An Experimental Model for the Rapid Screening of Compounds with Potential Use Against Mycobacteria

Abstract

Infections caused by Mycobacterium tuberculosis and other mycobacteria are major challenges for global public health. Particularly worrisome are infections caused by multidrug-resistant bacteria, which are increasingly difficult to treat because of the loss of efficacy of the current antibacterial agents, a problem that continues to escalate worldwide. There has been a limited interest and investment on the development of new antibacterial agents in the past decades. This has led to the current situation, in which there is an urgent demand for innovative therapeutic alternatives to fight infections caused by multidrug-resistant pathogens, such as multidrug-resistant tuberculosis. The identification of compounds that can act as adjuvants in antimycobacterial therapeutic regimens is an appealing strategy to restore the efficacy lost by some of the antibiotics currently used and shorten the duration of the therapeutic regimen. In this work, by setting Mycobacterium smegmatis as a model organism, we have developed a methodological strategy to identify, in a fast and simple approach, compounds with antimycobacterial activity or with potential adjuvant properties, by either inhibition of efflux or other unrelated mechanisms. Such an approach may increase the rate of identification of promising molecules, to be further explored in pathogenic models for their potential use either as antimicrobials or as adjuvants, in combination with available therapeutic regimens for the treatment of mycobacterial infections. This method allowed us to identify a new molecule that shows promising activity as an efflux inhibitor in M. smegmatis.

Introduction

Antimicrobial resistance is currently recognized as a global public health threat. Nowadays, an alarming number of common bacterial pathogens display phenotypes of multidrug resistance, for which the existing therapeutic options are limited, more toxic, and associated with higher costs for public health services.¹

Bacterial resistance to antimicrobials can be mediated by several mechanisms, such as antibiotic inactivation/degradation, alteration of the cell target(s), and reduction of the antimicrobial intracellular concentration by changes in the membrane permeability and/or through the activity of efflux systems.² Efflux pumps are present in all bacteria, contributing to the extrusion of diverse molecules, including toxins, secondary metabolites, and antimicrobials.³ Several recent studies have shown that efflux plays a crucial role in the early stages of development of antimicrobial resistance in the major pathogen Mycobacterium tuberculosis.^4

–9

One strategy to overcome antimicrobial resistance is to use a combined therapy by which an antibiotic is administered together with an adjuvant molecule that inhibits its efflux, thus keeping its intracellular concentration within a lethal interval, which prevents the emergence of resistant strains.¹⁰ Therefore, multiple efforts have been made toward the identification of such molecules and to test their efficacy against clinically relevant pathogens, such as mycobacteria.^{11

–18}

In this work, we present a methodological approach to rapidly identify molecules with antimicrobial and/or efflux inhibitory properties against mycobacteria, using Mycobacterium smegmatis as an experimental model. This species, rather than the more expensive and slow-growing M. tuberculosis, has already been established as a useful surrogate to study mycobacterial efflux activity,^19
–21 the effect of efflux inhibitors,^{13,22

–26} or even mycobacterial outer membrane components.²⁷ M. smegmatis has also been used successfully in the screening of molecules with antituberculosis activity.^22,28,29

The methodological approach proposed consists in a step-wise selective process, in which a candidate compound is sequentially evaluated and classified in one of the following four possible categories: (a) potential antimycobacterial, (b) potential adjuvant by inhibition of efflux, (c) potential adjuvant by mechanism(s) other than efflux inhibition, or (d) with no relevant biological activity. To study efflux activity, we have used the fluorescent dye ethidium bromide (EtBr) as a marker. The properties of this molecule as a common substrate of efflux pumps and a fluorochrome have been widely explored in studies on efflux activity in several bacteria.^{19,26,30
–32} The methodological model was validated with three compounds whose effect on mycobacteria has been well established, namely the phenothiazines thioridazine and chlorpromazine, and the ion channel blocker verapamil.^{4,6,19,33

–36} To further corroborate the efficacy of the presented algorithm, two compounds belonging to the class of the 2-aminothiazoles were tested to evaluate and categorize their antimycobacterial properties.

We have deeply investigated the structure–activity relationship of substituted 2-aminothiazoles with regard to their antitubercular activity.^12,14 In this work, reasoning on the versatility of this scaffold, which has been found to be the central core of many bioactive compounds, we have also investigated its potential as an inhibitor of efflux in mycobacteria. To avoid biases because of the intrinsic activity of the tested molecules, we selected a substituted 4-phenyl-2-aminothiazole [N-(3,5-dichlorophenyl)-4-(2-fluoro-5-trifluoromethylphenyl)thiazol-2-amine, UPAR-109] with a medium-low activity toward M. tuberculosis (minimum inhibitory concentration [MIC] = 15 μM)¹² and a 4-(isoxazol-5-yl)thiazol-2-amine (N-adamantan-1-yl)-5-(2-(3,5-dichlorophenylamino)thiazol-4-yl)isoxazole-3-carboxamide, UPAR-331), showing no activity against the same strain (MIC = >128 μM, article under preparation) (Fig. 1). The latter compound possesses an adamantane moiety, which, along with the 2-aminothiazole, is a widely distributed feature in medicinal chemistry, especially with regard to inhibition of transporters.^37

–40

Fig. 1.

Structure of compounds UPAR-109 and UPAR-331 tested in this work.

Materials and Methods

Reagents

(a) Commercially available compounds

EtBr, chlorpromazine (CPZ), thioridazine (TZ), verapamil (VP), erythromycin (ERY), clarithromycin (CLR), ofloxacin (OFX), rifampicin (RIF), ethambutol (EMB), streptomycin (STR), and isoniazid (INH) were purchased from Sigma-Aldrich (St. Louis, MO). Ciprofloxacin (CIP) was purchased from Fluka Chemie GmbH (Buchs, Switzerland). All solutions were prepared on the day of the experiment in deionized water, except for ERY, CLR, and RIF, prepared in dimethyl sulfoxide (DMSO).

(b) Synthesis of compounds UPAR-109 and UPAR-331

The synthesis of compound UPAR-109 has already been described in detail.¹² Compound UPAR-331 was prepared according to a straightforward protocol reported in detail in the Supplementary Data (see Experimental Section, Supplementary Fig. S1; Supplementary Data are available online at www.liebertpub.com/adt). The isoxazole ring was obtained through a 1,3-dipolar cycloaddition between the suitable commercially available oxime and 3-butyn-2-one. Subsequently, the bromination of ketone with bromine and glacial acetic acid in chloroform at 50°C afforded the α-bromoketone 4, which underwent the Hantzsch reaction with 3,5-dichlorophenylthiourea in absolute ethanol to yield the 2-aminothiazole 5. The ethyl ester was then hydrolyzed with LiOH in THF/MeOH/H₂O at room temperature to give the acid that is reacted with amantadine hydrochloride and Et₃N exploiting O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate and N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride as the coupling agents to yield the corresponding amide in good overall yield. Compounds UPAR-109 and UPAR-331 were resuspended in DMSO to obtain 10 mg/mL stock solutions. The structures of both compounds are depicted in Figure 1.

M. smegmatis growth conditions

The reference strain M. smegmatis mc²155 (ATCC^® 700084™) was used in this study. The strain was grown at 37°C, 180 rpm, in Middlebrook 7H9 (Becton, Dickinson and Company, Sparks, MD) supplemented with 10% (v/v) of oleic acid–albumin–dextrose–catalase (OADC, Becton, Dickinson and Company) and 0.05% (v/v) Tween 80 (Sigma, for fluorometric assays only).

MIC determination

MICs for EtBr, antibiotics, and compounds were determined by the twofold broth microdilution method.⁴¹ An inoculum equivalent to McFarland 0.5 standard was prepared from a midexponential culture (optical density [OD]₆₀₀ of 0.8) in 7H9. A final inoculum was then prepared by diluting 1:100 the cellular suspension in 7H9 plus 10% (v/v) OADC. Aliquots of 0.1 mL were transferred to each well of a 96-well plate containing 0.1 mL of twofold serial dilutions of each antimicrobial in 7H9 plus 10% (v/v) OADC. The MIC values were registered as the lowest concentration of antimicrobial that inhibited visible growth after an incubation of 72 hs at 37°C.

Evaluation of Efflux Inhibitory Activity

MIC determination in the presence of compounds

To evaluate the effect of compounds on the MICs of antibiotics and EtBr, MICs were redetermined in the presence of the compounds at one-fourth their MIC, a concentration by which their direct antimicrobial activity is residual and guarantees cell viability. MICs were determined as described above with an additional step consisting in the transfer of an aliquot of 0.01 mL of each compound to the wells of the microplate already containing twofold serial dilutions of each antimicrobial. Plates were inoculated and the MICs were registered as described above. MIC reductions equal or higher than fourfold were considered indicative of the presence of efflux inhibitory activity.^4,6

Synergism testing by “checkerboard assays”

The combination assays between antibiotics/EtBr and compounds were prepared in 96-well plates containing in the x-axis 0.05 mL of twofold dilutions of an antimicrobial ranging from 2 × up to 1/256× the MIC and in the y-axis 0.05 mL of twofold dilutions of a compound ranging from the 1× to 1/32× the MIC. The dilutions of the antimicrobials and compounds were prepared in 7H9 medium. For each assay, controls were included consisting in the testing of each antimicrobial and compound alone. The plates were inoculated with an inoculum equivalent to the one used for MIC determination. The combination results were registered visually after incubation at 37°C for 72 h. The effect of each compound was evaluated by determination of the fractional inhibitory concentration (FIC) calculated by the following formula: FIC_{antimicrobial} = MIC_{antimicrobial in combination}/MIC_{antimicrobial alone}. Values of FIC ≤0.25 were considered as synergy, FIC ≥2 as antagonism, and FIC = 0.5–1.0 as indifference.^4,13,23,42

Real-time fluorometry

The accumulation/efflux of EtBr was assessed in a Rotor-Gene™ 3000 (Corbett Research, Sydney, Australia) using real-time software analysis, as previously described by Viveiros et al. ³² The assays were conducted at 37°C and the fluorescence of EtBr was measured using 530 and 585 nm as excitation and detection wavelengths, respectively. The EtBr accumulation and efflux assays were performed as previously described.^19,35

(i) EtBr accumulation assays

Cultures were grown to midexponential growth phase (OD₆₀₀ of 0.8) and the cells were collected by centrifugation, washed in phosphate-buffered saline pH7.4 (PBS, Sigma), and resuspended in PBS to achieve a cellular suspension with an OD₆₀₀ of 0.8. Several reactions were prepared in 0.1 mL (final volume) containing 0.05 mL of the cellular suspension (final OD₆₀₀ of 0.4) plus 0.05 mL of (i) PBS, (ii) 2 × EtBr solution, and (iii) 2 × EtBr solution plus 2× compound solution. All solutions were prepared in PBS and the final concentrations of the compounds were equal to one-fourth their MIC. The EtBr concentration used, 0.125 mg/L, corresponds to steady-state equilibrium, in which EtBr efflux equals EtBr influx.¹⁹ If necessary, the accumulation assays can also be carried out with glucose as source of energy, as described previously.^32,43 The EtBr fluorescence was measured each minute for a total period of 60 min. The activity of the compounds on EtBr accumulation was evaluated according to the relative final fluorescence (RFF) values, calculated by the formula: RFF = (F_{plus compound}–F_{no compound})/F_{no compound}, where F_{plus compound} and F_{no compound} correspond to the fluorescence at the last time point (min 60) of the EtBr accumulation curve obtained in the presence of a compound and in its absence, respectively.²³ These accumulation assays, and consequently the RFF parameter, allow differentiating intraassay and interassay the relative strength of several compounds.

(ii) EtBr efflux assays

Cultures were grown in the same conditions and a cellular suspension was prepared similarly but adjusted to an OD₆₀₀ of 0.4. EtBr-loaded cells were prepared by incubating the cellular suspension in the presence of the EtBr steady-state concentration (0.125 mg/L) and the most efficient efflux inhibitor (VP) at 200 mg/L (one-fourth the MIC) for 1 h at room temperature to reach maximum EtBr loading. Cells were then collected by centrifugation and resuspended in PBS to achieve an OD₆₀₀ of 0.8. Several tubes were prepared in 0.1 mL containing 0.05 mL of cellular suspension and 0.05 mL of (i) PBS, (ii) glucose 0.8%, (iii) 2× VP solution, (iv) 2× VP solution plus glucose 0.8%, (v) 2× compound solution, and (vi) 2× compound solution plus glucose 0.8%. All solutions were prepared in PBS and the final concentrations of compounds were equal to one-fourth their MIC. The tube with glucose (ii) corresponds to the maximum efflux condition, whereas the tube with VP (iii) corresponds to the minimum efflux condition. The EtBr fluorescence was measured each 30 s for a total period of 30 min. The raw data obtained were normalized against the data of non-effluxing cells (control with VP only) at each time point; these time points correspond to the maximum fluorescence values that can be achieved during the assay. Thus, the relative fluorescence corresponds to the ratio of fluorescence that remains per unit of time relative to the EtBr-loaded cells. For each efflux assay, we considered that the assay run in the presence of glucose only corresponded to the conditions of maximum efflux, whereas the assays run in the presence of VP only corresponded to minimum efflux.^32,43 The efflux assays confirm the responsiveness of the efflux systems and provide an insight into the inhibitory strength of a compound relative to a strong efflux inhibitor—VP.

Results and Discussion

The aim of this study was to design and validate a methodological approach to enable the rapid screening and identification of potential antimycobacterial agents or adjuvants that can be used to potentiate the chemotherapy toward infections caused by mycobacteria (Table 1 and Fig. 2).

Fig. 2.

Algorithm designed for the rapid screening of new compounds with potential applications against mycobacteria.

Table 1.

Protocol Design for the Identification of Potential Antimycobacterial Adjuvants

Step	Parameter	Value	Description
1	MIC_c	20 μM	Twofold broth microdilution method
2	MIC_{antibacterial} w/wo compound	fourfold reduction	Twofold broth microdilution method in the presence/absence of compound at one-fourth the MIC_c
3	RFF	1	Real-time fluorometric detection of EtBr accumulation in the presence of compounds at one-fourth MIC (in Rotor-Gene™ 3000)
4	FIC	0.25	Checkerboard synergy assays in 96-well plates
5	EtBr efflux	Higher than the negative control	Real-time fluorometric detection of EtBr efflux in the presence of compounds at one-fourth MIC (in Rotor-Gene 3000)

Step Notes

1. An MIC_c <20 μM indicates a potential antimycobacterial.

2. An MIC_antibiotic or MIC_EtBr reduction by, at least, fourfold is indicative of a potential adjuvant.

3. An RFF >1 and MIC_EtBr reduction by at least fourfold indicate a potential adjuvant by efflux inhibition.

4. An FIC ≤0.25 indicates synergy between antibiotic/EtBr and the potential adjuvant.

5. Confirms an activity of EtBr efflux inhibition.

FIC, fractional inhibitory concentration; MIC, minimum inhibitory concentration; RFF, relative final fluorescence.

Rationale of the Methodological Model

In general, the synthesis of a given compound or set of compounds follows a rationale to achieve a specific functional goal, such as antimicrobial activity or inhibition of efflux. In our experience, a pre-established functional classification of a compound often does not reflect the biological outcome, potentially originating confusion in the attribution of a given potential use to a compound, namely as an antibacterial, adjuvant, and/or efflux inhibitor. These are very important functional characteristics that display some overlapping properties. For example, a compound may demonstrate a good adjuvant effect with established antibacterial agents through the inhibition of efflux pumps or by other unrelated mechanisms. However, this same compound at concentrations near its MIC will act as an antibacterial itself.

The methodological flowchart presented here proposes an approach in which the different biological effects of a given compound are evaluated in successive but nonexclusive steps, yielding maximum functional information for that molecule and its potential use. This approach allows optimizing several resources associated with the evaluation of a new molecule, such as time, laboratorial costs, and amount of the compound available for biological testing.

The design of this flowchart was based on establishing cutoff values for classifying compounds in different categories: antimycobacterial agents, adjuvants by inhibition of efflux activity, adjuvants by an unrelated mechanism, and compounds with no biological interest (Fig. 2). The cutoff values considered were already described in the literature or arbitrarily defined in this work. A compound is classified as a potential antimycobacterial agent when displaying an MIC lower than 20 μM, a value adapted from previous studies;²² this cutoff value allows identifying either a molecule with significant antibacterial properties or a lead molecule to be exploited for further development. Compounds that do not meet this criterion, but are able to reduce by at least fourfold the MIC values of either EtBr or established antimycobacterial agents, are considered as potential adjuvants. If this fourfold MIC reduction is observed for EtBr, together with significant intracellular accumulation of EtBr in fluorometric assays (arbitrarily defined as RFF values higher than 1), then the adjuvant properties of the compound are attributable to inhibition of efflux. In contrast, if the compound promotes reduction of antimycobacterial MICs, shows an RFF value less than 1, and an FIC equivalent to or less than 0.25, then it is classified as a non efflux inhibitor adjuvant.

Therefore, steps 1 to 3 of the proposed flowchart allow the classification of compounds in the established functional categories and the exclusion of compounds with no relevant effect (Fig. 2). The following steps, 4 and 5, will then confirm this classification and further explore the molecule's biological effect(s). Compounds displaying relevant biological properties will be selected for further studies in pathogenic mycobacteria, to be complemented with detailed mechanistic studies, assaying, for example, cytotoxicity, mutagenicity, enhancement of killing activity in human macrophages, and pK/pD studies in light of their potential future clinical use.^13,25,34,44

Validation of the Methodological Model

To validate the methodological flowchart, three compounds known to work as efflux inhibitors were analyzed according to the established process: the phenothiazines TZ and CPZ, and VP. Phenothiazines have long been recognized as presenting moderate antibacterial effects as well as moderate efflux inhibitory capacity against mycobacteria, whereas VP is considered a potent efflux inhibitor although without antibacterial effect per se.^{6,8,15,34
–36,45}

After validating the algorithm (Fig. 2), the compounds were first evaluated by determination of their MICs against the M. smegmatis strain in study (step 1)—Table 2. These MIC values will also define the concentration of compound to be used in subsequent steps—equivalent to one-fourth the MIC—to guarantee no effect upon cell viability during the time frame of the experiments. The phenothiazines CPZ and TZ showed MICs lower than MICs of VP, as expected (Table 2). However, taking into consideration the cutoff value established (20 μM), none of the compounds would be classified as a potential antibacterial.

Table 2.

MIC Values (mg/L and μM) of the Five Compounds Tested for M. smegmatis mc²155

	MIC
Compound	mg/L	μM
TZ	15	37
CPZ	60	169
VP	800	1630
UPAR-109	8	19.5
UPAR-331	>64	>130.8

TZ, thioridazine; CPZ, chlorpromazine; VP, verapamil.

Next, the three compounds were evaluated according to their potential (at one-fourth the MIC) to reduce the MIC values of relevant antibiotics and/or EtBr (step 2) or to promote intracellular accumulation of EtBr (step 3). Reduction of the MIC value of an antibacterial agent by at least fourfold is considered significant to attribute a potential adjuvant property to a given compound.^4,6 Overall, in step 2, the three compounds showed similar effects, except for EtBr whose MIC was most affected by VP (16-fold reduction vs. 4-fold decrease with TZ or CPZ) and STR, which was significantly affected only by TZ (Table 3). The MICs for the macrolides CLR and ERY were equally affected by the three compounds. The capacity of the three compounds to promote EtBr accumulation was tested by fluorometric assays (step 3)—Figure 3. As expected from the literature and the previous results of step 2, VP demonstrated a higher ability to promote EtBr accumulation than TZ or CPZ. This graphic observation is supported by the RFF values determined for each compound (Table 4), with VP presenting RFF values higher than 1, in opposition to TZ and CPZ.

Fig. 3.

Effect of the five compounds, at one-fourth their minimum inhibitory concentration, on the accumulation of EtBr [0.125 mg/L] for M. smegmatis mc²155.

Table 3.

Effect of the Compounds Tested on the MIC Values (mg/L) of Selected Antibacterial Agents for M. smegmatis mc²155

	MIC
Antibacterial	No compound	+TZ	+CPZ	+VP	+UPAR-109	+UPAR-331
EtBr^a	12.5	3.13 (4↓)	3.13 (4↓)	0.78 (16↓)	12.5 (-)	1.56 (8↓)
CIP	0.25	0.125 (2↓)	0.125 (2↓)	0.125 (2↓)	0.25 (-)	0.125 (2↓)
OFX	0.5	0.5 (-)	0.5 (-)	0.5 (-)	0.5 (-)	0.5 (-)
CLR	8	2 (4↓)	2 (4↓)	2 (4↓)	8 (-)	4 (2↓)
ERY	256	64 (4↓)	64 (4↓)	64 (4↓)	256 (-)	64 (4↓)
STR	0.5	0.125 (4↓)	0.25 (2↓)	0.25 (2↓)	0.5 (-)	0.125 (4↓)
RIF	16	16 (-)	16 (-)	8 (2↓)	16 (-)	8 (2↓)
INH	32	16 (2↓)	16 (2↓)	32 (-)	32 (-)	32 (-)
EMB	1	2 (2↑)	>2 (>2↑)	>2 (>2↑)	1 (-)	1 (-)

The compounds were used at one-fourth their MIC values: TZ: 3.75 mg/L (9.25 μM); CPZ: 15 mg/L (42.3 μM); VP: 200 mg/L (408 μM); UPAR-109: 2 mg/L (4.9 μM); UPAR-331: 32 mg/L (65.4 μM), assuming an MIC of, at least, 128 mg/L (261.6 μM). ↓: n-fold MIC reduction; ↑: n-fold MIC increase; (-) no change in MIC value. Values in bold type correspond to reductions in MIC values ≥fourfold.

EtBr was tested as a marker for efflux activity.

EtBr, ethidium bromide; CIP, ciprofloxacin; OFX, ofloxacin; CLR, clarithromycin; ERY, erythromycin; STR, streptomycin; RIF, rifampicin; INH, isoniazid; EMB, ethambutol; TZ, thioridazine; CPZ, chlorpromazine; VP, verapamil.

Table 4.

RFF Values for the Five Compounds Tested Against M. smegmatis mc²155

Compound	RFF
TZ	0.44 ± 0.01
CPZ	0.69 ± 0.01
VP	1.69 ± 0.02
UPAR-109	−0.02 ± 0.05
UPAR-331	1.90 ± 0.10

The compounds were used at one-fourth their MIC values: TZ: 3.75 mg/L (9.25 μM); CPZ: 15 mg/L (42.3 μM); VP: 200 mg/L (408 μM); UPAR-109: 2 mg/L (4.9 μM); and UPAR-331: 32 mg/L (65.4 μM), assuming an MIC of at least 128 mg/L (261.6 μM).

TZ, thioridazine; CPZ, chlorpromazine; VP, verapamil.

At this point, the three compounds were categorized according to the established model (Fig. 2). VP was clearly classified as an efflux inhibitor according to proposed criteria, showing an MIC value more than 20 μM, a capacity to reduce by 16-fold the MIC of EtBr, and presenting RFF values more than 1 (1.69 ± 0.02). Conversely, TZ and CPZ displayed MIC values more than 20 μM, a moderate capacity to reduce MICs of antibiotics and EtBr, presenting RFF values less than 1 (mild efflux inhibitory activity), being putatively classified as potential adjuvants by a mechanism not only/other than efflux inhibition. This classification befits the class of phenothiazines, for which different functional roles have been described in the literature.^18,46 In fact, they have been described as inhibitors of the enzyme type-II NADH-menaquinone dehydrogenase, essential in the M. tuberculosis respiratory chain,⁴⁷ as inhibitors of calcium–calmodulin binding,⁴⁸ and as inhibitors of mycobacterial efflux pumps.^6,19,34,35 In addition, phenothiazines have also been described as causing changes in the mycobacterial cell membrane fluidity and homogeneity^49,50 and affecting DNA replication and repair processes.^51,52 The results obtained reflect this hybrid effect, particularly for TZ, because, despite its relatively low MIC (37 μM), it is able to reduce by fourfold the MIC of EtBr as well as other effluxable antibiotics, although showing an RFF less than 1 (0.44 ± 0.01).

The assessment of a synergistic effect between these three compounds and antibacterial agents was performed by checkerboard assays (step 4). Confirming our previous classification, VP showed to be effective in reducing the MICs of EtBr, presenting FIC values of 0.25 or less at concentrations as low as 1/32 of the MIC (Supplementary Fig. S2, Table 5 and Supplementary Table S1). VP also presented significant synergy with ERY, an effluxable antibiotic.⁵³ CPZ and TZ presented a similar synergistic effect with ERY, but no significant effect with EtBr. No synergistic effect was observed with other antibiotics (Supplementary Fig. S2 and Supplementary Table S1). These results highlight the adjuvant properties of all three compounds. Fluorometric EtBr efflux assays (step 5) confirmed the high efflux inhibitory activity of VP in comparison with the two phenothiazines—Figure 4.

Fig. 4.

Effect of the compounds on the efflux of EtBr of M. smegmatis mc²155 strain in the absence (A) and presence (B) of 0.4% glucose.

Table 5.

FIC Values Corresponding to the Effect of Thioridazine, Chlorpromazine, Verapamil, and UPAR-331 on the MIC Values of Macrolides and EtBr for M. smegmatis mc²155

	FIC
	ERY			CLR			EtBr
[Compound] ( × MIC)	TZ	CPZ	VP	TZ	CPZ	VP	TZ	CPZ	VP	UPAR-331
1	—	—	—	—	—	—	—	—	—	—
1/2	0.0625	—	—	0.5	—	—	0.5	—	0.016	—
1/4	0.125	0.125	0.125	0.5	0.25	0.5	0.5	0.25	0.06	0.125
1/8	0.25	0.125	0.125	1	0.5	0.5	0.5	0.5	0.125	0.125
1/16	0.25	0.25	0.125	1	0.5	0.5	0.5	0.5	0.25	0.125
1/32	0.5	0.25	0.25	1	1	1	0.5	0.5	0.25	0.25

Values in bold type are indicative of a synergistic effect (FIC ≤0.25) between the compound and the antibacterial agent.

ERY, erythromycin; TZ, thioridazine; CPZ, chlorpromazine; VP, verapamil; —, FIC value could not be determined.

Application of the Model to New Compounds

We also applied the methodological model developed to two new molecules of the 2-aminothiazoles group, a substituted 4-phenyl-2-aminothiazole (UPAR-109),¹⁴ and a (thiazol-4-yl)isoxazole-3-carboxamide (UPAR-331). Following the rationale established, the first step of the evaluation process was the determination of the MIC of compounds against M. smegmatis mc²155 (step 1)—Table 2. UPAR-331 showed no antimycobacterial activity (MIC more than 130 μM), whereas UPAR-109 presented mild antimycobacterial activity (MIC = 19.5 μM), therefore, it was preliminarily classified as a potential antimycobacterial. The compounds were then evaluated regarding their ability to reduce the MICs of antibiotics and EtBr (step 2)—Table 3. UPAR-109 failed to show any interesting synergistic activity, whereas UPAR-331 decreased the MIC of EtBr as well as the MICs of ERY and STR by at least eightfold and fourfold, respectively. The evaluation of the capability to promote intracellular accumulation of EtBr (step 3) revealed distinct effects for the two molecules—Figure 3 and Table 4. The data obtained highlights the ability of UPAR-331 to augment significantly EtBr accumulation inside the mycobacterial cells. Interestingly, this effect was higher than that displayed by VP and was shown to be concentration dependent (Supplementary Fig. S3 and Supplementary Table S2). To further explore the relative strength of the compounds in promoting intracellular accumulation of EtBr, all compounds were tested at equivalent concentrations, 10 μM (data not shown) and 20 μM (Fig. 5). At both concentrations tested, UPAR-331 demonstrated a higher effect than VP, TZ, or CPZ. Molecule UPAR-109 showed no effect upon intracellular accumulation of EtBr.

Fig. 5.

Effect of thioridazine, chlorpromazine, verapamil, and UPAR-331, at 20 μM, on the accumulation of EtBr [0.125 mg/L] for M. smegmatis mc²155 strain.

In summary, the model developed allowed us to establish that compound UPAR-109 is a potential antimycobacterial agent with no efflux inhibitory activity. This is in agreement with the data already reported for this compound and the related series,¹² supporting the categorization proposed with our methodological model. In contrast, compound UPAR-331 is a potential adjuvant that exerts its activity by inhibition of efflux. Synergy testing by checkerboard synergy assays (step 4) was performed for UPAR-331 and EtBr, revealing that this molecule is able to reduce the MIC of EtBr (FIC ≤0.25) at concentrations as low as 1/32 its MIC (Table 5). Synergy testing by checkerboard assays (step 4) was not performed for UPAR-109 because of the lack of effect in step 2. The efflux inhibition properties of the thiazoisoxazole were confirmed in fluorometric EtBr efflux assays (step 5). Indeed, compound UPAR-331 demonstrated, at one-fourth the MIC, a remarkable efflux inhibitory activity, higher than that of VP, as evaluated visually and by the RFF value (1.90 ± 0.10) (Fig. 4 and Table 4). This effect confirmed the highest activity of UPAR-331 observed when both molecules were tested at the same concentration (20 μM) in an EtBr accumulation assay (Fig. 5).

Conclusions

In this work we describe a novel methodological approach to test and screen new molecules aimed at identifying new agents either endowed with antimycobacterial or with adjuvant properties. The results obtained with known compounds contributed to validate the algorithm that was, therefore, successfully applied to newly designed molecules, so as to corroborate this approach as an effective strategy to aid in the rapid screening of compounds that hold promise in antimycobacterial therapy. The methodologies applied are simple, time-effective, and cost-effective, when compared with the direct evaluation of compounds on pathogenic species such as M. tuberculosis, which have higher requirements in terms of biosafety, quantity of compounds, and time consumption. Genetic analyses have indicated that M. smegmatis and M. tuberculosis share many efflux systems, including 8 out of 12 M. tuberculosis ABC transporters, 4 out of 8 MFS transporters, and 6 out of 7 RND transporters.⁵⁴ Nevertheless, the efflux systems present in the two species are not completely superimposable and, together with differences in the cellular envelope, require some caution in the transposition of the results to M. tuberculosis. Once a compound is selected for further assays, it can be tested in M. tuberculosis or other species of interest by following the same experimental rationale, now adjusted for the new species, by using appropriate cutoff values and control strains. The advantages of the model were highlighted in the evaluation and categorization of a set of new compounds, enabling a simple decision-making process that leads to the identification of a potential antimycobacterial agent and a new potential adjuvant molecule with efflux inhibitory activity, even when the quantity of compound available is limited. These compounds may be interesting lead molecules and set the basis for the rational design of new molecules with improved properties.¹³

Footnotes

Acknowledgments

This work has been partially supported by Fundação para a Ciência e a Tecnologia (FCT), Portugal, through projects PTDC/BIA-MIC/121859/2010 and UID/Multi/04413/2013. S.S.C. and D.M. were supported by grants SFRH/BPD/97508/2013 and SFRH/BPD/100688/2014, respectively, from FCT. The Centro Interdipartimentale Misure “G. Casnati” is kindly thanked for the analytical determination of the molecules synthesized.

Disclosure statement

No competing financial interests exist.

Abbreviations Used

References

World Health Organization (WHO): Antimicrobial Resistance: Global Report on Surveillance 2014. Geneva, Switzerland, 2014.

Blair

, Webber

, Baylay

, Ogbolu

, Piddock

: Molecular mechanisms of antibiotic resistance. Nat Rev Microbiol, 2015; 13:42–51.

Poole

: Efflux pumps as antimicrobial resistance mechanisms. Ann Med, 2007; 39:162–176.

Coelho

, Machado

, Couto

, et al.: Enhancement of antibiotic activity by efflux inhibitors against multidrug resistant Mycobacterium tuberculosis clinical isolates from Brazil. Front Microbiol, 2015; 6:330.

Escribano

, Rodríguez

, Llorca

, García-Pachon

, Ruiz

, Royo

: Importance of the efflux pump systems in the resistance of Mycobacterium tuberculosis to fluoroquinolones and linezolid. Chemother, 2007; 53:397–401.

Machado

, Couto

, Perdigão

, et al.: Contribution of efflux to the emergence of isoniazid and multidrug resistance in Mycobacterium tuberculosis . PLoS One, 2012; 7:e34538.

Ramón-García

, Martín

, Thompson

, Aínsa

: Role of the Mycobacterium tuberculosis P55 efflux pump in intrinsic drug resistance, oxidative stress responses, and growth. Antimicrob Agents Chemother, 2009; 53:3675–3682.

Rodrigues

, Machado

, Couto

, Amaral

, Viveiros

: Contribution of efflux activity to isoniazid resistance in the Mycobacterium tuberculosis complex. Infect Genet Evol, 2012; 12:695–700.

Rodrigues

, Villellas

, Bailo

, Viveiros

, Aínsa

: Role of the Mmr efflux pump in drug resistance in Mycobacterium tuberculosis . Antimicrob Agents Chemother, 2013; 57:751–757.

10.

Worthington

, Melander

: Combination approaches to combat multi-drug resistant bacteria. Trends Biotechnol, 2013; 31:177–184.

11.

Gupta

, Tyagi

, Bishai

: Verapamil increases the bactericidal activity of bedaquiline against Mycobacterium tuberculosis in a mouse model. Antimicrob Agents Chemother, 2015; 59:673–676.

12.

Pieroni

, Wan

, Cho

, Franzblau

, Costantino

: Design, synthesis and investigation on the structure-activity relationships of N-substituted 2-aminothiazole derivatives as antitubercular agents. Eur J Med Chem, 2014; 72:26–34.

13.

Pieroni

, Machado

, Azzali

, et al.: Rational design and synthesis of thioridazine analogues as enhancers of the antituberculosis therapy. J Med Chem, 2015; 58:5842–5853.

14.

Pieroni

, Wan

, Zuliani

, Franzblau

, Costantino

, Rivara

: Discovery of antitubercular 2,4-diphenyl-1H-imidazoles from chemical library repositioning and rational design. Eur J Med Chem, 2015; 100:44–49.

15.

Pule

, Sampson

, Warren

, et al.: Efflux pump inhibitors: targeting mycobacterial efflux systems to enhance TB therapy. J Antimicrob Chemother, 2016; 71:17–26.

16.

Soares de Melo

, Feng

, van der Westhuyzen

, et al.: Aminopyrazolo[1,5-a]pyrimidines as potential inhibitors of Mycobacterium tuberculosis: structure activity relationships and ADME characterization. Bioorg Med Chem, 2015; 23:7240–7250.

17.

Tegos

, Haynes

, Strouse

, et al.: Microbial efflux pump inhibition: tactics and strategies. Curr Pharm Des, 2013; 487:109–113.

18.

Viveiros

, Martins

, Rodrigues

, et al.: Inhibitors of mycobacterial efflux pumps as potential boosters for anti-tubercular drugs. Expert Rev Anti Infect Ther, 2012; 10:983–998.

19.

Rodrigues

, Ramos

, Couto

, Amaral

, Viveiros

: Ethidium bromide transport across Mycobacterium smegmatis cell-wall: correlation with antibiotic resistance. BMC Microbiol, 2011; 11:e35.

20.

, Villellas

, Koul

, et al.: The ATP synthase inhibitor bedaquiline interferes with small-molecule efflux in Mycobacterium smegmatis . J Antibiot (Tokyo), 2014; 67:835–837.

21.

Luthuli

, Purdy

, Balagaddé

: Confinement-induced drug-tolerance in Mycobacteria mediated by an efflux mechanism. PLoS One, 2016; 10:e0136231.

22.

Altaf

, Miller

, Bellows

, O'Toole

: Evaluation of the Mycobacterium smegmatis and BCG models for the discovery of Mycobacterium tuberculosis inhibitors. Tuberculosis, 2010; 90:333–337.

23.

Machado

, Cannalire

, Costa

, et al.: Boosting effect of 2-phenylquinoline efflux inhibitors in combination with macrolides against Mycobacterium smegmatis and Mycobacterium avium . ACS Infect Dis, 2015; 1:593–603.

24.

Lechner

, Gibbons

, Bucar

: Plant phenolic compounds as ethidium bromide efflux inhibitors in Mycobacterium smegmatis . J Antimicrob Chemother, 2008; 62:345–348.

25.

Ngwane

, Panayides

, Chouteau

, et al.: Design, synthesis, and In vitro antituberculosis activity of 2(5H)-Furanone derivatives. IUBMB Life, 2016; 68:612–620.

26.

Jin

, Zhang

, Guo

, et al.: The plant alkaloid piperine as a potential inhibitor of ethidium bromide efflux in Mycobacterium smegmatis . J Med Microbiol, 2011; 60:223–229.

27.

Bansal-Mutalik

, Nikaido

: Mycobacterial outer membrane is a lipid bilayer and the inner membrane is unusually rich in diacyl phosphatidylinositol dimannosides. Proc Natl Acad Sci U S A, 2014; 111:4958–4963.

28.

Andries

, Verhasselt

, Guillemont

, et al.: A diarylquinoline drug active on the ATP synthase of Mycobacterium tuberculosis . Science, 2005; 307:223–227.

29.

Chaturvedi

, Dwivedi

, Tripathi

, et al.: Evaluation of Mycobacterium smegmatis as a possible surrogate screen for selecting molecules active against multi-drug resistant Mycobacterium tuberculosis . J Gen Appl Microbiol, 2007; 53:333–337.

30.

Costa

, Viveiros

, Rosato

, Melo-Cristino

, Couto

: Impact of efflux in the development of multidrug resistance phenotypes in Staphylococcus aureus . BMC Microbiol, 2015; 15:e232.

31.

Mullin

, Mani

, Grossman

: Inhibition of antibiotic efflux in bacteria by the novel multidrug resistance inhibitors biricodar (VX-710) and Timcodar (VX-853). Antimicrob Agents Chemother, 2004; 48:4171–4176.

32.

Viveiros

, Rodrigues

, Martins

, et al.: Evaluation of efflux activity of bacteria by a semi-automated fluorometric system. In: Antibiotic Resistance Protocols. Gillespie

, McHugh

(eds., 2nd ed.), pp. 159–172 [In: Methods in Molecular Biology, vol. 642] Humana Press, New York, 2010.

33.

, Zhang

, Guo

, et al.: Study of efflux pump gene expression in rifampicin-monoresistant Mycobacterium tuberculosis clinical isolates. J Antibiotics, 2015; 68:431–435.

34.

Machado

, Pires

, Perdigão

, et al.: Ion channel blockers as antimicrobial agents, efflux inhibitors, and enhancers of macrophage killing activity against drug resistant Mycobacterium tuberculosis . PLoS One, 2016; 11:e0149326.

35.

Rodrigues

, Wagner

, Viveiros

, et al.: Thioridazine and chlorpromazine inhibition of ethidium bromide efflux in Mycobacterium avium and Mycobacterium smegmatis . J Antimicrob Chemother, 2008; 61:1076–1082.

36.

Rodrigues

, Sampaio

, Couto

, et al.: The role of efflux pumps in macrolide resistance in Mycobacterium avium complex. Int J Antimicrob Agents, 2009; 34:529–533.

37.

Lamoureux

, Artavia

: Use of the adamantane structure in medicinal chemistry. Curr Med Chem, 2010; 17:2967–2978.

38.

Lun

, Guo

, Onajole

, et al.: Indoleamides are active against drug-resistant Mycobacterium tuberculosis . Nat Commun, 2013; 4:2907.

39.

Grzegorzewicz

, Pham

, Gundi

, et al.: Inhibition of mycolic acid transport across the Mycobacterium tuberculosis plasma membrane. Nat Chem Biol, 2012; 19:334–341.

40.

Tahlan

, Wilson

, Kastrinsky

: SQ109 targets MmpL3, a membrane transporter of trehalose monomycolate involved in mycolic acid donation to the cell wall core of Mycobacterium tuberculosis . Antimicrob Agents Chemother, 2012; 56:1797–1809.

41.

Clinical and Laboratory Standards Institute (CLSI): Performance Standards for Antimicrobial Susceptibility Testing of Mycobacteria, Nocardie and Other Aerobic Actynomicetes. CLSI M24-A2, Wayne, PA, 2011.

42.

Pillai

, Moellering

, Eliopoulos George

: Antimicrobial combinations. In: Antibiotics in Laboratory Medicine. Lorian

(ed., 5th ed.), pp. 365–441. Wlliams & Wilkins, Baltimore, 2005.

43.

Paixão

, Rodrigues

, Couto

, et al.: Fluorometric determination of ethidium bromide efflux kinetics in Escherichia coli . J Biol Eng, 2009; 3:18.

44.

Szumowski

, Adams

, Edelstein

, et al.: Antimicrobial efflux pumps and Mycobacterium tuberculosis drug tolerance: evolutionary considerations. Curr Top Microbiol Immunol, 2013; 374:81–108.

45.

Amaral

, Viveiros

, Molnar

: Antimicrobial activity of phenothiazines. In Vivo, 2004; 18:725–732.

46.

Sharma

, Singh

: Phenothiazines as anti-tubercular agents: mechanistic insights and clinical implications. Expert Opin Investig Drugs, 2011; 20:1665–1676.

47.

Yano

, Li

, Weinstein

, Teh

, Rubin

: Steady-state kinetics and inhibitory action of antitubercular phenothiazines on Mycobacterium tuberculosis type-II NADH-menaquinone oxidoreductase (NDH-2). J Biol Chem, 2006; 281:11456–11463.

48.

Salih

, Kaushik

, Sharma

, Choudary

, Murthy

, Venkitasubramanian

: Calmodulin-like activity in mycobacteria. Indian J Biochem Biophys, 1991; 28:491–495.

49.

Jutila

, Söderlund

, Pakkanen

, Huttunen

, Kinnunen

: Comparison of the effects of clozapine, chlorpromazine, and haloperidol on membrane lateral heterogeneity. Chem Phys Lipids, 2001; 112:151–163.

50.

Rosso

, Zachowski

, Devaux

: Chlorpromazine induced changes in the lipid fluidity of erythrocyte cell membranes. Biochim Biophys Acta, 1988; 942:271–279.

51.

Lialiaris

, Pantazaki

, Sivridis

, Mourelatos

: Chlorpromazine-induced damage on nucleic acids: a combined cytogenetic and biochemical study. Mutat Res, 1992; 265:155–163.

52.

Viola

, Latterini

, Vedaldi

, et al.: Photosensitization of DNA strand breaks by three phenothiazine derivatives. Chem Res Toxicol, 2003; 16:644–651.

53.

Louw

, Warren

, Gey van Pittius

, McEvoy

, Van Helden

, Victor

: A balancing act: efflux/influx in mycobacterial drug resistance. Antimicrob Agents Chemother, 2009; 53:3181–3189.

54.

Da Silva

, Machado

, Ramos

, et al.: Efflux pumps in Mycobacteria: antimicrobial resistance, physiological functions, and role in pathogenicity. In: Efflux-Mediated Antimicrobial Resistance in Bacteria. Li

X-Z

, Elkins

, Zgurskaya

(eds.). Springer International Publishing, Switzerland, 2016. In press. ISBN 978-3-319-39658-3.