Thioridazine Is an Efflux Pump Inhibitor in Mycobacterium avium Complex but of Limited Clinical Relevance.

SUSCEPTIBILITY crossm Thioridazine Is an Efflux Pump Inhibitor in Mycobacterium avium Complex but of Limited Clinical Relevance Mike Marvin Ruth,a Lian J. Pennings,a Valerie A. C. M. Koeken,b Jodie A. Schildkraut,a Aria Hashemi,a Heiman F. L. Wertheim,a Wouter Hoefsloot,c Jakko van Ingena a Radboud Center for Infectious Diseases, Department of Medical Microbiology, Radboud University Medical Center, Nijmegen, The Netherlands b Radboud Center for Infectious Diseases, Department of Internal Medicine, Radboud University Medical Center, Nijmegen, The Netherlands c Radboud Center for Infectious Diseases, Department of Pulmonary Diseases, Radboud University Medical Center, Nijmegen, The Netherlands ABSTRACT Treatment of Mycobacterium avium complex pulmonary disease (MACPD) is challenging partly due to high efflux pump expression. Thioridazine might block these efflux pumps. We explore the efficacy of thioridazine against M. avium isolates using MICs, time-kill combination assays, ex vivo macrophage infection assays, and efflux assays. Thioridazine is bactericidal against M. avium, inhibits intracellular growth at 2⫻ MIC, and blocks ethidium bromide efflux. However, its toxicity and low plasma concentrations make it unlikely to add efficacy to MAC-PD therapy. KEYWORDS M. avium, efflux pumps, thioridazine N ontuberculous mycobacteria (NTM) are highly drug-resistant opportunistic pathogens primarily causing pulmonary disease (NTM-PD) (1). NTM-PD caused by Mycobacterium avium complex (MAC-PD) accounts for ⬃80% of NTM-PD cases (1). Treatment of MAC-PD is lengthy and based on an azithromycin-ethambutol-rifamycin backbone but only achieves prolonged culture conversion in ⬃50% to 70% of patients (1, 2). MAC-PD is highly drug resistant partly due to high efflux pump induction, increasing drug tolerance (3). A strategy to sensitize MAC bacteria to multidrug therapy is neutralizing these efflux pumps. Thioridazine has anecdotal effectivity against M. tuberculosis (4, 5). It inhibits its efflux pumps and blocks its respiratory chain (6). The potential role of thioridazine in NTM treatment has been explored in vitro using hollow-fiber experiments with promising results (7). However, in vitro studies combining thioridazine with drugs used in the standard treatment regimen of MAC are limited (8). Here, we explore whether thioridazine might be a viable option in NTM treatment based on in vitro and ex vivo studies. For susceptibility testing, NTM reference strains were obtained from their respective culture collections (Table 1). Clinical MAC isolates were acquired from the collection of the Department of Medical Microbiology at Radboud University Medical Center, Nijmegen, The Netherlands. MICs and minimum bactericidal concentrations (MBCs) of thioridazine were determined by broth microdilution in cation-adjusted Mueller-Hinton broth (CAMHB) (BD Bioscience, Drachten, The Netherlands) per CLSI guidelines (9), and we defined bactericidal activity as previously described (10). Synergy between thioridazine and established antimycobacterial drugs was assessed using the checkerboard method, subsequently calculating the fractional inhibition concentration index (FICI). We interpreted a FICI of ⬍0.5 as synergistic, 0.5 to 4 as no interaction, and ⬎4 as antagonistic (11). The combination with the lowest FICI value was selected for subsequent combination time-kill assays (TKs). A dose-response TK of thioridazine against M. avium ATCC 700898 was performed in CAMHB as previously described (10), using a concentration range of 0.25⫻ to 32⫻ MIC. Combination TKs were performed between thioridazine and clarithromycin or rifampin as described previously (10), using concenJuly 2020 Volume 64 Issue 7 e00181-20 Antimicrobial Agents and Chemotherapy Citation Ruth MM, Pennings LJ, Koeken VACM, Schildkraut JA, Hashemi A, Wertheim HFL, Hoefsloot W, van Ingen J. 2020. Thioridazine is an efflux pump inhibitor in Mycobacterium avium complex but of limited clinical relevance. Antimicrob Agents Chemother 64:e00181-20. https://doi.org/10.1128/AAC .00181-20. Copyright © 2020 American Society for Microbiology. All Rights Reserved. Address correspondence to Jakko van Ingen, . Received 27 January 2020 Returned for modification 26 March 2020 Accepted 13 April 2020 Accepted manuscript posted online 20 April 2020 Published 23 June 2020 aac.asm.org 1 Ruth et al. Antimicrobial Agents and Chemotherapy TABLE 1 MIC values of thioridazine against rapidly growing and slowly growing mycobacteria Species and isolate Rapidly growing mycobacteria M. abscessus CIP 104536 M. fortuitum ATCC 6841 M. peregrinum ATCC 700686 Slowly growing mycobacteria M. avium ATCC 700898 B16106707 B16107228 B16105577 B17019074 M. chimaera DSM 44623 B16015538 B16018489 B17072535 M. intracellulare DSM 43223 B16125524 B16116883 B16029695 M. simiae ATCC 25221 M. xenopi ATCC 19250 MIC (␮g/ml) 16 8 4 16 4 2 32 16 16 8 4 4 4 2 16 1 1 4 trations ranging from 0.5⫻ to 2⫻ MIC for the companion drug and the combination of the two. Bliss independence calculations to assess synergistic interactions over time, as well as interpretation of the data, were performed as described previously (10). For ex vivo assays, peripheral blood mononuclear cells of three healthy volunteers were obtained with their informed consent as described previously (12). Infection of the differentiated macrophages was performed as described previously (12). Based on the suggestion of synergy in the TK data, thioridazine, thioridazine plus clarithromycin, and clarithromycin were added at 2⫻ MIC after 1 h of incubation and were present for the remainder of the experiment. After 6 days, quantification of extracellular and intracellular bacteria was performed as described previously (12). For fluorometric kinetics, filter-ventilated bottles containing 10 ml of CAMHB were inoculated with M. avium ATCC 700898 as in TKs and incubated until log-phase growth (7 log10 CFU/ml). Ethidium bromide (EtBr) was added to a final concentration of 1 ␮g/ml. Bottles were incubated for 1 h at 37°C, and their content was centrifuged, washed, and resuspended to original volume in phosphate-buffered saline containing 0.5% glucose (wt/vol) and thioridazine at 0.5⫻ MIC. One hundred microliters of bacterial suspension was added to a 96-well plate. Thioridazine-free and glucose-free controls were also used. Fluorescence was measured using a Tecan 10M light cycler (Tecan Benelux, Giessen, The Netherlands) at 533 to 630 nm as described previously (13). Emission was measured every 2 min for 1 h at 37°C and was normalized at time zero (T0). The MICs of thioridazine against NTM are shown in Table 1. MICs of reference strains of rapidly growing mycobacteria ranged from 4 to 16 ␮g/ml. The MIC of thioridazine against M. avium ATCC 700898 was 16 ␮g/ml, and the MBC was 32 ␮g/ml, defining it as bactericidal (MBC/MIC ⫽ 2). The MIC range for clinical MAC isolates was 1 to 32 ␮g/ml. Thioridazine had the lowest FICI with rifampin and clarithromyc (...truncated)