significant.

Fig. 6. In vivo activity of BZT against Mtb and STm.

a Oral treatment with 10 mg/kg and 20 mg/kg BZT reduced Mtb load in the lungs of Mtb-infected mice. Rifampin (RIF) at 10 mg/kg and 25 mg/kg, served as a positive control. BZT did not reduce Mtb levels in mouse; b livers; c spleens. BZT (5 mg/kg, subcutaneous) reduced; d abscess size; and e bacterial recovery from abscesses in a STm high-density abscess model. f BZT reduced clinical signs of morbidity resulting from high-density subcutaneous infection with STm. Data represent the mean ± SEM of two or three independent experiments containing 2–3 biological replicates each (N = 5–6). Statistical significance was determined using the Mann–Whitney U-test; ^****p < 0.0001, ^***p < 0.001, ^**p < 0.005, ns non-significant, NT non-treated controls.