Antibacterial Activity of combinatorial treatments composed of transition-metal/antibiotics against Mycobacterium tuberculosis

Notwithstanding evidence that tuberculosis (TB) is declining, one of the greatest concerns to public health is the emergence and spread of multi-drug resistant strains of Mycobacterium tuberculosis (MDR-TB). MDR-TB are defined as strains which are resistant to at least isoniazid (INH) and rifampicin, the two most potent TB drugs, and their increasing incidence is a serious concern. Recently, notable efforts have been spent on research to pursue novel treatments against MDR-TB, especially on synergistic drug combinations as they have the potential to improve TB treatment. Our research group has previously reported promising synergistic antimicrobial effects between transition-metal compounds and antibiotics in Gram-negative and Gram-positive bacteria. In this work, we evaluated antimycobacterial activity of transition-metals/antibiotics combinatorial treatments against first-line drug resistant strains of Mycobacterium tuberculosis. Our data showed that INH/AgNO3 combinatorial treatment had an additive effect (bactericidal activity) in an isoniazid-resistant clinical strain of Mycobacterium tuberculosis. Moreover, in vitro evaluation of cytotoxicity induced by both, the individual tratments of AgNO3 and INH and the combinatorial treatment of INH/AgNO3 in murine RAW 264.7 macrophages and human A549 lung cells; showed no toxic effects. Together, this data suggests that the INH/AgNO3 combinatorial treatment could be used in the development of new strategies to treat resistant strains of Mycobacterium tuberculosis.

Mycobacterium tuberculosis (M. tuberculosis) causes tuberculosis (TB), which is the leading cause of death by infectious diseases worldwide, with an estimated 10.4 million new TB cases in 2016 1 . One of the main challenges in TB drug development is that the treatment scheme is a combined regimen, not a single drug. Therefore, the research of new treatments for drug-resistant tuberculosis should be based on novel mechanisms of action relative to the current TB therapy, considering the least amount of undesirable interactions between drugs and the least possible number of side effects; this will lead to obtaining a treatment with a high therapeutic potential in patients with MDR-TB 2 . Consequently, in recent years the development of metallo-pharmaceuticals has increased since they are compounds that include metals due to their therapeutic action, and currently belong to a class of promising antimicrobial compounds aimed to overcome resistant strains [3][4][5][6][7][8][9][10] . Transition metal species, and especially silver compounds, lie among the most studied metallo-pharmaceuticals 11,12 .
Recent literature has reported on the antimycobacterial activity of transition-metals (ions, salts or complexes) in combination with antibiotics. They have reported interesting advances on the antimycobacterial effects of organic compounds; and antibiotics [13][14][15][16] in M. tuberculosis strain H37Rv, which is the most commonly used control for M. tuberculosis identification in the clinical and research laboratory setting, and drug-resistant clinical isolates of M. tuberculosis. In this study, we tested the antimycobacterial activity of transition-metals (CuSO 4 , AgNO 3 , ZnSO 4 , and NiSO 4 ) in combination with antibiotics against first-line anti-tuberculosis drug resistant isolates.

Results and Discussion
One of the central strategies of the tuberculosis control program is early detection of drug-resistant Mycobacterium tuberculosis strains. No clinical strains with a monoresistance profile to ethambutol (EMB) were found in the present study. It has been reported that the frequency EMB resistance is lower than that for other antimycobacterial agents. EMB is an alternative drug in the standard four-drug combination therapy since it prevents treatment failure in resistant strains to other antimicrobial agents (i.e. streptomycin) and avoids the risk of side effects 17 . Ziehl Neelsen stain showed that all the strains were acid-alcohol-resistant bacilli and they also presented slow growth, ability to produce niacin and nitrate reduction, which confirmed they were M. tuberculosis strains 18 .
We Antimycobacterial effect of transition-metal salts/first-line tB drugs combinatorial treatments. We tested the combination of transition-metal salts (that showed best inhibitory effect) with the corresponding drug to which each isolate was resistant, via a checkerboard assay, a widely used methodology to test synergistic effects between antimicrobial agents 19-22 . Our results showed that the transition-metal salts/drugs combinations did not inhibit cell viability for strains OxPS-22 ( Supplementary Fig. S1), OxPS-4 and OxPS-19 ( Supplementary Fig. S2); nevertheless, the INH/AgNO 3 combination showed a positive effect in strain 152589, since a reduction of 50% of the MIC of both treatments was achieved, from 0.25 µg/ml and 25 µM to 0.125 µg/ml and 12.5 µM, respectively. Regarding strain H37Rv, the INH/AgNO 3 combination showed the same effect observed in strain 152589 ( Supplementary Fig. S3).
To determine the interaction between two or more drugs intended to be used in combination, the Fractional Inhibitory Concentration (FIC) Index was used 23,24 . It was interpreted as follows: FIC index of 0.5 was considered for synergism, FIC index of 1 was defined as additive effect, and antagonism when FIC index was 2 or 4. For INH/ AgNO 3 combinatorial treatment the FIC was 1 which means an additive effect. Moreover, the effect of the INH/ AgNO 3 combination in strains 152589 and H37Rv was found to be bactericidal.
The INH/AgNO 3 combination can be considered as a potential antimycobacterial agent since we have shown it can inhibit bacterial growth at a lower concentration than each one as a separate treatment. These would allow decreasing in dose both, isoniazid and AgNO 3 , and at the same time reduce the potential toxic effects in mammalian cells. Some other authors have reported the use of transitions metals/drug complexes as antibacterial agents since the combination improves its antimycobacterial activity [25][26][27][28] .
It is well known that silver affects the permeability of the bacterial membrane 29 , whereas isoniazid is a pro-drug that requires activation by the catalase/peroxidase enzyme KatG, encoded by the KatG gene to exert its effect 30 . The active form acts by inhibiting the synthesis of mycolic acid through the NADH-dependent enoyl-acyl carrier protein (ACP)-reductase 31 . Nevertheless, the mechanism through which the INH/AgNO 3 combination affects the cell viability in an INH-resistant strain has not been elucidated.
Silver induces the folding of proteins that are secreted from the cytoplasm and transported to the outer membrane which can lead to membrane destabilization and increased permeability. Our research group has previously reported 7 that the combination of sublethal concentrations of antibiotics with silver salts alters multiple cellular processes, including the formation of disulfide bonds, central metabolism, iron homeostasis, and these changes are associated with an increase in the production of ROS and permeability of the bacterial membrane. Therefore, we hypothesize that the addition of sublethal doses of silver to antibiotic induces a marked increase in ROS, where silver contributes to the production of ROS, oxidative stress and bacterial cell death.
As a complementary study, cytotoxic effects at 24 and 48 h treatment with isoniazid, silver nitrate and the combinatorial treatment were tested in two relevant respiratory cells lines. The results of the cytotoxic evaluation www.nature.com/scientificreports www.nature.com/scientificreports/ of INH, AgNO 3 and the combinatorial treatment on cell viability of murine RAW 264.7 macrophages and human A549 lung cells is presented in Fig. 1. Our results indicate that INH did not induce any toxic effect on the cell viability of both A549 and RAW 264.7 cells after 24 and 48 h treatment. INH/AgNO 3 combinatorial treatment did not increase significantly cell viability in A549 cells. Concerning RAW 264.6 cells, a significant increase of cell viability was observed upon exposure to 50-6.2 mM AgNO 3 at 24 h (30% increase; p < 0.05). We observed same effect when RAW 264.7 cells were treated for 24 h with 0.25ug/mL-25mM and 0.12ug/mL-12.5 mM, INH/ AgNO 3 respectively. Nevertheless, we did not observe changes in growth after 48 h. This effect could be related to cellular stress which increased oxidative metabolism. Since the MTT assay is an assay for assessing cell metabolic activity, this would explain the observed increase in RAW 264.7 cells. These results suggest that INH, AgNO 3 and combinatorial treatment have no toxic effect on A549 lung epithelial cells and RAW 264.7 macrophage cells.

Conclusions
Mono-resistance to isoniazid is the most common first-line drug resistance in tuberculosis; therefore, it has been a challenge the development of more efficient and effective antimycobacterial drugs that show less toxicity against mammalian cells. Here, we have identified a combination of compounds (isoniazid/AgNO 3 ) with antimycobacterial activity against an isoniazid-resistant clinical strain (strain 152589) of M. tuberculosis. The combinatorial treatment had a significant additive effect at 0.125 µg/ml isoniazid and 12.5 µM AgNO3 (50% decrease of individual MICs) and in vitro evaluation of cytotoxicity in RAW 264.7 and A549 cells showed no toxic effects. We have previously described transition-metals key role in several cellular processes and their antimicrobial effects; therefore, the combination of antibiotics with transition metals results in an excellent treatment alternative since antibacterial effect is enhanced with the combination of drugs, the concentrations used of both drugs are reduced, and the possibility of adverse effects is considerably reduced, resulting in a better outcome for the patient. Thus, this combination could be used in the development of new strategies to treat resistant strains of Mycobacterium tuberculosis.

Methods
Biosafety criteria and Microorganisms. All procedures involving M. tuberculosis specimens were carried out in a Class II A2 biological safety cabinet in a BSL-2 level containment facility located at the Regional Center for the Control of Infectious Diseases (CRCEI), Faculty of Medicine, Autonomous University of Nuevo León. Twelve strains with previous report of resistance to first-line drugs were selected from databases from the Regional Center for the Control

strains reactivation and biochemical characterization. The selected strains were reactivated in
Middlebrook 7H9 broth added with OADC (oleic acid, albumin, dextrose and catalase) at 10%, 5% Tween 80 and 0.2% glycerol. After their reactivation, Ziehl-Neelsen stain was performed to identify acid-alcohol-resistant Results are expressed as mean values ± SD (three independent experiments, three replicates per experiment at each concentration). INH (isoniazid), AgNO 3 (silver nitrate), mock (untreated control), Etoposide (positive toxicity control). +y++ p < 0.05 was considered statistically significant. + refers to statistical significance with mock and ++ refers to statistical significance between the same time period.
www.nature.com/scientificreports www.nature.com/scientificreports/ bacilli. The twelve reactivated strains were phenotypically identified using biochemical assays of niacin accumulation and nitrate reduction previously described by Bernardelli et al. 18 . A positive control (M. tuberculosis H37Rv) and negative control (a tube with broth without inoculum) were included. Drug susceptibility testing. Antimicrobial susceptibility test for STR, INH, RIF, EMB was carried out by triplicate using critical concentration and minimum inhibitory concentration categories according to the recommendations of the World Health Organization in its technical report on the critical concentrations for drug susceptibility tests in medicines used in the treatment of drug-resistant tuberculosis 32 . The critical concentration of INH, RIF, EMB and STR were 0.2, 40, 4.0 and 2.0 μg/mL, respectively. The breakpoint concentrations (µg/ml) for resistance to STR, INH, RIF and EMB were defined as ≥4, 0.25, ≥0.5 and ≥4, respectively. Based on these results, from the twelve initially reactivated strains we selected strain 152589 (resistant to ISO), OxPs-4 (resistant to RIF), OxPs-19 (resistant to STR), and OxPs-22 (Multidrug-resistant; MDR) for subsequent experiments since they showed the resistance profile required. The rest of the strains were found to be sensitive therefore they were not considered for further analysis.

Determination of MIC for antibiotics and transition-metal salts.
MICs values were obtained using microplate Alamar Blue assay (MABA) as previously described 33 . Streptomycin, isoniazid, rifampicin and ethambutol, and metal salts: silver nitrate (AgNO 3 ), copper sulfate (CuSO 4 ), nickel sulfate (NiSO 4 ) and zinc sulfate (ZnSO 4 ) were evaluated. Clinic strains cultures were subcultured in Middlebrook 7H9 broth (added with 0.2% glycerol, 0.05% Tween 80 and 10% OADC) for 21 days at 37 °C. The resulting mycobacterial suspension was adjusted at a turbidity of 1.0 McFarland standard and then diluted 1:25 in Middlebrook 7H9 broth (added with 0.2% glycerol and 0.05% OADC). Antibiotics or transition-metal salts were aliquoted (100 μl) into the first row of wells of a 96-well microtiter plate in which wells were prefilled with 100 μl of Middlebrook 7H9 broth (added with 0.2% glycerol and 0.05% OADC), row 1 wells were mixed 8 to 10 times using a pipettor. Then, 100 μl was withdrawn and transferred to row 2. Row 2 wells were mixed 8 to 10 times, followed by a 100-μl transfer from row 2 to row 3. This procedure was used to serially dilute the rest of the rows of the microtiter plate. Finally, 100 μl of 1:25 diluted mycobacterial suspension was added into the wells containing specified antibiotic. The microtiter plate was incubated at 37 °C for 5 days.

Determination of MIC for transition-metals/antibiotics combinatorial treatments.
Checkerboard assays [19][20][21][22] were performed in 96-well polystyrene plates, in order to determine the antimycobacterial effects of transition-metals salts and first-line TB drugs. The MIC fractions (0, 0.5, 0.25, and 0.125) of transition-metals salts were prepared along the abscissa axis of the plate and the first-line TB drugs MIC fractions (0, 0.5, 0.25, and 0.125) were placed along the ordinate axis of the plate. Concentrated transition-metals salts (4×) and first-line TB drugs solutions (8×) were prepared in culture media, so that when added to the culture the needed volume of transition-metals salts, first-line TB drugs and inoculum MICs fractions was reached.
Clinical isolates cultures were grown for 21 days at 37 °C in Middlebrook 7H9 broth (added with 0.2% glycerol, 0.05% Tween 80 and 10% OADC) for 21 days at 37 °C. The resulting mycobacterial suspension was adjusted at a turbidity of 1.0 McFarland standard and then diluted 1:25 in Middlebrook 7H9 broth (added with 0.2% glycerol and 0.05% OADC). MIC fractions of 0.5, 0.25 and 0.125 of each transition-metal salt and first-line TB drug nominal concentrations were combined to achieve final concentrations in a final volume of 200 μL, including the bacteria inoculum. These combinations were diluted 2-fold and 100 μL were transferred each time after a thorough mixing and discarding the last 100 μL from the 3 rd dilution. The 96-well plates were incubated at 37 °C for 5 days. After incubation, 20 μl of AB solution and 12 μl of 10% sterile Tween 80 were added into drug control, transition-metal salt control, and negative and positive control. Plates were incubated again at 37 °C for 24 hours, after that, drug control, transition-metal salt control, and negative displayed no color change, while the positive control changed to pink, then we added 20 μl of AB solution and 12 μl of 10% sterile Tween 80 into the remaining wells and incubated again at 37 °C for 24 hours. The results were interpreted as non-viable cells if they displayed no color change or viable if they had the same intensity as the control well at 1%, or any shade of pink, violet or purple. The optical density (OD) of the control and the treated inoculums were measured, and the respective values were recorded. All tests and their respective control samples were performed in replicates of three.
Assay for cell viability. The effect of compounds in cell viability was determinate using MTT assay according to Freshney 34 . Briefly, RAW 264.7 or A549 cells (1 × 103 cell/ well in 200 µL of medium) were seeded in a 96-well plate and incubated for 24 h, the cells were treated with serial dilution of isoniazide (from 1 to 0.06 mg/ mL), AgNO3 (from 100 to 6 mM) prepared in PBS, a mix of both compounds, or etoposide (20 μM) and incubated (2019) 9:5471 | https://doi.org/10.1038/s41598-019-42049-5 www.nature.com/scientificreports www.nature.com/scientificreports/ for 24 or 48 h. Medium was changed and 0.2 ml of medium with 0.5 mg of MTT was added to each well, and the cells were incubated for another 4 h. The optical density was measured at 570 nm on a microplate reader. statistical analysis. The results were obtained at least by three independent experiments and are presented as means ± SD. Statistical analyses were performed by one-way analysis of variance (ANOVA) followed by t Student´s test. All statistical analyses were performed using the GraphPad Prism ® , version 6.0 software. P values ˂ 0.05 were considered to indicate statistical significance.