In vitro activity of bedaquiline against rapidly growing nontuberculous mycobacteria

Bedaquiline (BDQ) has been proven to be effective in the treatment of multidrug-resistant tuberculosis. We hypothesized that BDQ could be a potential agent to treat nontuberculous mycobacterial (NTM) infection. The objective of this study was to evaluate the in vitro activity of BDQ against rapidly growing mycobacteria by assessing the minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC) against 18 NTM strains. For MIC determination we performed the resazurin microtitre assay broth dilution, and for the MBC the c.f.u. was determined. BDQ exhibited a strong inhibitory effect against most NTM tested; however, for some NTM strains the MBC was significantly higher than the MIC. A new finding is that Mycobacterium flavescens has a mutation in the gene atpE associated with natural resistance to BDQ. These preliminary promising results demonstrate that BDQ could be potentially useful for the treatment of NTM.

The genus Mycobacterium comprises more than 150 different species of mycobacteria with the capacity to cause pathogenicity in humans [1]. Most important among these species, due to their airborne transmission and public health implications, are Mycobacterium tuberculosis and Mycobacterium leprae causing tuberculosis (TB) and leprosy, respectively. Among nontuberculous mycobacteria (NTM), Mycobacterium avium and Mycobacterium abscessus represent prevalent sources of infection not only in immunocompromised individuals but also in other susceptible populations, such as in cystic fibrosis patients [2]. More recently, other emerging NTM, such as Mycobacterium chimaera, have been reported as causes of outbreaks due to heating-cooling devices in surgical rooms [3].
Drug resistance is one of the key issues associated with the current burden of TB around the world, negatively impacting control of the disease [4,5]. Consequently, efforts have been devoted to the discovery and development of new anti-TB drugs [6]. As a result two new drugs, bedaquiline (BDQ) and delamanid, were recently approved for the treatment of multidrug-resistant TB (MDR-TB) [7,8]. BDQ has a broad antimycobacterial spectrum and a novel mode of action, inhibiting the ATP synthase [9]. We hypothesized that BDQ could also treat NTM infection.
For infections caused by NTM, combination antimicrobial chemotherapy is the treatment of choice in most cases [10,11]. Nevertheless, NTM is difficult to eradicate because most of them are naturally resistant to many common antibiotics and in many cases become refractory to the commonly recommended antibiotics [12]. In this context, BDQ has recently been used as off-label for salvage treatment in patients with Mycobacterium intracellulare lung disease, with encouraging results [13]. In order to shed light on the conditions and parameters guiding the potential use of BDQ for NTM infections, we evaluated its in vitro activity against a panel of rapidly growing NTM reference strains and clinical isolates, and explored the possible correlation of single nucleotide polymorphisms in the target gene and natural resistance to the drug.
Eighteen rapidly growing mycobacterial strains were used in this study (Table 1). Seventeen were obtained from the CCUG collection (http://www.ccug.se) and one strain from the UCL collection in Brussels, Belgium. Strains were cultured on Löwenstein-Jensen medium and the inoculum was prepared in distilled water, adjusted to McFarland 0.5 and diluted 1 : 10 in Mueller-Hinton (MH) broth medium. Mycobacterium smegmatis CCUG 28063 was used for quality control since its minimum inhibitory concentration (MIC) for BDQ of 0.015 µg ml À1 is well known [9]. To assess whether BDQ had a bacteriostatic or bactericidal effect, we determined the MIC and minimum bactericidal concentration (MBC) using the resazurin microplate assay (REMA) [14,15]. Briefly, twofold serial dilutions were made in MH in 96-well polystyrene plates. BDQ concentrations were 2.0-0.0035 µg ml À1 and each experiment was performed in triplicate. An inoculum equal to McFarland 0.5 diluted 1 : 10 was prepared. Growth controls without drug (positive control), a drug control and a sterile control (negative control) were also prepared for each assay. To prevent evaporation during incubation, 200 µl sterile distilled water was added to all perimeter wells. Plates were sealed and incubated at 37 C for 3 days before adding 30 µl of 0.01 % resazurin to all wells and incubating for a further 24 h. The MIC was determined as the lowest drug concentration that prevented growth and, therefore, a colour change from blue (oxidized state) to pink (reduced state). MIC values were scored for each isolate tested. The same plates were used for MBC determination. At day 4 of incubation and after the MIC reading, four blue wells were chosen to test the viability of the mycobacteria. One hundred microlitres from each well at the MIC, one concentration higher, and the previous two BDQ dilutions, were transferred to a tube and diluted in sterile distilled water to 10 À3 , 10 À4 and 10 À5 and plated in duplicate on Luria broth (LB) agar plates to determine the c.f.u. Also, c.f.u. were determined in duplicate for the positive control diluted 10 À4 , 10 À5 and 10 À6 . The plates were incubated for 4 days. The percentage of killed bacteria was calculated against the control, and the MBC was defined as the lowest drug concentration that killed 99.9 % of bacteria.
For investigation of the mutation in the gene atpE, DNA extraction was carried out according to Perez-Martinez et al. [16]. Briefly, a loopful of mycobacteria from a Löwenstein-Jensen culture was resuspended in 100 µl Milli-Q water, boiled for 5 min, placed on ice for 10 min, centrifuged at room temperature (13 600 g, 5 min) and the supernatants were used for PCR. The atpE gene was amplified using degenerated primers atpE forward (degenerated) 5¢-TGTA YTTCAGCCARGCSATGG-3¢ and atpE reverse (degenerated) 5¢-CCGTTSGGDABGAGGAAGTTG-3¢ [17]. However, if these primers did not amplify atpE, a second set of primers was used: atpE forward 5¢ -TGTACTTCAGC-CAAGCGATGG-3¢ and atpE reverse 5¢-CCGTTGGGAA TGAGGAAGTTG-3¢ [18]. For the degenerated primers, the PCR was run with an initial pre-denaturation at 95 C for 5 min, followed by 30 cycles of denaturation at 95 C for 1 min, annealing at 57 C for 1 min and elongation at 72 C for 1 min. The reaction was finished with 7 min final elongation at 72 C. Amplicons were detected by agarose (1.5 %) gel electrophoresis and ethidium bromide staining.
For the second set of primers, the PCR was run with an initial pre-denaturation at 95 C for 5 min, followed by 30 cycles of denaturation at 95 C for 1 min, annealing at 62 C for 1 min and elongation at 72 C for 1 min. The reaction was finished with 7 min final elongation at 72 C.
In both cases, identical primers were used for sequencing PCR (BigDye Terminator Sequencing Kit; Applied sensitivity of NTM to BDQ. In addition, more research is needed to understand why in some species the MIC and MBC of BDQ are very close and for other species the MBC is much higher than the MIC, and to elucidate the most promising companion drugs. In conclusion, to the best of our knowledge, this is the first study to have assessed the MIC and MBC values for BDQ against a large number of rapidly growing NTM. We also described for the first time that M. flavescens is naturally resistant to BDQ and its high MIC correlates with the mutation found at amino acid 63 in AtpE (alanine replaced by methionine). However, despite this finding, BDQ exhibited a strong inhibitory effect against all NTM tested, suggesting the potential to treat NTM infections. These preliminary results warrant further research in this area.

Funding information
The authors received no specific grant from any funding agency.