Discovery of novel oral protein synthesis inhibitors of mycobacterium tuberculosis that target leucyl-tRNA synthetase

Article abstract of DOI:10.1128/AAC.01339-16

The recent development and spread of extensively drug-resistant and totally drug-resistant resistant (TDR) strains of Mycobacterium tuberculosis highlight the need for new antitubercular drugs. Protein synthesis inhibitors have played an important role in

Full text of DOI:10.1128/AAC.01339-16

AAC Accepted Manuscript Posted Online 8 August 2016
Antimicrob. Agents Chemother. doi:10.1128/AAC.01339-16
Copyright © 2016 Palencia et al.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license.
1
2
Discovery of novel oral protein synthesis inhibitors of
Mycobacterium tuberculosis that target leucyl-tRNA synthetase
3
4
*Andrés Palencia1†, *Xianfeng Li2, Wei Bu2, Wai Choi2, Charles Z. Ding2, Eric E. Easom2,
Lisa Feng2, Vincent Hernandez2, Paul Houston2, Liang Liu2, Maliwan Meewan2, Manisha
Mohan2, Fernando L. Rock2, Holly Sexton2, Suoming Zhang2, Yasheen Zhou2, Baojie Wan3,
Yuehong Wang3, Scott G. Franzblau3, Lisa Woolhiser4, Veronica Gruppo4, Anne J. Lenaerts4,
Theresa O’Malley5, Tanya Parish5, Christopher B. Cooper6, M. Gerard Waters6, Zhenkun
Ma6, Thomas R. Ioerger7, James C. Sacchettini7, Joaquín Rullas8, Iñigo Angulo-Barturen8,
Esther Pérez-Herrán8, Alfonso Mendoza8, David Barros8, Stephen Cusack1, Jacob J. Plattner2
and M.R.K. Alley2#
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
1European Molecular Biology Laboratory (EMBL), Grenoble, France
2Anacor Pharmaceuticals, Palo Alto, CA, USA
3Institute for Tuberculosis Research, University of Illinois at Chicago, Chicago, IL, USA
4Mycobacteria Research Laboratories, Department of Microbiology, Immunology and
Pathology, Colorado State University, Fort Collins, CO, USA
5TB Discovery Research, Infectious Disease Research Institute, Seattle, WA, USA
6Global Alliance for TB Drug Development, New York, USA
7Department of Biochemistry & Biophysics, Texas A&M University, College Station, TX,
USA
8Tres Cantos Medicines Development Campus (TCMDC), GlaxoSmithKline (GSK), Tres
Cantos, Spain
23
24
25
26
27
28
29
30
31
32
33
*Co-first authors
# Correspondence should be addressed to M.R.K. Alley (dickon_alley@mac.com)
†Present address: Andrés Palencia, Institut for Advanced Biosciences, Team Host-pathogen
interactions & immunity to infection, INSERM U1209, CNRS UMR5309, Université Grenoble Alpes,
Grenoble, France
RUNNING TITLE: Antitubercular leucyl-tRNA synthetase inhibitors
1

34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
ABSTRACT
The recent development and spread of extensively (XDR) and totally resistant (TDR)
strains of Mycobacterium tuberculosis, highlights the need for new antitubercular drugs.
Protein synthesis inhibitors have played an important role in the treatment of tuberculosis
(TB) starting with the inclusion of streptomycin in the first combination therapies. Although
the parenteral aminoglycosides are a key component in multidrug-resistant (MDR) TB
therapy, the oxazolidinone, linezolid, is the only orally available protein synthesis inhibitor
that is effective against TB. Herein, we show that small molecule inhibitors of aminoacyl-
tRNA synthetases (AARS), known to be excellent antibacterial protein synthesis targets, can
be designed that are orally bioavailable and effective against M. tuberculosis in TB mouse
infection models. We applied the oxaborole tRNA trapping (OBORT) mechanism, which was
first developed to target fungal cytoplasmic leucyl-tRNA synthetase (LeuRS), to M.
tuberculosis LeuRS. X-ray crystallography was used to guide design of LeuRS inhibitors
that have good biochemical potency and excellent whole cell activity against M. tuberculosis.
Importantly, their good oral bioavailability translates into in vivo efficacy in both the acute
and chronic mouse models of TB with comparable potency to the frontline drug isoniazid.
2

51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
INTRODUCTION
The aminoacyl-tRNA synthetases (AARS) are a family of essential enzymes that are
required for protein synthesis in all cells (38). Although various family members have been
targeted for the design of novel antibacterial (37) only the isoleucyl-tRNA synthetase
inhibitor, mupirocin, is an FDA-approved antibiotic (36). However, mupirocin is only
approved for the topical treatment of staphylococcal and streptococcal skin infections (36)
and M. tuberculosis is naturally resistant to this agent (31). Leucyl-tRNA synthetase
(LeuRS) is a class I AARS that has two active sites separated by a distance of 30 Å; a
synthetic site that aminoacylates tRNALeu and an editing site that ensures the fidelity of
translation by a proofreading mechanism (7, 10, 19, 26). Recently boron-containing
compounds known as oxaboroles have been shown to inhibit LeuRS by the oxaborole tRNA
trapping (OBORT) mechanism (29), which exploits the ability of the boron atom to bond to
the cis-diols of the 3´-terminal adenosine nucleotide, Ade76, of tRNALeu. The resulting
covalent adduct traps the 3´-end of tRNALeu in the editing site in a non-productive complex,
inhibiting leucylation and thereby protein synthesis (29). Here we report the discovery of
novel 3-aminomethyl derivatives that have potent antitubercular activity.
3

68
MATERIALS AND METHODS
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Chemical Synthesis. Starting materials used were either available from commercial sources
or prepared according to literature procedures and had experimental data in accordance with
those reported. The syntheses of the compounds are described in detail in the Supplementary
Material.
Expression, purification and crystallization of M. tuberculosis LeuRS editing domain. A
DNA fragment coding for the region G309 to I513 of M. tuberculosis LeuRS (Uniprot
P67510) was cloned into pETM-11 using the NcoI and XdeI restriction sites (EMBL). The
protein containing an N-terminal six-histidine tag was prepared and purified following a
similar protocol as for the E. coli LeuRS (26), except that the nickel affinity chromatography
was conducted at pH 8.0. Protein was stored in buffer comprising 20 mM Tris-HCl (pH 7.4),
100 mM NaCl, 5 mM MgCl2 and 5 mM 2-mercaptoethanol. Crystallization was performed at
20°C by the hanging drop vapor diffusion method. The solutions for the ternary complexes
were prepared with 10 mg/mL LeuRS, 5 mM AMP and 1 mM of the corresponding
benzoxaborole compound (provided by Anacor Pharmaceuticals, Palo Alto, CA). Initial
crystals were obtained at 15 mg/mL LeuRS, 5 mM AMP and 1 mM of the corresponding
benzoxaborole compound (provided by Anacor Pharmaceuticals, Palo Alto, CA). Crystals
were obtained by mixing 2 μL of this solution with 2 μL of reservoir solution containing 0.1
M Bis-TRIS (pH 5.5), 22% (w/v) PEG 10000 and 0.2 M ammonium acetate. Quality and size
of final diffracting crystal was improved by decreasing LeuRS concentration to 10 mg/mL
and PEG 10000 to 17% (w/v). The crystals were frozen directly in liquid nitrogen in the
mother liquor containing 15% (v/v) ethylene glycol as a cryoprotectant.
Structure determination and refinement. All diffraction data sets were collected at the
European Synchrotron Radiation Facility (ESRF, Grenoble, France). Data were integrated
and scaled with the XDS suite (15). Further data analysis was performed with the CCP4 suite
(3). The structure of the LeuRS:AMP-compound 6 complex was initially solved by molecular
replacement with PHASER (22) using the E. coli LeuRS editing domain structure(20) (PDB
2AJG) as a model. The model was improved by automatic building using ARP-wARP (28)
and manual adjustments were made with COOT (9). The structures of the complexes with
the compounds 14 and 16 were solved using the editing domain of M. tuberculosis LeuRS
(described above) as a model. All models were refined using REFMAC5 with anisotropic B-
factors.
Structure
quality
was
analyzed
with
MOLPROBITY(4)
(http://molprobity.biochem.duke.edu/) and showed all residues in allowed regions (with 95.1-
4

101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
98.0% of residues in favored regions) for the different models. Figures were drawn with
PYMOL (http://www.pymol.org/).
Aminoacylation assay. An N-terminal six histidine-tagged LeuRS from M. tuberculosis
H37Rv, which was codon-optimized for E. coli (GenScript, Piscataway NJ, USA), was over-
expressed and purified according to Novagen (Madison, WI, USA) using an E. coli
BL21(DE3) T7 RNA polymerase over-expression strain. Experiments were performed in 96-
well microtiter plates, using 80 μL reaction mixtures containing 50 mM HEPES-KOH (pH
8.0), 30 mM MgCl2 30 mM KCl, 13 μM L-[14C]leucine (306 mCi/mmol, Perkin-Elmer), 15
µM total E. coli tRNA (Roche, Switzerland), 0.02% (w/v) BSA, 1 mM DTT, 0.2 pM LeuRS
and 4 mM ATP at 300C. Reactions were started by the addition of 4 mM ATP. After 7
minutes, reactions were quenched and tRNA was precipitated by the addition of 50 μL of
10% (w/v) TCA and transferred to 96-well nitrocellulose membrane filter plates (Millipore
Multiscreen HTS, MSHAN4B50). Each well was then washed three times with 100 μL of
5% TCA. Filter plates were then dried under a heat lamp and the precipitated L-[14C]leucine
tRNALeu were quantified by liquid scintillation counting using a Wallac MicroBeta Trilux
model 1450 liquid scintillation counter (PerkinElmer, Waltham, MA, USA).
IC50 determination. To determine the inhibitor concentration, which reduces enzyme
activity by 50% (IC50), increasing concentrations of compound inhibitors that covered the
IC50 value were incubated with LeuRS enzyme, tRNA and L-leucine for 20 minutes.
Reactions were initiated by the addition of 4 mM ATP. Reactions were stopped after 7
minutes then precipitated and counted to quantify radioactivity. IC50 values were determined
using a 4-parameter logistic nonlinear regression model (Graphpad Software Inc. (La Jolla,
CA, USA).
Isothermal titration calorimetry (ITC) experiments. ITC experiments were performed at
250C using an ITC200 system (MicroCal, GE Healthcare). The editing domain protein was
dialyzed for 12 hours against the titration buffer (50 mM HEPES-KOH, 30 mM KCl and 30
mM MgCl2, pH 8.0) at 40C. Protein solutions at 50 μM plus AMP at 10 mM in the
calorimetric cell were titrated with the appropriate compound dissolved in dialysis buffer.
Compound solutions at 1-5mM plus AMP at 10 mM were incubated at 37°C during 1 hour
before titrations. The heat evolved after each ligand injection was obtained from the integral
of the calorimetric signal. The resulting binding isotherms were analyzed by nonlinear least-
squares fitting of the experimental data to a single site model. Analysis of the data was
performed using Microcal Origin software (OriginLab version 7). Experiments were
5

134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
performed at least twice. The variability in the binding experiments was estimated to be 5%
in the binding enthalpy; and 10% in both the binding affinity and the number of sites.
Determination of minimum inhibitory concentration for M. tuberculosis. MIC
determinations were mainly determined using resazurin (Alamar Blue) as an indicator for cell
growth (6) with additional determinations as described by Ollinger et al (25).
Selection of M. smegmatis ATCC 700084 single-step mutants. Resistant mutants to
compound 1 were isolated on Middlebrook 7H10 medium plus 10% (v/v) oleic albumin
dextrose catalase (OADC) supplement (Becton Dickinson) containing compound 1 at 4xMIC.
Resistance was confirmed by measuring the mutants MIC value essentially as described by
Collins et al (6).
Selection of M. tuberculosis single-step mutants. M. tuberculosis mutants resistant to
compound 1 and 13 were isolated as described by Ioerger et al (14). Mutants were isolated on
Middlebrook 7H10 medium plus 10% (v/v) OADC (Becton Dickinson) containing compound
1 and 13 at 5X or 10X MIC99. Resistance was confirmed by measuring MIC99 on solid
medium – defined as minimum concentration that inhibits 99% of CFU (34). Genome
sequencing and identification of polymorphisms were essentially carried out as described by
by Ioerger et al (14). In order to determine the rate of spontaneous resistant mutants, M.
tuberculosis H37RV was grown at 37°C in fresh Middlebrook 7H9-ADC-Tween 80 to mid-
exponential phase and then diluted in fresh Middlebrook 7H9-ADC-Tween 80 to 5x108
CFU/mL. Middlebrook 7H10-OADC plates with 4 and 10-fold MIC of each compound were
inoculated with 108, 107, 106, and 105 CFU/plate, and the plates were incubated at 37°C for 3
to 4 weeks. The frequency of appearance of resistant mutants was calculated, and isolated
colonies were restreaked onto Middlebrook 7H10-OADC agar containing the drugs and on
plates without the drug.
Time-kill assay. Compounds were added at 20xMIC to a 10 mL exponential culture of M.
tuberculosis H37Rv (~5x105 cfu/mL) in Middlebrook 7H9 with 10% (v/v) OADC and 0.05%
(v/v) Tween-80. At specified time points, aliquots of cultures were withdrawn, serially
diluted and plated on solid culture medium. Plates were then incubated at 37oC and CFU
were counted after 3 to 4 weeks.
In vitro cytotoxicity assay. Vero epithelial cells (from African green monkey; ATCC CCL-
81) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10%
fetal bovine serum (FBS) and maintained in a humidified incubator (370C, 5% CO2). Cells
were dislodged with a cell scraper, collected by centrifugation, resuspended in fresh medium
at ~106 cells/mL, dispensed into 96-well microtitre plates (100mL/well) and incubated for
6

168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
18h at 370C. Two-fold serial dilutions of test compounds (800–0.4 mg/L) in DMEM with
FBS were subsequently added and cells incubated for another 72h. From triplicate studies,
the cytopathic effects of compounds were evaluated colorimetrically using the MTT cell
proliferation assay (ATCC). IC50 data were obtained from dose–response curves plotted using
Graphpad prism 5.
Mitochondrial protein synthesis assay. Human liver carcinoma derived HepG2 cell line
was obtained from the ATCC (HB-8065). HepG2 cells were grown in Dulbecco's Modified
Eagle's medium containing 10% Fetal Calf Serum, 1 mM sodium pyruvate, 0.1 mM non
essential amino acids and 50 units/mL penicillin-streptomycin at 370C with 5% CO2. HepG2
cells were seeded in 96-well plates at 3000 cells/200 μL/well in cell culture medium. Cells
were then grown in the presence of the compounds at 37 °C in 10% CO2 for 7 days in at least
duplicate concentrations, with the medium and compounds being replaced on the fourth day.
After 7 days the levels of SDHA and COX1 were determined by using the MitoSciencesTM
In-Cell ELISA kit. Janus Green staining was used to determined cell viability after 7 days
(In-cell ELISA kit, cat # MS643).
Mouse pharmacokinetic analysis. The studies were conducted using female CD-1 for
compounds 1, 11 and 12, while BALB/C mice were used for compound 13. Mice body
weights were 19-28 g and on the morning of dosing, mice were split randomly into 3 dosing
groups to receive test article by either tail-vein injection (IV) or oral gavage (PO). After
dosing, blood samples were collected via cardiac puncture at specific time points (n=3
mice/time point) through 24 hours (K2EDTA as anticoagulant) and processed for plasma.
Antibiotic concentrations in the plasma samples were analysed by LC/MS/MS. The
LC/MS/MS analysis was conducted using analyte/internal standard peak area methods. The
minternal standard was AN3365 (13) and the instrument was a API4000 QTRAP (AB Sciex).
The limit of quantitation (LOQ) was 1 or 2 ng/mL. Pharmacokinetic analyses of the mean
plasma concentration-time profiles were performed using WinNonlin Pro version 5.2. A
compartmental model was used for the IV data and non-compartmental model for PO data.
The time-concentration curve after an IV dose showed a bi-exponential decline with first-
order elimination. Compound 1 and 13 were formulated in saline (0.9% w/v NaCl) at 7.5
mg/mL and the pH adjusted to >5 by the addition of NaOH. Compound 11 was formulated
to 6.5 mg/mL in Water/Dimethylacetamide/EtOH (76/19/5) and the pH adjusted to >5 by the
addition of NaOH. Compound 12 was formulated to 7.5 mg/mL in PEG300/PG/water
(55/25/20) and the pH adjusted to >5 by the addition of NaOH.
7

201
202
203
204
205
206
207
208
209
210
211
212
Mouse plasma protein binding determination. Compounds were added to 1.5-mL aliquots
of mouse plasma and plasma ultrafiltrate to the following concentrations: 1 µg/mL and 10
µg/mL, and then incubated in a shaking water-bath at 37 °C for 15 minutes. Both samples
were treated similarly and a 0.5-mL aliquot was removed from each tube and added to the
filter reservoir of the Microcon® centrifugal filter devices (Ultracel YM-30, MWCO=30K
Da, Bedford, MA). The devices were centrifuged at 1000 x g for 10 minutes and 100 µL of
filtrate was transferred to the 96-well plate and diluted 5-fold. Ten-µL volumes of the
samples were injected and analyzed with the LC/MS/MS system. All samples were analysed
in duplicate. Quantitation was based on peak area ratio of analyte over internal standard and
all integrations were performed with peak areas using Analyst version 1.4.1 (Applied
Biosystems, Foster City, CA 94404, USA). Plasma protein binding was calculated, based on
the following equation:
Spiked
PlasmaUltrafiltrate
Filtrate
)
Plasma
(PeakArea
− PeakArea
213
Plasma Protein Binding (%) =
*100
Spiked
PeakArea
PlasmaUltrafiltrate
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
Murine model of acute TB infection using C57/BL GKO IFNγ mice. Eight- to 10-week-
old female specific-pathogen-free C57BL/6-Ifngtm1ts mice (IFNγ gene-disrupted [GKO]
mice) were purchased from Jackson Laboratories, Bar Harbor, ME. The mice were infected
via a low-dose aerosol exposure with M. tuberculosis Erdman in a Middlebrook aerosol
generation device (Glas-Col Inc., Terre Haute, IN) as described previously (18). One day
post-aerosol, three mice were sacrificed to verify the uptake of 50 to 100 CFU of bacteria per
mouse. Each treatment group consisted of five mice and treatment was started 10-13 days
post-infection and continued for nine or fourteen consecutive days. Five infected mice were
sacrificed at the start of treatment as pretreatment controls. Drugs were administered daily by
oral gavage. Lungs were harvested 24 hours after the last administration and all lung lobes
were aseptically removed, homogenized and frozen. Homogenates were plated on 10%
OADC-7H11 medium for 21 days at 37ºC. All animal studies strictly adhered to the protocols
and regulations approved by their respective Animal Care and Use Committees of University
of Illinois at Chicago and Colorado State University.
Murine model of acute and chronic TB infection using C57/BL6J mice. Specific
pathogen-free, 8-10 week-old female C57BL/6 mice were purchased from Harlan
Laboratories and were allowed to acclimate for one week. The experimental design for the
acute assay has been previously described (30). In brief, for the acute assay, mice were
intratracheally infected with 100,000 CFU/mouse (M. tuberculosis H37Rv strain). Products
8

233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
were administered for 8 consecutive days starting one day after infection. For the chronic
assay, mice were intratracheally infected with 100 CFU/mouse and the products administered
daily (7 days a week) for 8 consecutive weeks starting 6 weeks after infection. Lungs were
harvested 24 hours after the last administration. All lung lobes were aseptically removed,
homogenized and frozen. Homogenates were plated on 10% OADC-7H11 medium and
incubated for 21 days at 37ºC. The viable colony forming units were converted to logarithms,
which were then evaluated by a one-way analysis of variance, followed by a multiple-
comparison analysis of variance by a one-way Tukey test (SigmaStat software program).
Differences were considered significant at the 95% level of confidence. All animal studies
were ethically reviewed and carried out in accordance with European Directive 2010/63/EU
and the GSK Policy on the Care, Welfare and Treatment of Animals.
Murine model of chronic TB infection using BALB/c mice. Six- to 8-week-old female
specific-pathogen-free immunocompetent BALB/c mice (Charles River, Wilmington, MA)
were infected via a low-dose aerosol exposure to M. tuberculosis Erdman as described (18).
One day post-aerosol, three mice from each run were sacrificed to verify the uptake of 50 to
100 CFU of bacteria per mouse. Each group consisted of five to six mice at each time point.
Treatment was started 3 weeks post-infection and continued for 12 weeks. Five infected mice
were sacrifice at the start of treatment as pretreatment controls. Drugs were administered 5
days per week by oral gavage, for four weeks. Lungs were harvested 72 hours after the last
administration. All lung lobes were aseptically removed, homogenized and frozen.
Homogenates were plated on 10% OADC-7H11 medium and incubated for 21 days at 37ºC.
All animal studies strictly adhered to the protocols and regulations approved by Colorado
State University’s Animal Care and Use Committee.
9

257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
RESULTS AND DISCUSSION
Discovery of antitubercular LeuRS inhibitors. A focus library of 20 benzoxaboroles was
initially screened against M. tuberculosis H37Rv, which yielded AN3016 and AN3017 with
minimal inhibitory concentration (MIC) of 1 µg/mL and 1.8 µg/mL and LeuRS half-maximal
inhibitory concentration (IC50) of 3.5 µM and 0.64 µM, respectively (Fig. 1). When we
combined both the 3-aminomethyl and the 7-ethoxy substitutions in one moiety, compound 1,
it gave significantly better activity with a LeuRS IC50 of 0.28 µM and an MIC of 0.26 µg/mL
(Fig. 1). To confirm that compound 1 was targeting LeuRS in the cell, we obtained resistant
mutants of M. tuberculosis H37Rv and Mycobacterium smegmatis ATCC 700084. The leuS
gene, which codes for LeuRS, was sequenced from selected resistant mutants and mutations
were found in both organisms, which was consistent for an OBORT LeuRS inhibitor (Fig. 2
and Table 1). Therefore, we progressed compound 1 into an in vivo mouse pharmacokinetic
study to determine suitability for testing in an acute TB mouse model. Compound 1 was
dosed by oral administration (PO) at 30 mg/kg, which yielded an area under the curve over
24 hrs (AUC0-24) of 15 h*µg/mL with a maximum plasma level (Cmax) of 4.33 µg/mL and
oral bioavailability of 55% (data not shown). Since the existing in vivo efficacy mouse model
used M. tuberculosis Erdman and not M. tuberculosis H37Rv, the MIC of compound 1 was
confirmed against the Erdman strain as well as some drug-resistant isolates (Table 2). The M.
tuberculosis Erdman MIC value was 0.127 µg/mL, which was not affected by resistance
mechanisms to rifampin, isoniazid or streptomycin. Compound 1 was then tested in a
BALB/c acute model of TB dosed at 100 mg/kg twice-a-day (BID) for 3 weeks with weekend
drug holidays. The control drug, PA-824 (Pretomanid) (35), was dosed at 100 mg/kg once-a-
day (QD), which gave a 1.9 log10 reduction in colony forming units (CFU) from mouse lungs
compared to only a 0.4 log10 reduction in CFU for compound 1 (data not shown). Therefore,
we separated the two enantiomers in compound 1 by chiral HPLC and determined their
activities (Fig. 1). The active enantiomer was the (S)-isomer, compound 2, which had a
LeuRS IC50 of 0.13 µM and an MIC of 0.13 µg/mL, while the (R)-isomer was barely active
with an IC50 of 21 µM. The active enantiomer, compound 2, was then tested at 200 mg/kg
BID for 9 days in the acute TB model using an IFN-γ gene knock-out (GKO) mouse (17),
which showed a 2 log10 reduction in lung CFU and a 1.5 log10 reduction in spleen CFU
compared to the control mice (Fig. 3). However, this was not as good as the frontline drug
isoniazid (INH), which gave a 2.8 and 3.4 log10 reduction in CFU from lungs and spleen,
respectively, when dosed at 25 mg/kg.
10

290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
M. tuberculosis LeuRS inhibitor co-crystalization. In order to improve potency we
performed structural and biophysical studies to understand the binding mode of these novel
3-aminomethyl benzoxaboroles to M. tuberculosis LeuRS. Crystallization trials with different
editing domain constructs of M. tuberculosis LeuRS were attempted in the presence of
compound 2 with either AMP or longer nucleotides such as CytAde or CytCytAde, which
might act as surrogates for the 3´-end of the tRNA acceptor stem (32). An editing domain
construct encompassing residues G309-I513 gave co-crystals with compound 2 and AMP that
diffracted to 1.3 Å resolution, which permitted structure determination (Fig. 4A, 4B and
Table S1). Compound 2 forms a bidentate covalent adduct with AMP (Fig. 4B), which
mimics Ade76 of the tRNA acceptor end (13, 19, 26). The amino acid residues, T336-T337,
of the threonine-rich region provides multiple H-bonding interactions to the covalent adduct,
and L432 and Y435 of the AMP binding loop have extensive H-bonding and hydrophobic
contacts with AMP (Fig. 4B). In addition, the amino group of compound 2 makes three key
interactions with the carboxylic acid side chains of D447 and D450 and the carbonyl of
M441. The 7-ethoxy enables not only a new interaction to R449 but also packs with the
Ade76 ribose, thus further stabilizing the boron-tRNA adduct (Fig. 4B). Superposition of the
compound 2 adduct bound structure with that of the E. coli LeuRS editing domain with
methionine bound (20) shows that the 3-aminomethyl benzoxaborole moiety occupies the
same position as the non-cognate amino acid (Fig. 4C). Although this moiety mimics the
interactions established by the amino and the oxygen carbonyl groups of methionine, it lacks
atoms at the positions of the Sδ-Cε atoms of methionine (Fig. 4C), which suggests that there is
additional space to make further interactions.
SAR of potent antituberculars. Several derivatives were synthesized with different
substitutions at position 4 as well as at positions 5 and 6 to explore this hypothesis (Fig. 1).
The halogen substitutions 5-Cl (compound 5) and 6-F (compound 7) were not well tolerated
with LeuRS IC50 values worse than the original compound 1 and MIC values of 1.1 µg/mL or
greater. The most potent analogs were compounds with halogen substitutions at position 4,
bromo (compound 11), chloro (compound 4) and fluoro (compound 6), which improved MIC
values more than 5-fold over compound 1 (Fig. 1). The phenyl (compound 10) substitution
was not tolerated with LeuRS IC50 values of 28 µM (Fig. 1). However, it must be noted that
the significant improvements in MIC values for compounds 4, 6 and 11 were not fully
reflected in their IC50 values as determined using an aminoacylation assay with M.
tuberculosis LeuRS, which could be due to the way that OBORT inhibitors indirectly inhibit
11

323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
aminoacylation by preventing Ade76 binding to the aminoacylation active site (29). We
therefore decided to measure the direct binding of the compounds to the editing domain using
isothermal titration calorimetry (ITC) and found that the 4-Cl and 4-Br substitutions
significantly enhanced the affinity of the compounds to the M. tuberculosis LeuRS editing
domain (Table 3). The increased affinity is due to a significant gain in the enthalpic
contribution (3.1-4.5 Kcal mol-1), which is consistent with additional favourable interactions
being established by the halogen atoms in the editing site. To confirm whole cell activity was
derived from inhibition of LeuRS we selected 6 M. tuberculosis mutants resistant to
compound 13 and sequenced their leuS gene, while two additional mutants were selected for
whole genome sequencing. All 8 resistant mutants had SNP in their leuS genes and the
mutations were located in the editing domain as expected for OBORT LeuRS inhibitors (13,
24, 29) (Table S2). To further explore interactions at 4-position, we co-crystallised
compounds with 4-Cl and 4-Br substitutions in the presence of AMP and solved the
structures of the ternary complexes at 1.45 and 1.47 Å resolution, respectively (Table S1, Fig.
S1). The structures showed that the halogenated compounds bind to the editing site without
major structural changes and as predicted, the 4-Cl/Br atoms now occupy the position of the
sulphur in bound methionine (Fig. S2A) allowing van der Waals interactions with the
neighbouring protein atoms (Fig. S2B). These results confirmed the importance of the size
and nature of the substitution at position 4, and agreed well with the in vitro activities and
thermodynamic analysis (Fig. S2C).
We selected the three most potent compounds, 11, 12 and 13, for in vivo murine
pharmacokinetic analysis and we dosed mice both intravenously (IV) and orally (PO). All
three compounds showed improvements in plasma exposure as measured by AUC after oral
administration over compound 2 (Table 4). Therefore, we tested them in a GKO mouse
model of acute TB, which showed all to be very efficacious with the racemate compound 11
having similar efficacy to isoniazid (Fig. 5AB). In a chronic TB BALB/c mouse model, all
compounds showed good efficacy (Fig. 5C) with compound 14, the (S)-isomer of compound
11, being the most potent. In addition, we observed that compounds 13 and 14 did not show
any cross-resistance against multidrug-resistant isolates (Table S3).
Inhibition of mitochondrial protein synthesis. Although protein synthesis inhibitors are
validated TB drugs they are associated with some safety concerns, for example
myelosuppression and neuropathy observed with linezolid (16) and deafness induced by
aminoglycosides (12). The similarity between the bacterial and mitochondrial protein
synthesis machinery (2) and their subsequent inhibition of mitochondrial protein synthesis is
12

357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
thought to drive these toxicities. Therefore, we tested the ability of compound 14 and some
close analogues to inhibit mitochondrial protein synthesis in the human liver carcinoma cell
line HepG2 (Table 5). The ribosomal protein synthesis inhibitors linezolid, chloramphenicol
and doxycycline inhibited the synthesis of the mitochondrial derived COX1 protein with EC50
values of between 23 and 31 µM, while erythromycin and compounds 12, 13 and 14 had
EC50 values of >150 µM. Although this could be due to poor mitochondrial penetration, it is
interesting to note that human mitochondrial LeuRS is known to be editing defective as it
lacks key conserved amino acid residues in the AMP binding loop and amine binding pocket
(21).
Compound 14 in vitro and in vivo activity. Since the racemate 11 had similar activity to
isoniazid, which is an in vitro bactericidal compound (8), in the acute GKO mouse model
(Fig. 5B) we tested the enantiomer pure compound 14 for in vitro bactericidal activity over
14 days at 20-fold its MIC (Fig. 6). The profile for compound 14 was very similar to the
bacteriostatic protein synthesis inhibitor, linezolid, which was different from moxifloxacin a
known bactericidal compound (33). Therefore, further tests of compound 14 in a murine
chronic TB model (Fig. 7) were performed in parallel with the protein synthesis inhibitor
linezolid. Compound 14 at 30 mg/kg QD showed good efficacy resulting in a 2.4 log10
reduction in CFU compared with a 2.6 log10 reduction in CFU for 100 mg/kg QD of linezolid.
In order to establish the optimal dosing regimen for compound 14 we tested it in the acute TB
murine model and compared the efficacy from the following dosing regimes, BID, QD and
q48h (Fig. 8). Similar to results in the chronic model, compound 14 was more active at lower
doses than linezolid and dosing every other day (q48h) was as efficacious as QD or even
BID, which suggests that the preliminary pharmacodynamic driver for efficacy was
AUC/MIC.
Resistance and LeuRS inhibitors.
Emergence of resistance during streptomycin
monotherapy (1) and its noted reduction by the addition of p-aminosalicyclic acid (11, 23)
lead to the paradigm of combination TB drug therapy. The current core TB regimen calls for
a four-drug combination of isoniazid, rifampin, ethambutol and pyrazinamide. Although
compound 14 has a lower in vitro resistance frequency than isoniazid (Supplementary Table
7), the emergence of resistance to epetraborole (GSK2251052/AN3365), another 3-
aminomethylbenzoxaborole LeuRS inhibitor, in a minority of patients in a complicated
urinary tract infection trial might suggest some caution (24). However, the addition of
trimethoprim to rifampicin, which has a similar resistance problem, in an urinary tract
infection trial demonstrated the benefit of combination therapy in overcoming emergence of
13

391
392
393
394
395
396
397
398
399
400
rifampin-resistant strains (27). This suggests that the risk from emergence of resistance to
OBORT LeuRS inhibitors will likely be mitigated when used in combination therapy.
Beneficial properties. Since combination therapy necessitates a larger armamentarium than
regular monotherapy, the demonstration for the first time that an oral AARS inhibitor can be
a potent antitubercular adds a potential new tool to fight TB, which is timely noting the recent
onset of TDR-TB. In addition, the combination of low plasma protein binding, molecular
weight (207-285) and logD7.4 (-0.04-0.76), like the frontline TB drugs isoniazid,
pyrazinamide and ethambutol, suggests that this novel chemical class deserves further
optimisation and hopefully progression into clinical trials.
14

401
402
REFERENCES
1.
2.
3.
4.
1948. Streptomycin treatment of pulmonary tuberculosis. British medical journal
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
2:769-782.
Benne, R., and P. Sloof. 1987. Evolution of the mitochondrial protein synthetic
machinery. Bio Systems 21:51-68.
C.C.P.N. 1994. The CCP4 suite: programs for protein crystallography. Acta
Crystallogr D D50:760-763.
Chen, V. B., W. B. Arendall, 3rd, J. J. Headd, D. A. Keedy, R. M. Immormino,
G. J. Kapral, L. W. Murray, J. S. Richardson, and D. C. Richardson. 2010.
MolProbity: all-atom structure validation for macromolecular crystallography. Acta
Crystallogr D Biol Crystallogr 66:12-21.
5.
6.
Cheng, A. F., W. W. Yew, E. W. Chan, M. L. Chin, M. M. Hui, and R. C. Chan.
2004. Multiplex PCR amplimer conformation analysis for rapid detection of gyrA
mutations in fluoroquinolone-resistant Mycobacterium tuberculosis clinical isolates.
Antimicrobial agents and chemotherapy 48:596-601.
Collins, L., and S. G. Franzblau. 1997. Microplate alamar blue assay versus
BACTEC 460 system for high-throughput screening of compounds against
Mycobacterium tuberculosis and Mycobacterium avium. Antimicrobial agents and
chemotherapy 41:1004-1009.
7.
8.
Cusack, S., A. Yaremchuk, and M. Tukalo. 2000. The 2 A crystal structure of
leucyl-tRNA synthetase and its complex with a leucyl-adenylate analogue. The
EMBO journal 19:2351-2361.
de Steenwinkel, J. E., G. J. de Knegt, M. T. ten Kate, A. van Belkum, H. A.
Verbrugh, K. Kremer, D. van Soolingen, and I. A. Bakker-Woudenberg. 2010.
Time-kill kinetics of anti-tuberculosis drugs, and emergence of resistance, in relation
15

426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
to metabolic activity of Mycobacterium tuberculosis. The Journal of antimicrobial
chemotherapy 65:2582-2589.
9.
Emsley, P., and K. Cowtan. 2004. Coot: model-building tools for molecular
graphics. Acta Crystallogr D Biol Crystallogr 60:2126-2132.
10.
Englisch, S., U. Englisch, F. von der Haar, and F. Cramer. 1986. The proofreading
of hydroxy analogues of leucine and isoleucine by leucyl-tRNA synthetases from E.
coli and yeast. Nucleic acids research 14:7529-7539.
11.
12.
13.
Graessle, O. E., and J. J. Pietrowski. 1949. The in vitro effect of para-
aminosalicylic acid (PAS) in preventing acquired resistance to streptomycin by
Mycobacterium tuberculosis. Journal of bacteriology 57:459-464.
Hamasaki, K., and R. R. Rando. 1997. Specific binding of aminoglycosides to a
human rRNA construct based on a DNA polymorphism which causes
aminoglycoside-induced deafness. Biochemistry 36:12323-12328.
Hernandez, V., T. Crepin, A. Palencia, S. Cusack, T. Akama, S. J. Baker, W. Bu,
L. Feng, Y. R. Freund, L. Liu, M. Meewan, M. Mohan, W. Mao, F. L. Rock, H.
Sexton, A. Sheoran, Y. Zhang, Y. K. Zhang, Y. Zhou, J. A. Nieman, M. R.
Anugula, M. Keramane el, K. Savariraj, D. S. Reddy, R. Sharma, R. Subedi, R.
Singh, A. O'Leary, N. L. Simon, P. L. De Marsh, S. Mushtaq, M. Warner, D. M.
Livermore, M. R. Alley, and J. J. Plattner. 2013. Discovery of a novel class of
boron-based antibacterials with activity against Gram-negative bacteria.
Antimicrobial agents and chemotherapy 57:1394-1403.
14.
Ioerger, T. R., T. O'Malley, R. Liao, K. M. Guinn, M. J. Hickey, N. Mohaideen,
K. C. Murphy, H. I. Boshoff, V. Mizrahi, E. J. Rubin, C. M. Sassetti, C. E. Barry,
3rd, D. R. Sherman, T. Parish, and J. C. Sacchettini. 2013. Identification of new
16

450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
drug targets and resistance mechanisms in Mycobacterium tuberculosis. PloS one
8:e75245.
15.
16.
Kabsch, W. 1993. Automatic processing of rotation diffraction data from crystals of
initially unknown symmetry and cell constants. J. Appl. Cryst. 26:795-800.
Lee, M., J. Lee, M. W. Carroll, H. Choi, S. Min, T. Song, L. E. Via, L. C.
Goldfeder, E. Kang, B. Jin, H. Park, H. Kwak, H. Kim, H. S. Jeon, I. Jeong, J. S.
Joh, R. Y. Chen, K. N. Olivier, P. A. Shaw, D. Follmann, S. D. Song, J. K. Lee, D.
Lee, C. T. Kim, V. Dartois, S. K. Park, S. N. Cho, and C. E. Barry, 3rd. 2012.
Linezolid for treatment of chronic extensively drug-resistant tuberculosis. The New
England journal of medicine 367:1508-1518.
17.
18.
Lenaerts, A. J., V. Gruppo, J. V. Brooks, and I. M. Orme. 2003. Rapid in vivo
screening of experimental drugs for tuberculosis using gamma interferon gene-
disrupted mice. Antimicrobial agents and chemotherapy 47:783-785.
Lenaerts, A. J., V. Gruppo, K. S. Marietta, C. M. Johnson, D. K. Driscoll, N. M.
Tompkins, J. D. Rose, R. C. Reynolds, and I. M. Orme. 2005. Preclinical testing of
the nitroimidazopyran PA-824 for activity against Mycobacterium tuberculosis in a
series of in vitro and in vivo models. Antimicrobial agents and chemotherapy
49:2294-2301.
19.
20.
Lincecum, T. L., Jr., M. Tukalo, A. Yaremchuk, R. S. Mursinna, A. M. Williams,
B. S. Sproat, W. Van Den Eynde, A. Link, S. Van Calenbergh, M. Grotli, S. A.
Martinis, and S. Cusack. 2003. Structural and mechanistic basis of pre- and
posttransfer editing by leucyl-tRNA synthetase. Molecular cell 11:951-963.
Liu, Y., J. Liao, B. Zhu, E. D. Wang, and J. Ding. 2006. Crystal structures of the
editing domain of Escherichia coli leucyl-tRNA synthetase and its complexes with
17

474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
Met and Ile reveal a lock-and-key mechanism for amino acid discrimination. The
Biochemical journal 394:399-407.
21.
22.
Lue, S. W., and S. O. Kelley. 2005. An aminoacyl-tRNA synthetase with a defunct
editing site. Biochemistry 44:3010-3016.
McCoy, A. J., R. W. Grosse-Kunstleve, L. C. Storoni, and R. J. Read. 2005.
Likelihood-enhanced fast translation functions. Acta Crystallogr D Biol Crystallogr
61:458-464.
23.
24.
Nagley, M. M. 1950. The combined use of para-aminosalicylic acid (PAS) and
streptomycin in pulmonary tuberculosis. Tubercle 31:151-155.
O'Dwyer, K., A. T. Spivak, K. Ingraham, S. Min, D. J. Holmes, C. Jakielaszek, S.
Rittenhouse, A. L. Kwan, G. P. Livi, G. Sathe, E. Thomas, S. Van Horn, L. A.
Miller, M. Twynholm, J. Tomayko, M. Dalessandro, M. Caltabiano, N. E.
Scangarella-Oman, and J. R. Brown. 2015. Bacterial resistance to leucyl-tRNA
synthetase inhibitor GSK2251052 develops during treatment of complicated urinary
tract infections. Antimicrobial agents and chemotherapy 59:289-298.
Ollinger, J., M. A. Bailey, G. C. Moraski, A. Casey, S. Florio, T. Alling, M. J.
Miller, and T. Parish. 2013. A dual read-out assay to evaluate the potency of
compounds active against Mycobacterium tuberculosis. PloS one 8:e60531.
Palencia, A., T. Crepin, M. T. Vu, T. L. Lincecum, Jr., S. A. Martinis, and S.
Cusack. 2012. Structural dynamics of the aminoacylation and proofreading functional
cycle of bacterial leucyl-tRNA synthetase. Nature structural & molecular biology
19:677-684.
25.
26.
27.
Palminteri, R., and D. Sassella. 1979. Double-blind multicenter trial of a rifampicin-
trimethoprim combination and rifampicin alone in urinary tract infections.
Chemotherapy 25:181-188.
18

499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
28.
29.
Perrakis, A., R. Morris, and V. S. Lamzin. 1999. Automated protein model
building combined with iterative structure refinement. Nat Struct Biol 6:458-463.
Rock, F. L., Mao, W., A. Yaremchuk, M. Tukalo, T. Crepin, H. Zhou, Y. K.
Zhang, V. Hernandez, T. Akama, S. J. Baker, J. J. Plattner, L. Shapiro, S. A.
Martinis, S. J. Benkovic, S. Cusack, and M. R. Alley. 2007. An antifungal agent
inhibits an aminoacyl-tRNA synthetase by trapping tRNA in the editing site. Science
316:1759-1761.
30.
Rullas, J., J. I. Garcia, M. Beltran, P. J. Cardona, N. Caceres, J. F. Garcia-
Bustos, and I. Angulo-Barturen. 2010. Fast standardized therapeutic-efficacy assay
for drug discovery against tuberculosis. Antimicrobial agents and chemotherapy
54:2262-2264.
31.
32.
Sassanfar, M., J. E. Kranz, P. Gallant, P. Schimmel, and K. Shiba. 1996. A
eubacterial Mycobacterium tuberculosis tRNA synthetase is eukaryote-like and
resistant to a eubacterial-specific antisynthetase drug. Biochemistry 35:9995-10003.
Seiradake, E., W. Mao, V. Hernandez, S. J. Baker, J. J. Plattner, M. R. Alley,
and S. Cusack. 2009. Crystal structures of the human and fungal cytosolic leucyl-
tRNA synthetase editing domains: A structural basis for the rational design of
antifungal benzoxaboroles. Journal of molecular biology 390:196-207.
Shandil, R. K., R. Jayaram, P. Kaur, S. Gaonkar, B. L. Suresh, B. N. Mahesh, R.
Jayashree, V. Nandi, S. Bharath, and V. Balasubramanian. 2007. Moxifloxacin,
ofloxacin, sparfloxacin, and ciprofloxacin against Mycobacterium tuberculosis:
evaluation of in vitro and pharmacodynamic indices that best predict in vivo efficacy.
Antimicrobial agents and chemotherapy 51:576-582.
33.
19

522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
34.
35.
Sirgel, F. A., I. J. Wiid, and P. D. van Helden. 2009. Measuring minimum
inhibitory concentrations in mycobacteria. Methods in molecular biology 465:173-
186.
Stover, C. K., P. Warrener, D. R. VanDevanter, D. R. Sherman, T. M. Arain, M.
H. Langhorne, S. W. Anderson, J. A. Towell, Y. Yuan, D. N. McMurray, B. N.
Kreiswirth, C. E. Barry, and W. R. Baker. 2000. A small-molecule
nitroimidazopyran drug candidate for the treatment of tuberculosis. Nature 405:962-
966.
36.
37.
38.
Sutherland, R., R. J. Boon, K. E. Griffin, P. J. Masters, B. Slocombe, and A. R.
White. 1985. Antibacterial activity of mupirocin (pseudomonic acid), a new antibiotic
for topical use. Antimicrobial agents and chemotherapy 27:495-498.
Vondenhoff, G. H., and A. Van Aerschot. 2011. Aminoacyl-tRNA synthetase
inhibitors as potential antibiotics. European journal of medicinal chemistry 46:5227-
5236.
Woese, C. R., G. J. Olsen, M. Ibba, and D. Soll. 2000. Aminoacyl-tRNA
synthetases, the genetic code, and the evolutionary process. Microbiology and
molecular biology reviews : MMBR 64:202-236.
539
20

540
ACKNOWLEDGMENTS
541
542
543
544
545
546
547
548
The authors would like to thank Dr. Ding Zhou and his team at PharmaResources, Dr.
Michael Xu and his team at WuXi AppTec, and Dr. James Nieman and his team at NAEJA
Pharmaceuticals for their contributions to medicinal chemistry. The authors thank the staff of
the EMBL HTX (high-throughput crystallization) platform within the Grenoble Partnership
for Structural Biology (PSB) and of the ESRF-EMBL Joint Structural Biology Group for
access to ESRF beamlines. AL acknowledges the NIH/NIAID Contract NO1 AI-95385
(TAACF)) for the efficacy testing in acute and chronic mouse models of infection. We thank
James Ahn for technical assistance.
549
550
551
ACCESSION CODES
Atomic coordinates and structure factors for the compound 2, 11 and 13 have been deposited
in wwPDB with the following codes, 5AGR, 5AGS, 5AGT.
552
553
21

554
Table 1. MIC values for compound 1 resistant mutants.
Organism/Mutant
/SNP
Agar MIC
(µg/mL)
Organism/Mutant
/SNP
Liquid MIC
(µg/mL)
M. tuberculosis
H37Rv
M. smegmatis
ATCC 700084
0.6-1.3
1
RM1leuS Y435C
RM2leuS S311L
RM3leuS D450Y
RM4leuS S311L
21
21
21
21
RM1leuSΔ421-462
RM2leuS A428T
RM3leuS R435C
RM4leuS A428T
>256
64
8
64
555
556
557
558
559
560
The residues Y435 and D450 stabilize the adduct formed by compound 1 with AMP in the
editing site of M. tuberculosis LeuRS (Figure 4b). The residue S311, like the equivalent E.
coli LeuRS residue (20), interacts with phosphate of Ade76 thus stabilizing the adduct in the
editing site. However, S311 is located at the flexible N-terminal part of our editing domain
construct of M. tuberculosis LeuRS and thus is not visible in the crystal structure (Figure 4b).
SNP = single nucleotide polymorphism, RM = resistant mutant.
22

561
562
Table 2. Compound 1 MIC (µg/mL) against M. tuberculosis Erdman and monoresistant
isolates compared with known standards
Strain
Cmp 1
0.127
PA-824
0.116
RIF
INH
STR
0.018
0.244
0.369
M. tuberculosis
Erdman
0.128
≤0.063
0.189
>4
0.383
>8
0.216
0.202
>16
rRIF
0.120
0.037
0.082
rINH
rSTR
0.059
0.113
0.344
563
564
r, resistance to RIF = Rifampin, INH= Isoniazid or STR= Streptomycin
23

565
566
Table 3. Thermodynamic analysis of the interaction between M. tuberculosis LeuRS and
benzoxaborole compounds.
*
Compound/
parameter
Kd
ΔGap
(Kcal mol-1)
ΔHap
(Kcal mol-1)
-TΔSap
(Kcal mol-1)
n
(µM)
2
3.7
- 7.4
-9.7
-1.1
-4.2
-5.5
-6.3
-5.5
-4.6
1.05
13
11
0.075
0.040
1.19
1.02
-10.0
567
568
569
* The error in the thermodynamic binding parameters is about 5% for the apparent binding enthalpy, and 10%
for the apparent binding constant and the number of sites (n). Values in the table are the average of at least 2
independent experiments.
570
24

571
Table 4. Murine Pharmacokinetic Parameters
Cmp2
30
Cmp11
15
Cmp12
30
Cmp13
30
Cmp14
30
*IV
Cmax (µg/mL) @ 5 min
CL (mL/h/kg)
Vss (mL/kg)
8.9
2180
2116
2.1
18.0
328
968
13.7
1119
3805
13.6
582
3142
17.1
687
3221
MRT (h)
3.0
3.4
5.4
4.7
AUC0-∞ (h*µg/mL)
α-t1/2 (hr) [%AUC]
β-t1/2 (hr) [%AUC]
13.8
0.06[5]
1.5[95]
45.8
0.09 [2]
2.08 [98]
26.8
51.6
0.10 [2]
3.83 [98]
43.7
0.05 [5]
3.40 [95]
0.11 [7]
2.53 [93]
#PO (mg/kg)
30
30
30
30
30
Cmax (µg/mL)
Tmax (h)
AUC0-24 (h*µg/mL)
Terminal t1/2 (h)
Bioavailability (%)
Mouse PPB(%)
3.4
0.50
13.2
1.8
96
7.2
1.00
35.9
2.7
39
50
5.0
1.00
23.8
2.7
89
16
6.4
0.25
47.5
3.1
92
6.3
0.50
57.6
3.6
100
-
6
23
572
573
*WinNonlin two-compartment analysis iterative weighting.
#WinNonlin non-compartment analysis with uniform weighing.
25

574
575
Table 5. Mitochondrial protein synthesis inhibition
Cell
Viability
EC50
SDHA
EC50
(µM)
COX1
EC50
(µM)
Compound
(µM)
Compound 12
Compound 13
Compound 14
Linezolid
>150
>150
39.5 ± 9.2
20.5 ± 2.1
21.5 ± 6.4
>150
23.0 ± 1.4
80.0 ± 5.7
106 ± 37.5
>150
>150
27.3 ± 10.8
31.4 ± 23.2
Chloramphenicol
>150
110 ± 14.1
Doxycycline
Erythromycin
23.7 ± 6.4
>150
109 ± 29
>150
118 ± 35.5
>150
576
577
578
579
580
COX1 is cytochrome c oxidase, which is a mitochondrial protein that is
synthesized by mitochondrial ribosomes. SDHA is subunit A of succinate
dehydrogenase complex, which is a mitochondrial protein that is
synthesized by cytoplasmic ribosomes. Janus Green staining was used to
determined cell viability after 7 days.
581
26

582
Table 6. In vitro resistance frequency
Compound
4xMIC
10xMIC
Compound 14
4.6 x 10-6
3.9 x 10-6
Isoniazid
ND
1.8 x 10-5
1.1 x 10-8
Moxifloxacin
1.7 x 10-7
583 The MIC values for compound 14, isoniazid and moxifloxacin were
584 determined on Middlebrook 7H10 agar as 0.2, 0.06 and 0.08 µg/mL,
585 respectively.
586
27

587
588
FIGURE LEGENDS
Figure 1. In vitro structure-activity relationship. Mtb = M. tuberculosis, NT= not tested.
589
590
591
592
593
Figure 2. Compound 1 resistant mutants bear mutations in the editing domain of LeuRS.
(A) Domain map of M. tuberculosis LeuRS. (B) Amino acid alignment of part of the editing
domain of LeuRS from M. tuberculosis and M. smegmatis, identical residues are colored in
blue, non-identical residues are colored in red with arrows indicating where the mutations
were found.
594
595
596
597
598
599
600
Figure 3. In vivo efficacy of compound 2 in a murine GKO (C57BL/6-Ifngtm1ts) model of
acute TB. Oral treatment was started 15 days (Start) after infection with a low dose aerosol of
M. tuberculosis Erdman lux and continued for 9 consecutive daily treatments until day 23
when mice were euthanized on day 24 (End). CFU were determined from lungs (black) and
spleens (grey) and means from five mice for drug treated groups and 6 mice per group for the
untreated controls. *P < 0.001 by pairwise multiple comparison procedures (Tukey test)
compared to control.
601
602
603
604
605
606
607
608
609
610
611
Figure 4. X-ray co-crystal structure of LeuRS with compound 2. (A) Crystal structure of M.
tuberculosis LeuRS editing domain in complex with compound 2 (carbon atoms are colored
in green)-AMP (carbon atoms are colored in magenta). Color code is the same throughout all
figures with blue for nitrogen, red for oxygen, pink for boron, orange for phosphorus, yellow
for sulfur. (B) Zoomed view into the editing site of M. tuberculosis LeuRS showing the
compound 2-AMP adduct and the key residues establishing important hydrogen bonds (red
dashed lines) with only the H-bond from the 3-aminomethyl to M441 being omitted for
clarity. (C) Overlay of the LeuRS editing domain of M. tuberculosis and E. coli in complex
with methionine colored in yellow (PDB: 2AJF). The 3-aminomethyl group of compound 2
mimics the amino group of methionine, including the interaction to the bacterial specific
residue D447.
612
613
614
615
616
617
Figure 5. In vivo efficacy of compounds 11, 12, 13 and 14 in acute and chronic models of
TB infection. (A) In vivo efficacy in a murine GKO (C57BL/6-Ifngtm1ts) model of acute TB.
Compounds were dosed orally daily for 14 days after 10 days infection (Start) with a low
dose aerosol of M. tuberculosis Erdman. Mean lung CFU were determined from five mice at
End. (B) In vivo efficacy in a murine GKO (C57BL/6-Ifngtm1ts) model of acute TB. Oral
treatment was started 13 days after infection (Start) with low dose aerosol of M. tuberculosis
28

618
619
620
621
622
623
624
Erdman lux and continued for 9 consecutive daily treatments until day 21 when mice were
sacrificed on day 22 (End). Mean lung CFU were determined from five mice at End. (c) In
vivo efficacy in a murine BALB/c model of chronic TB infection. Compounds were dosed
orally 5 days a week for 4 weeks after infecting with M. tuberculosis Erdman with a low dose
aerosol 21 days prior (Start). Lung (black) and spleen (grey) CFU were determined from six
mice at End. **P< 0.01, *P < 0.001 by pairwise multiple comparison procedures (Tukey
test) compared to control.
625
626
627
628
629
630
Figure 6. M. tuberculosis H37Rv in vitro kill kinetics. Cells were incubated with compounds
at 20-fold their MIC values for different times over 14 days in 10 mL of 7H9 10% (v/v) ADC
and 0.05% (v/v) Tween 80 medium. The MIC values used in this experiment were as
follows, 0.013 µg/mL, 0.6 µg/mL and 0.06 µg/mL for compound 14, linezolid and
moxifloxacin, respectively. The mean and the standard deviations of at triplicate cultures of
each point are shown.
631
632
633
634
635
636
Figure 7. Efficacy of compound 14 in a mouse model of chronic TB infection. C57 BL/6J
mice were infected with M. tuberculosis H37Rv intratracheally (~102 CFU) and were dosed
once daily for 8 weeks starting 6 weeks after infection. Mice were sacrificed 24 hours after
the last drug administration. Every column represents the mean value +/- SD of 7 mice per
group for untreated and Linezolid treated groups and 3 mice for Compound 14 treated mice.
*P < 0.001 by pairwise multiple comparison procedures (Tukey test) compared to control.
637
638
639
640
641
642
643
Figure 8. Efficacy of compound 14 in a mouse model of acute TB infection under different
dosing regimes of once a day (QD), twice a day (BID) or every other day (q48h). C57 BL/6J
mice were infected with M. tuberculosis H37Rv intratracheally (~105 CFU) and were dosed
starting on the following day after infection for 8 days. Only one dose was administered on
day 8 under the BID schedule. Mice were sacrificed at least 24 hours after the last drug
administration. Every dot represents one mouse data point except for Linezolid (mean of 5
mice +/- SD).
29

FIG. 1
7
4
g/mL)
1
6
5
3
7.5
1.0
1.8
0.26
0.13
NT
0.04
1.2
0.05
>1.1
0.04
0.29
17
0.05
0.05
0.02
0.02