7/27/2019 screening on non culturable MTB.pdf
1/26
1
New 2-thiopyridines as potential candidates for killing both actively growing and1
dormantMycobacterium tuberculosis2
3
Elena Salina#, Olga Ryabova, Arseny Kaprelyants and Vadim Makarov4
5
Institution of theRussian Academy of Sciences A.N. Bach Institute of Biochemistry6
RAS, Moscow, Russia7
8
Running title: New 2-thiopyridines9
10
# Address for correspondence to Elena Salina, [email protected]
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
AAC Accepts, published online ahead of print on 14 October 2013
Antimicrob. Agents Chemother. doi:10.1128/AAC.01308-13
Copyright 2013, American Society for Microbiology. All Rights Reserved.
7/27/2019 screening on non culturable MTB.pdf
2/26
2
Abstract27
From in vivo observations, majority of M. tuberculosis cells in latently infected28
individuals are in a dormant and probably non-culturable state, display little metabolic29
activity and are phenotypically resistant to antibiotics. Despite many attempts no specific30
antimicrobials effective for latent tuberculosis have yet been found partially because of lack of31
reliable and adequate in vitromodels for screening for drug candidates. We propose here a32
novel in vitro model of M. tuberculosis dormancy which meets the important criteria of33
latency, namely non-culturability of cells, considerable reduction of metabolic activity and34
significant phenotypic resistance to the first-line antibiotics rifampicin and isoniazid. Using35
this model we found a new group of 2-thiopyridine derivatives which had potent antibacterial36
activity against both actively growing and dormant M. tuberculosis cells. By means of the37
model of M. tuberculosis non-culturability several new 2-thiopyridine derivatives were38
found to have potent antitubercular activity. The compounds are effective against both active39
and dormant M. tuberculosis cells. The bactericidal effect of compounds for dormant M.40
tuberculosiswas confirmed by using three different in vitromodels of tuberculosis dormancy.41
The model of non-culturability could be used as a reliable tool for screening drug candidates42
and 2-thiopyridine derivatives may be regarded as prominent compounds for further43
development of new drugs for curingM. tuberculosislatent infection.44
45
46
47
48
49
50
51
52
7/27/2019 screening on non culturable MTB.pdf
3/26
3
Introduction53
Curing latent tuberculosis (TB) is a big challenge for modern chemotherapy (1). One54
third of the entire world's population carries latent TB infection with a lifelong risk of disease55
reactivation, which increases several times in immunocompromised patients (1, 2).56
Mycobacterium tuberculosis cells associated with latent tuberculosis in the human host are57
believed to be in a special non-replicating (dormant) state characterized by the development of58
considerable antibiotic tolerance (2). This tolerance, described as phenotypic drug resistance,59
is due to changes in the physiological state of the bacteria characterized by the low metabolic60
status of cells (2).61
Since specific and highly effective TB drugs for latency are still absent, traditional62
antibiotics such as isoniazid (INH) and rifampicin (RIF) are used in current chemotherapy of63
latent TB. There are some therapeutic regimens based on long-term cure of latently infected64
patients by INH or RIF (3, 4) although INH is weakly effective for latent TB infection because65
it targets the processes of bacterial cell wall biosynthesis (1). Probably the role of INH may be66
to kill emerging bacilli that grow actively as a result of dormant cells reactivation or67
conversion of slowly growing persisters to actively growing bacilli susceptible to INH68
treatment (5). Significant tolerance of dormant cells to RIF was shown in the in vivoCornell69
model of TB persistence (6) thus questioning the efficacy of RIF to cure latent TB. Moreover,70
it was postulated that dormant M. tuberculosis cells are phenotypically resistant to the71
sterilizing activity of RIF and this feature is one of the hallmarks of dormant TB (7).72
Interestingly, in accordance with the Cornell model and some clinical studies, dormant73
cells in latently infected individuals are characterized by non-culturability i.e. a74
phenomenon of transient inability to divide and grow on non-selective solid media (8-10). A75
resuscitation procedure is required for such cells to enter into an active growth state (8, 11).76
Consequently, there is a need for new drugs against latent TB infection. Since genome-77
derived, target-based approaches have had little success in the antibacterial therapeutic area in78
7/27/2019 screening on non culturable MTB.pdf
4/26
4
general (1),whole bacterial cell screening is believed to be a preferable approach for finding79
new drugs for TB as it allows to test drug candidates in a relevant physiological state of the80
pathogen (1, 12).81
Several in vitromodels ofM. tuberculosisdormancy that were developed recently are82
currently being used to searching for anti-latency drugs (7, 13-15), however, dormant cells83
obtained in the vast majority of these in vitromodels are fully culturable and metabolically84
active and are characterized by limited antibiotic resistance that may compromise their use in85
the search for drugs for latent TB.86
We propose here a new in vitromodel that mimics the phenomenon of TB latency. In87
this model we managed to eliminate many of the limitations of the in vitromodels mentioned88
above. In particular, dormant cells in this model are non-culturable (NC), of low metabolic89
activity and characterized by phenotypic resistance both to INH and RIF. Therefore, this in90
vitromodel meets the key criteria of true dormancy and may be applied as a relevant tool for91
latent TB drug discovery. Using this new model we found a new group of 2-thiopyridine92
derivatives which had potent antibacterial activity against both actively growing and dormant93
M. tuberculosiscells.94
95
Materials and methods96
Chemistry97
All reagents and solvents were purchased from commercial suppliers and used without98
further purification. Melting points were determined on Electothermal 9001 and are99
uncorrected.1
H NMR was measured in DMSO-d6at 400 MHz on a Varian Unity +400 (USA).100
Shifts for NMR are reported in ppm downfield from TMS (). A Waters Micromass ZQ101
detector (USA) was used in ESI MS for identification of various products. Elemental analyses102
were carried out on a Carlo-Erba 5500 elemental analyzer for C, H, N, and the results are103
7/27/2019 screening on non culturable MTB.pdf
5/26
5
within 0.3% of the theoretical values. Merck silica gel 60 F254plates were used for analytical104
TLC; column chromatography was performed on Merck silica gel 60 (70-230 mesh).105
Compounds 11026101 and 10026127 were obtained by a series of successive106
transformations of 2-hydroxynicotinamide which was first reacted with fuming nitric acid in107
concentrated sulfuric acid to give the corresponding 5-nitroderivative. The reflux of the latter108
with excess thionyl chloride results in 2-chloro-5-nitronicotinic acid chloroanhydride which109
was immediately used in a reaction with 12% ammonia solution in water and the substituted110
nicotinamide was isolated with high yield. Subsequently, boiling with thionyl chloride111
resulted in the key intermediate 2-chloro-3-cyano-5-nitropyridine. A mixture of 2-chloro-3-112
cyano-5-nitropyridine and 1.15 mol of potassium thiocyanate was refluxed for 4 hours and113
dissolved in water. Light yellow crystal of 3-cyano-5-nitro-2-thiocyanopyridine 11026101114
was formatted and filtered off. Yield 67% (ethanol). MS m/z+.
206.1H NMR (dmso-d6) 9.68115
(1H, s, C(6)H), 8.66 (H, s, C(4)H) ppm. Anal. C7H2N4O2S, C,H,N. The same synthesis116
procedure was used to obtain 3-cyano-2-diethyldithiocarbamoyl-5-nitronopyridine 10026127117
using sodium diethyldithiocarbamate as nucleophylic agent in the last step. Yield 82%118
(ethanol). MS m/z+.
296.1H NMR (dmso-d6) 9.73 (1H, s, C(6)H), 8.75 (H, s, C(4)H), 4,22119
and 3.93 (4H, two q, N(CH2CH3)2), 1.32 (6H, t N(CH2CH3)2) ppm. Anal. C11H12N4O2S2,120
C,H,N.121
A series of 1-oxidopyridines 11026103, 11026115 and 11026114 was synthesized122
from the corresponding 5-R-2-chloropyridines, which were oxidized by complex123
thiourea/H2O2 and trifluoroacetic acid anhydride in methylene chloride. These124
chloroderivatives were treated with potassium thiocyanate for synthesis of compounds125
11026103 and 11026115, and sodium thioacetic acid for synthesis of compound 11026114.126
All reactions were done in ethanol as solvent. 1-Oxido-5-(trifluoromethyl)pyridin-2-yl127
thiocyanate (11026103). The yield was 53% (etanol), MS m/z+.
220.1H NMR (dmso-d6) 128
8.39 (1H, s, C(6)H), 7.79 (H, d, C(3)H), 7.44 (H, d, C(4)H) ppm. Anal. C 7H3F3N2OS, C,H,N.129
7/27/2019 screening on non culturable MTB.pdf
6/26
6
Ethyl 6-thiocyanatonicotinate 1-oxide (11026115). The yiels was 42 % (ethyl acetate). MS130
m/z+.
224.1H NMR (dmso-d6) 8.51 (1H, s, C(6)H), 8.02 (H, d, C(3)H), 7.88 (H, d, C(4)H),131
4.38 (2H, q, COOCH2CH3), 1.41 (3H, t, COOCH2CH3) ppm. Anal. C9H8N2O3S, C,H,N. Ethyl132
6-(acetylthio)nicotinate 1-oxide (11026114). The yiels was 36 % (ethanol). MS m/z+.
241.1H133
NMR (dmso-d6) 8.53 (1H, s, C(6)H), 8.07 (H, d, C(4)H), 7.52 (H, d, C(3)H), 4.38 (2H, q,134
COOCH2CH3), 2.35 (3H, s, SC(O)CH3), 1.41 (3H, t, COOCH2CH3) ppm. Anal. C10H11NO4S,135
C,H,N.136
Bacterium and media.137
M. tuberculosisstrain H37Rv was grown from frozen stocks for 1012 days in Sauton138
medium, containing: KH2PO4, 0.5 g; MgSO47H2O, 1.4 g; L-asparagine, 4 g; glycerol, 60 ml;139
ferric ammonium citrate, 0.05 g; sodium citrate, 2 g; 1% ZnSO47H2O, 0.1 ml; H2O, to l L; pH140
7.0 (adjusted with 1 M NaOH) (16) and supplemented with ADC (albumin, glucose and NaCl)141
and 0.05% of Tween-80 (37 C, 200 rpm).142
Dormant cells preparation.143
To obtain non-culturable (dormant) cells the culture grown in Sauton medium to144
OD600=5 served as an inoculum that was added to potassium-deficient Sauton medium145
supplemented with ADC and 0.05% of Tween-80 (37 C, 200 rpm) at a concentration 5x105146
cells per ml. Potassium-deficient Sauton medium contains: Na2HPO412H2O, 8.9 g;147
MgSO47H2O, 1.4 g; L-asparagine, 4 g; glycerol, 60 ml; ferric ammonium citrate, 0.05 g;148
sodium citrate, 2 g; 1% ZnSO47H2O, 0.1 ml; H2O, to l L; pH 7.0 (adjusted with 1 M NaOH)149
To obtain dormant M. tuberculosis cells in the Wayne hypoxia model and Betts150
starvation model the appropriate conditions were applied (13, 14). Briefly, for Wayne151
dormancy model M. tuberculosiswas grown in Dubos media at 37 C in sealed tubes. Cells152
ceased replicating after 7-8 days when oxygen concentrations decrease to the microaerobic153
level (1% oxygen saturation) and entered a non-replicating persistence state. With continued154
incubation the oxygen tension decrease to anaerobic level (0.06% oxygen saturation) and155
7/27/2019 screening on non culturable MTB.pdf
7/26
7
anaerobic culture was harvested at 14 days. For Betts starvation model cultures grown for 7156
days in nutrient-rich media were pelleted, washed twice with PBS and then resuspended in157
PBS and left standing at 37 C in sealed bottles for a 6 week period.158
Resuscitation of non-culturable cells.159
Resuscitation procedure was performed in liquid Sauton diluted medium (1:1, v/v, the160
final concentration of glycerol 0.6%) supplemented with ADC. Non-culturable cells161
obtained in potassium-deficient Sauton medium were harvested, washed twice with fresh162
media, resuspended in ADC-supplemented Sauton diluted medium and left standing at 37 C.163
The number of resuscitated cells was estimated by Most Probable Number (MPN) assay (17).164
Viability estimation.165
For CFU-counting ten-fold dilutions of M. tuberculosis suspensions were plated in166
triplicate on agar-solidified ADC-supplemented Sauton medium (limit of detection is 5 CFU167
per ml). Plates were incubated at 37 C for 21-25 days. For MPN assay (17) ten-fold bacterial168
dilutions were resuspended in ADC-supplemented liquid Sauton diluted medium in 48-well169
Corning microplates. Microplates were incubated at 37 C for 30 days without agitation.170
Wells with visible bacterial growth were counted as positive, and MPN values were calculated171
using standard statistical methods (17).172
Incorporation of radioactive uracil.173
Culture samples (1 ml) were placed in 2-ml screw cup tubes with 5,6-3H uracil (1 Ci)174
and incubated at 37 C with agitation for 20 h. Afterwards, 0.2 ml of the culture was placed in175
a 15-ml Falcon tube with 3 ml of 10% ice-cold Cl3and incubated in ice for 15 min.176
The mixture was then filtered through a Whatman glass microfiber filter (GFC) and washed177
with 3 ml 7% Cl3 and 6 ml 96% ethanol. Filters were then placed in 10 ml of178
scintillation mixture, and counts determined using a LS analyzer (Beckman Instruments, Inc).179
MICs determination.180
7/27/2019 screening on non culturable MTB.pdf
8/26
8
The determination of the MIC (Minimal Inhibitory Concentration) for M. tuberculosis181
H37Rv was determined according to the NCCLS guidelines(18) using a broth microdilution182
method in Middlebrook 7H9 media supplemented with ADC with a final inoculum of 5x105183
cell/mL. The compounds were dissolved in DMSO (1 mg/ml) and used as a stock solution.184
Concentrations ranging from 31.25 to 4000 ng/ml (at these concentrations the compounds185
were soluble in the medium) were used to assess the effectiveness of compounds. Microtitre186
plates were incubated at 37oC for 72 h, the MIC value represents the lowest dilution of the187
compound in which no bacterial growth was detected.188
Estimation of bactericidal effect.189
Both log-phase and dormant cells were exposed to different concentrations of RIF,190
INH and 2-thiopyridines for 7 days (37 C, 200 rpm). Both treated and untreated ten-fold191
diluted suspensions were employed in triplicate for MPN assays in ADC-supplemented liquid192
Sauton diluted medium in 48-well Corning microplates at 37 C for 30 days without agitation.193
MPN values were calculated using standard statistical methods (17).194
195
Results and discussion196
Themodel of M. tuberculosis non-culturability197
As previously demonstrated, potassium-deficiency led to the development of non-198
culturability in Mycobacterium smegmatis cells (19). The application of a similar approach199
induced non-culturability inM. tuberculosisbacilli: the cultivation of cells under potassium-200
deficient conditions led to their transition to a reversible NC state after a 30-day incubation201
period (Fig. 1). More than 99% of M. tuberculosis cells were unable to form colonies on202
standard solid medium after a 30-day incubation period in potassium-limiting conditions but203
the NC cells remained viable and could be resuscitated. Resuscitation of cells starved for 15-204
30 days by incubation in liquid Sauton diluted medium demonstrated the presence of 108
205
potentially viable cells per ml in the starved culture according to MPN assay. However, about206
7/27/2019 screening on non culturable MTB.pdf
9/26
9
106cells/ml (or less than 1% of the population) in such a 30-day-starved population remained207
fully culturable.208
The main problem of all in vitro dormancy models is the presence of cells with209
different physiological status in the population (11). Therefore, elimination of210
replicating/active cells is a crucial point for establishing a model with a high proportion of211
truly dormant cells. In particular, penicillin treatment was used for enrichment ofMicrococcus212
luteus dormant cells(20) and Hu and colleagues have suggested applying RIF to kill viable213
cells in the population of long-stationary-phase M. tuberculosis cells (21). As a result, they214
obtained a population of persistent M. tuberculosis cells, however, their concentration was215
very low (ca 102cells /ml).216
In order to obtain a population enriched with dormant NC cells and to remove217
metabolically active cells a 15-day-starvedM. tuberculosisculture was additionally incubated218
in the presence of a moderate concentration of RIF (5 g/ml), which should not affect viability219
of dormant cells (7). After such additional treatment of a 15-day-starved culture by RIF for220
10-12 days, a zero-CFU population of NC cells was obtained. These cells were unable to221
produce colonies on solid medium but were still characterized by a high recovery potential222
estimated by MPN assay when cultivated in Sauton diluted medium, 1x108 cells/ml (close to223
MPN for starved cells without rifampicin) were able to resuscitate and transit into a fully224
culturable state (Fig. 1). Such a high recovery potential of the zero-CFU population was225
maintained for at least the following 10-14 days afterM. tuberculosiscells entered in a zero-226
CFU state.227
Despite the ability to recover from the NC state, the metabolic activity of cells in the228
zero-CFU population was very low according to the rate of radioactive uracil incorporation229
correlating with transcriptional activity of cells. While viable log-phase M. tuberculosiscells230
demonstrated a rate of radioactivity incorporation at approx. 40 000 cpm estimated for 1x107
231
cells, this parameter was at least 100-fold less for the same amount of NC cells. This was in232
7/27/2019 screening on non culturable MTB.pdf
10/26
10
contrast to the metabolic activity of persisters obtained after exposure to a high concentration233
of RIF (100 g/ml) which was just 4 times less that parameter in log-phase cells (21), which234
means that Hus persisters did not demonstrate a dramatic metabolic reduction. This235
discrepancy could be explained by the different physiological state of cells in these two236
models.237
Apart from reduced metabolic activity, NC cells were morphologically distinct from238
actively growing cells (they were sphere-shape) and characterized by considerable resistance239
to anti-TB drugs. Namely, the treatment of NC cells in the zero-CFU population by two240
first-line anti-TB drugs RIF and INH (both at 5 g/ml) resulted in no effect on their ability to241
recover from the NC state in diluted liquid Sauton medium (Fig. 2). Poor killing effect was242
demonstrated even after treating the cells with ten-fold higher doses (Fig. 2). The sensitivity243
of cells obtained after the resuscitation to RIF and INH and the suppression of resuscitation244
from the NC state in the presence of 5 g/ml RIF and INH (not shown) showed that the245
antibiotic resistance that developed in NC cells is phenotypic.246
We compared the sensitivity to antibiotics of cells obtained in different in vitro247
dormancy models with the cells produced in the model described in this study (Table 1).248
Given that the population of dormant cells possibly contains a fraction of NC cells and that it249
is impossible to reveal susceptibility of NC cells to drugs by plating, we applied cell counting250
in a liquid medium (MPN assay) along with a standard CFU counting to estimate the killing251
effect of antibiotics. Cells in all dormancy models were expected to be characterized by their252
considerable resistance to INH (7, 13-15) because the latter impacts the biosynthesis of cell253
wall mycolic acids, a process which is evidently inactive in dormant and even in aged cells.254
However, dormant cells in all previously published M. tuberculosis dormancy models255
demonstrated significant susceptibility to RIF, which, in turn, may indicate some256
transcriptional activity in such cells. Resistant dormant NC cells in this study are evidently257
characterized by more deep dormancy (11) in comparison to previously published models.258
7/27/2019 screening on non culturable MTB.pdf
11/26
11
Similarly, deep persistence was reported to be characterised by very low metabolic activity259
and ability to survive under longer antibiotic exposure and their high concentrations (22)260
261
2-thiopyridine derivatives are effective against both M. tuberculosis active and262
dormant cells.263
Earlier, some 2-thiopyridine derivatives were found to have antileprosy activity (23).264
We therefore proposed that this group of chemical compounds may also possess anti-TB265
activity. In the present study we removed the potentially toxic nitro group and introduced a266
thiocyanato group as a result of different chemical modifications of the basic structures of 2-267
thiopyridine together with SAR studies. During SAR research we also suggested that one of268
the key factors of the antimicrobial activity of this group in the compounds is the level of269
positive charge on the second carbon atom of the pyridine ring. To confirm this hypothesis we270
synthesized several compounds where the pyridine nitrogen atom was oxidized. Then, the271
anti-TB activity of five 2-thiopyridine derivatives was studied (Fig. 3). Two compounds from272
this series, namely #11026103 and #11026115 demonstrated MICs of 0.25 g/ml (Table 2).273
The effectiveness of these five compounds was checked in the model of non-274
culturability of M. tuberculosiscells. As these NC cells are unable to produce colonies on275
non-selective agar-solidified media, we applied the MPN assay to check the effect of these276
compounds on the viability of dormant NC cells. In contrast to CFU counting this approach is277
able to uncover a sub-population of cells with decreased culturability which may arise as a278
result of antibiotic treatment.279
It was found that incubation of NC M. tuberculosis cells with some of the above280
mentioned compounds did affect their viability. In particular, incubation of NC cells with 10281
g/ml of both #11026114 and #11026115 for 7 days led to a more than 2-log killing effect282
(Fig. 4). At the same time NC cells are highly resistant to RIF and INH even at a high283
concentration (50 g/ml) (Fig. 2). The susceptibility of NC cells to the 2-thiopyridine284
7/27/2019 screening on non culturable MTB.pdf
12/26
12
derivatives may indicate the existence of some residual metabolic activity in the latent state of285
the pathogen, which may be potentially targeted by novel antimicrobials. Whilst the target(s)286
of studied compounds in M. tuberculosisis not clear, some thiopyridines were identified as287
inhibitors of M. tuberculosis alanine racemase which plays an essential role in cell wall288
synthesis (24), or enoyl-acyl carrier protein reductase which catalyzes an essential step in fatty289
acid biosynthesis (25). However, MIC reported forM. tuberculosisfor the first inhibitor is 50290
times higher then for the compounds described in present study, and the second inhibitor was291
not tested on mycobacteria. Thus, the mechanism of action of the compounds discovered in292
the present work remains to be elucidated.293
For further studying the effectiveness of 2-thiopyridine derivatives on dormant M.294
tuberculosiscells we selected the compound #11026115 and performed two other well-known295
in vitrodormancy models: the Wayne hypoxia model (13) and the Betts starvation model (14).296
Dormant cells obtained in different models were treated with 10 g/ml of #11026115 for 7297
days. As the population of dormant cells in these models may contain a fraction of NC cells298
(arising before or after treatment) we applied the MPN assay for estimation of the killing299
effect of #11026115 instead of a standard CFU counting. According to the MPN assay,300
dormant cells in these models are characterized by slightly higher resistance to both RIF and301
INH than was reported previously (7, 13-15).Probably, antibiotic treatment may induce the302
partial transition of cells to a NC state instead of cell death, which was not noticed earlier303
because of applying standard CFU counting. We also observed a significant killing effect of304
compound #11026115 on these dormantM. tuberculosiscells obtained in the Wayne hypoxia305
model and the Betts starvation modelin vitro (Fig. 5). Despite the fact that dormant cells in306
both models are rather susceptible to RIF, treatment with compound #11026115 demonstrated307
an even stronger effect in comparison to RIF in all three of M. tuberculosisin vitrodormancy308
models and their effectiveness on actively growing cells was comparable to that of both RIF309
7/27/2019 screening on non culturable MTB.pdf
13/26
13
and INH. So, we may conclude that several 2-thiopyridinas are able to kill dormant M.310
tuberculosisbacteria.311
Despite the recent success with drug candidates such as diarylquinolines (TMC207),312
which target ATP synthesis (26, 27), and benzothiazinones (BTZ043), which target essential313
cell-wall arabinan synthesis (28), the efficacy of both candidates at killing dormant bacilli was314
found to be low. For example, BTZ043 failed to kill cells with low metabolic status (15, 28,315
29)and the specific TMC207 affected the viability of dormant cells in the Wayne model but316
not the viability of nutrient-starved organisms (30).317
By a combination of two approaches - potassium deficiency and RIF treatment - we318
obtained a population containing dormant NC cells in a high concentration. The in vitromodel319
proposed in this study produces dormant NC cells which are characterized by remarkable320
metabolic cessation and considerable phenotypic resistance to both RIF and INH. Therefore,321
this model meets the key criteria of latent tuberculosis and may be applied as a relevant tool322
for latent TB drug discovery. Here, we demonstrated for the first time that 2-thiopyridines are323
able to kill dormantM.tbcells obtained under three different types of stress: hypoxia, nutrient324
starvation and non-culturability under potassium limitation. The most active compound was325
#11026115, which kills dormantM. tuberculosisbacteria obtained in all three in vitromodels326
of dormancy. We suggest 2-thiopyridinederivatives as a prominent group of compounds for327
further development of drugs for curing latent TB.328
Funding329
This work was supported by the Program of the Presidium of the Russian Academy of330
Sciences Molecular and Cellular Biology, FP7 project More medicines for tuberculosis331
(Grant 260872) and Russian Foundation of Basic Research grant number 12-04-01760-.332
Acknowledgements333
We are grateful to Prof. Stewart Cole (Global Health Institute, EPFL) for his critical334
reading of the manuscript and valuable comments.335
7/27/2019 screening on non culturable MTB.pdf
14/26
14
References336
1.Koul A, Arnoult E, Lounis N, Guillemont J, Andries K.2011. The challenge of337
new drug discovery for tuberculosis. Nature. 469:483-90.338
2.Zhang Y.2004. Persistent and dormant tubercle bacilli and latent tuberculosis. Front339
Biosci. 9:1136-56.340
3.Lobue P, Menzies D.2010. Treatment of latent tuberculosis infection: An update.341
Respirology 15:603-22.342
4.Menzies D, Al Jahdali H, Al Otaibi B.2011. Recent developments in treatment of343
latent tuberculosis infection. Indian J. Med. Res. 133:257-66.344
5.Zhang Y, Yew WW, Barer MR.2012. Targeting persisters for tuberculosis control.345
Antimicrob Agents Chemother. 56:2223-30346
6.Dhillon J, Dickinson JM, Sole K, Mitchison DA.1996. Preventive chemotherapy of347
tuberculosis in Cornell model mice with combinations of rifampin, isoniazid, and348
pyrazinamide. Antimicrob. Agents Chemother. 40:552-5.349
7. Deb C, Lee CM, Dubey VS, Daniel J, Abomoelak B, Sirakova TD, Pawar S,350
Rogers L, Kolattukudy PE. 2009. A novel in vitro multiple-stress dormancy model for351
Mycobacterium tuberculosisgenerates a lipid-loaded, drug-tolerant, dormant pathogen. PLoS352
One. 4:e6077.353
8.Chao MC, Rubin EJ.2010. Letting sleeping dos lie: does dormancy play a role in354
tuberculosis? Annu Rev. Microbiol.64:293-311.355
9.Dhillon J, Lowrie DB, Mitchison DA. 2004. Mycobacterium tuberculosis from356
chronic murine infections that grows in liquid but not on solid medium. BMC Infect. Dis.357
4:51.358
10.Khomenko AG, Golyshevskaya VI. 1984. Filtrable forms of Mycobacteria359tuberculosis. Z. Erkr. Atmungsorgane. 162:147-54.360
7/27/2019 screening on non culturable MTB.pdf
15/26
15
11.Young M, Mukamolova GV, Kaprelyants AS.2005. Mycobacterial dormancy and361its relation to persistence. In: Parish T (ed). Mycobacterium: Molecular Microbiology.362
London: Horizon bioscience. 265320.363
12.Payne DJ, Gwynn MN, Holmes DJ Pompliano DL. 2007. Drugs for bad bugs:364confronting the challenges of antibacterial discovery. Nature Rev. Drug Discov. 6:2940.365
13.Wayne LG, Hayes LG.1996. An in vitromodel for sequential study of shiftdown366of Mycobacterium tuberculosis through two stages of nonreplicating persistence. Infect.367
Immun. 64:2062-2069.368
14.Betts JC, Lukey PT, Robb LC, McAdam RA, Duncan K.2002. Evaluation of a369nutrient starvation model of Mycobacterium tuberculosis persistence by gene and protein370
expression profiling. Mol. Microbiol. 43:717-731.371
15.Sala C, Dhar N, Hartkoorn RC, Zhang M, Ha YH, Schneider P, Cole ST. 2010.372Simple model for testing drugs against nonreplicating Mycobacterium tuberculosis.373
Antimicrob. Agents Chemother. 54:4150-4158.374
16.Connell ND. 1994. Mycobacterium: isolation, maintenance, transformation, and375mutant selection. Methods Cell Biol. 45:107-25.376
17.de Man JC.1975. The probability of most probable numbers. J. Appl. Microbiol. 3771:67-78.378
18.National Committee for Clinical Laboratory Standards. 2007. Methods for379Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically. Fourth380
Edition: Approved Standard M7-A4. NCCLS, Villanova, PA.381
19.Shleeva MO, Mukamolova GV, Young M, Williams HD, Kaprelyants AS.2004.382Formation of 'non-culturable' cells of Mycobacterium smegmatis in stationary phase in383
response to growth under suboptimal conditions and their Rpf-mediated resuscitation.384
Microbiology. 150:1687-1697.385
7/27/2019 screening on non culturable MTB.pdf
16/26
16
20.Kaprelyants AS, Kell DB. 1993. Dormancy in stationary-phase cultures of386Micrococcus luteus: flow cytometric analysis of starvation and resuscitation. Appl.387
Environment. Microbiol.59:3187-3196.388
21.Hu Y, Mangan JA, Dhillon J, Sole KM, Mitchison DA, Butcher PD, Coates AR.3892000. Detection of mRNA transcripts and active transcription in persistent Mycobacterium390
tuberculosisinduced by exposure to rifampin or pyrazinamide. J. Bacteriol.182:6358-6365.391
22.Ma C, Sim S, Shi W, Du L, Xing D, Zhang Y. 2010. Energy production genes392sucB and ubiF are involved in persister survival and tolerance to multiple antibiotics and393
stresses inEscherichia coli. FEMS Microbiol Lett. 303:33-40394
23.Makarov V, Riabova OB, Yuschenko A, Urlyapova N, Daudova A, Zipfel PF,395Mllmann U. 2006. Synthesis and antileprosy activity of some dialkyldithiocarbamates. J.396
Antimicrob. Chemother. 57:11341138.397
24.Anthony KG, Strych U, Yeung KR, Shoen CS, Perez O, Krause KL, Cynamon398MH, Aristoff PA, Koski RA.2011. New classes of alanine racemase inhibitors identified by399
high-throughput screening show antimicrobial activity against Mycobacterium tuberculosis.400
PLoS One. 6:e20374401
25.Ling LL, Xian J, Ali S, Geng B, Fan J, Mills DM, Arvanites AC, Orgueira H,402Ashwell MA, Carmel G, Xiang Y, Moir DT. 2004. Identification and characterization of403
inhibitors of bacterial enoyl-acyl carrier protein reductase. Antimicrob Agents Chemother.404
48:1541-7.405
26.Andries K, Verhasselt P, Guillemont J, Ghlmann HW, Neefs JM, Winkler H,406Van Gestel J, Timmerman P, Zhu M, Lee E, Williams P, de Chaffoy D, Huitric E,407
Hoffner S, Cambau E, Truffot-Pernot C, Lounis N, Jarlier V. 2005. A diarylquinoline408
drug active on the ATP synthase ofMycobacterium tuberculosis. Science. 307:2237.409
7/27/2019 screening on non culturable MTB.pdf
17/26
17
27.Koul A, Dendouga N, Vergauwen K, Molenberghs B, Vranckx L, Willebrords R,410Ristic Z, Lill H, Dorange I, Guillemont J, Bald D, Andries K. 2007. Diarylquinolines target411
subunit c of mycobacterial ATP synthase. Nature Chem. Biol. 3:323324.412
28.Makarov V, Manina G, Mikusova K, Mllmann U, Ryabova O, Saint-Joanis B,413Dhar N, Pasca MR, Buroni S, Lucarelli AP, Milano A, De Rossi E, Belanova M,414
Bobovska A, Dianiskova P, Kordulakova J, Sala C, Fullam E, Schneider P, McKinney415
JD, Brodin P, Christophe T, Waddell S, Butcher P, Albrethsen J, Rosenkrands I, Brosch416
R, Nandi V, Bharath S, Gaonkar S, Shandil RK, Balasubramanian V, Balganesh T,417
Tyagi S, Grosset J, Riccardi G, Cole ST. 2009. Benzothiazinones kill Mycobacterium418
tuberculosisby blocking arabinan synthesis. Science. 324:801814.419
29.Zhang M, Sala C, Hartkoorn RC, Dhar N, Mendoza-Losana A, Cole ST. 2012.420Streptomycin-starved Mycobacterium tuberculosis 18b, a drug discovery tool for latent421
tuberculosis. Antimicrob. Agents Chemother. 56:5782-5789.422
30.Gengenbacher M, Rao SP, Pethe K, Dick T. 2010. Nutrient-starved, non-423replicating Mycobacterium tuberculosis requires respiration, ATP synthase and isocitrate424
lyase for maintenance of ATP homeostasis and viability. Microbiology. 156:81-87.425
426
427
428
429
430
431
432
433
434
435
7/27/2019 screening on non culturable MTB.pdf
18/26
18
Figure legends.436
Figure 1. Formation of NC M. tuberculosis cells under potassium-limiting437
conditions. Filled squares - gradual CFU decreasing in prolonged stationary phase; filled438
circles formation of a zero-CFU population after an additional treatment a 15-day-starved439
culture with a moderate concentration of RIF (5 g/ml). A potential viability of NC cells was440
estimated after resuscitation in liquid Sauton diluted medium by MPN assay. The arrow shows441
MPN value for 25-days starved with the additional treatment with RIF. Similar MPN values442
were also found for 15-30 days starved culture without RIF treatment. This experiment was443
repeated five times with similar results. Typical experiment is shown. Standard deviation for444
CFU did not exceed 10-20% for CFU mean and 20-30% for MPN mean.445
446
Figure 2.Resistance of NC M. tuberculosiscells to INH and RIF treatment. NC447
cells were washed and treated by different concentrations of antibiotics for 7 days. Viability of448
both treated and untreated NC cells was tested by the concentration of cells, which were able449
to recover from NC state (by MPN assay). The no-drug data represent an untreated control.450
Results are displayed as means and standard deviations of three independent experiments.451
452
Figure 3. Chemical structures of 11026101, 11026103, 11026114, 11026115 and453
10026127.454
455
Figure 4. Bactericidal effect of 2-thiopyridines on NC (dormant) M. tuberculosis456
cells.NC cells were treated by the compounds (10 g/ml) for 7 days at 37 C. Viability of both457
treated and untreated NC cells was tested by MPN assay. Bars represent 95% confidence458
limits.459
460
7/27/2019 screening on non culturable MTB.pdf
19/26
19
Figure 5. The bactericidal effect of RIF, INH and compound #11026115 for log-461
phase and dormantM. tuberculosiscells obtained in different in vitromodels.Dark bars 462
the model of non-culturability in potassium limiting conditions; grey bars Wayne hypoxia463
model; light-grey bars Betts starvation model; white bars actively growing cells. Cells464
were treated by the 10 g/mlantimicrobials for 7 days. Viability of both treated and untreated465
dormant cells was estimated by the MPN assay. Bars represent 95% confidence limits.466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
7/27/2019 screening on non culturable MTB.pdf
20/26
20
Table 1. Resistance ofM. tuberculosisdormant cells obtained in different models486
of TB dormancy in vitro to RIF and INH (percentage of resistant cells after a 7-day-487
treatment at 37 C). Ten-fold dilutions of M. tuberculosis suspensions were plated in488
triplicate onto agar-solidified Sauton medium (CFU counting) or in liquid Sauton ADC-489
supplemented medium containing Tween-80 0.05% (v/v) (MPN counting) in 48-well Corning490
microplates, wells with visible bacterial growth were scored as positive. Figures obtained in491
this study are in bold and calculated as means of three independent experiments.492
Models
Percentage of resistant
cells calculated by CFU
counting
Percentage of resistant
cells calculated by MPN
counting
RIF
(5-10
g/ml)
INH
(0.5-1
g/ml)
RIF
(5 g/ml)
INH
(1 g/ml)
Wayne model (Wayne and Heys,
1996)
< 0.10 96 0.8 100
Nutrient starvation (Betts et al.,
2002)
< 0.01 701.0 100
Multiple stress (Deb et al., 2009)
12.5 84.4 ND ND
Non-replicating state (Sala et al.,
2010)
< 0.01 30 ND ND
Non-culturability in
potassium-limiting conditions
(zero-CFU population)
* * 100 100
* CFU counting could not be performed due to non-culturability of cells493
494
7/27/2019 screening on non culturable MTB.pdf
21/26
21
Table 2. Minimal Inhibitory Concentration of 2-thiopyridines forM. tuberculosis495
H37Rv cells. The MIC was determined according to the NCCLS guidelines using a broth496
microdilution method in Middlebrook 7H9 media supplemented with ADC with a final497
inoculum of 5x105cell/ml. The results of five independent experiments are shown.498
499
Compound MIC, g/ml
11026101 4.0
11026103 0.250
11026114 0.750
11026115 0.250
10026127 4.0
500
501
502
503
504
505
506
507
508
509
510
511
7/27/2019 screening on non culturable MTB.pdf
22/26
7/27/2019 screening on non culturable MTB.pdf
23/26
7/27/2019 screening on non culturable MTB.pdf
24/26
7/27/2019 screening on non culturable MTB.pdf
25/26
7/27/2019 screening on non culturable MTB.pdf
26/26