Development of methionyl-trna synthetase inhibitors as antibiotics for gram-positive bacterial infections

Article abstract of DOI:10.1128/AAC.00999-17

Antibiotic-resistant bacteria are widespread and pose a growing threat to human health. New antibiotics acting by novel mechanisms of action are needed to address this challenge. The bacterial methionyl-tRNA synthetase (MetRS) enzyme is essential for protein synthesis, and the type found in Gram-positive bacteria is substantially different from its counterpart found in the mammalian cytoplasm. Both previously published and new selective inhibitors were shown to be highly active against Gram-positive bacteria with MICs of 1.3 g/ml against Staphylococcus, Enterococcus, and Streptococcus strains. Incorporation of radioactive precursors demonstrated that the mechanism of activity was due to the inhibition of protein synthesis. Little activity against Gram-negative bacteria was observed, consistent with the fact that Gram-negative bacterial species contain a different type of MetRS enzyme. The ratio of the MIC to the minimum bactericidal concentration (MBC) was consistent with a bacteriostatic mechanism. The level of protein binding of the compounds was high (95%), and this translated to a substantial increase in MICs when the compounds were tested in the presence of serum. Despite this, the compounds were very active when they were tested in a Staphylococcus aureus murine thigh infection model. Compounds 1717 and 2144, given by oral gavage, resulted in 3- to 4-log decreases in the bacterial load compared to that in vehicle-treated mice, which was comparable to the results observed with the comparator drugs, vancomycin and linezolid. In summary, the research describes MetRS inhibitors with oral bioavailability that represent a class of compounds acting by a novel mechanism with excellent potential for clinical development.

Full text of DOI:10.1128/AAC.00999-17

AAC Accepted Manuscript Posted Online 28 August 2017
Antimicrob. Agents Chemother. doi:10.1128/AAC.00999-17
Copyright © 2017 American Society for Microbiology. All Rights Reserved.
1
2
3
4
Development of methionyl-tRNA synthetase inhibitors as antibiotics
for Gram-positive bacterial infections
5
6
7
Omeed Faghiha, Zhongsheng Zhangb, Ranae M. Ranadea, J. Robert Gillespiea, Sharon A.
Creasona, Wenlin Huangb, Sayaka Shibatab, Ximena Barros-Álvarezb,c, Christophe L. M. J.
Verlindeb, Wim G. J. Holb, Erkang Fanb, Frederick S. Bucknera
8
9
10
11
Department of Medicinea, and Department of Biochemistryb, University of Washington, Seattle,
Washington, USA; Laboratorio de Enzimología de Parásitos, Facultad de Ciencias, Universidad
de los Andes, Mérida, Venezuelac.
12
Address correspondences to Frederick S. Buckner, fbuckner@uw.edu
13
14
1

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
ABSTRACT
Antibiotic resistant bacteria are widespread and pose a growing threat to human health. New
antibiotics acting by novel mechanisms of action are needed to combat this threat. The
bacterial methionyl-tRNA synthetase (MetRS) enzyme is essential for protein synthesis and the
type found in Gram-positive bacteria is substantially different from its counterpart found in the
mammalian cytoplasm. Selective inhibitors, both previously published and new, were shown to
be highly active on Gram positive bacteria with minimum inhibitory concentrations (MICs) ≤ 1.3
µg/mL against Staphylococcus, Enterococcus, and Streptococcus strains. Incorporation of
radioactive precursors demonstrated the mechanism of activity was due to inhibition of protein
synthesis. Little activity was observed against Gram-negative bacteria, consistent with Gram-
negative species containing a different type of MetRS enzyme. The ratio of MIC to minimum
bactericidal concentrations (MBC) was consistent with a bacteriostatic mechanism. Protein
binding was high (>95%) for the compounds and this translated to a substantial increase in
MICs when tested in the presence of plasma. Despite this, the compounds were very active
when tested in the Staphylococcus aureus murine thigh infection model. Compounds 1717 and
2144 given by oral gavage resulted in 3-4-log decreases in bacterial load compared to vehicle-
treated mice which was comparable to results observed with the comparator drugs, vancomycin
and linezolid. In summary, the research describes MetRS inhibitors with oral bioavailability that
represent a class of compounds acting by a novel mechanism with excellent potential for clinical
development.
INTRODUCTION
Gram-positive bacteria such as Staphylococcus, Streptococcus, and Enterococcus are major
human pathogens responsible for a myriad of clinical syndromes. Antibiotic resistant strains
such as methicillin resistant Staphylococcus aureus (MRSA) and vancomycin resistant
2

41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
Enterococcus (VRE) are widespread and limit the effectiveness of available antibiotics.
Concern about the diminishing availability of effective antibiotics has led to urgent calls for the
development of new antibiotics(1). Targeting the prokaryotic protein synthesis machinery has
been a highly successful strategy for developing antibiotics. Aminoglycosides, tetracyclines,
macrolides, ketolides, and oxazolidinones are major classes of antibiotics that all interfere with
bacterial protein translation. Inhibition of tRNA synthetases represents another possible
approach to target prokaryotic protein translation. The widely used antibiotic, mupirocin, works
by inhibiting the bacterial isoleucyl-tRNA synthetase (2). Mupirocin is used as an ointment to
treat or decolonize patients with cutaneous infections due to Staphylococcus or Streptococcus,
however, its use is limited to the topical route of administration. Another bacterial tRNA
synthetase inhibitor, a boron-containing compound targeting the bacterial leucyl-tRNA
synthetase (GSK2251052) made it to phase 2 trials for Gram negative infections (3).
Unfortunately, its development was discontinued due to high rates of resistance occurring
during treatment, which may be related to the targeting of the editing domain of the enzyme
rather than the catalytic domain. Investigators at GlaxoSmithKline reported on inhibitors to the
S. aureus methionyl-tRNA synthetase (MetRS) over a decade ago (4-7). These inhibitors had
excellent antibiotic potency, but poor oral bioavailability that restricted its development (pre-New
Drug Application) to topical use for skin infections and to oral use for Clostridium difficile
infections where oral absorption is not needed (8, 9). The research in the current report also
focuses on MetRS inhibitors, building on compounds that are being developed as antiprotozoan
chemotherapies (10-13). The compounds have high selectivity (>1000-fold) for Trypanosoma
brucei cells over mammalian cell lines (14) Changes to the molecules have led to improved
oral bioavailability and pharmacokinetic properties (14), thus making them better candidates for
antibiotic development as will be described.
65
3

66
67
68
69
With respect to the target, bacteria and all living organisms contain a complement of tRNA
synthetases that are responsible for charging tRNAs with their corresponding amino acids for
subsequent delivery to the ribosome. tRNA synthetases, including MetRS, catalyze a two-step
reaction as follows:
70
71
E + aa + ATP  E•aa∼AMP + PPi
(1)
E•aa∼AMP + tRNA  E + aa-tRNA + AMP (2)
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
In the first step, a highly reactive aminoacyl adenylate (aa∼AMP) is formed through condensing
ATP with the carboxylate of the amino acid. The second step uses this activated species to
transfer the amino acid to the 3’-end of the tRNA (aa-tRNA) (15). The bacterial MetRS enzymes
are categorized in two forms (MetRS1 and MetRS2) based on sequence similarity and
sensitivity to inhibitors (16). Bacteria generally have a single MetRS enzyme with most Gram
positive genera containing the MetRS1 form (Staphylococcus, Streptococcus, Enterococcus,
Bacillus, Clostridium, and others) and most Gram negative bacteria containing the MetRS2 form
(Escherichia, Klebsiella, Pseudomonas, Haemophilus, Bacteroides, and others) (17).
Exceptions include Bacillus anthracis and a subset of Streptococcus pneumoniae which contain
both MetRS1 and MetRS2 isoforms (16, 18). In mammals, distinct tRNA synthetases typically
operate in the cytoplasm and the mitochondria. The human mitochondrial MetRS encoded in
the mitochondrial genome (19) has close sequence homology to bacterial enzymes of the
MetRS1 variety, whereas the human cytoplasmic MetRS is nuclear encoded with close
homology to the MetRS2 variety. As will be detailed below, the MetRS inhibitors under study in
this project are active on the S. aureus MetRS enzyme and show broad spectrum activity on
Gram positive bacteria and negligible activity on Gram negative bacteria, consistent with
targeting the MetRS1 form of the enzyme. Microbiological properties, murine pharmacology,
and efficacy in the murine S. aureus thigh infection model are described herein. The new
compounds represent promising antibiotic candidates that act by a novel mechanism of action.
4

91
92
93
RESULTS
94
95
MetRS inhibitors and lead optimization. The structures and properties of compounds under
study in these experiments are shown in Fig. 1. The synthetic procedures for compounds 1312-
1717 were published previously (10, 14). The synthetic procedures for new compounds 1962,
2062, 2093, 2114, and 2144 are described in the supplementary materials. The starting point for
these investigations was the aminoquinolone scaffold exemplified by 1312 (Fig. 1). In separate
research to develop MetRS inhibitors as antiprotozoan drugs, our group introduced changes to
the molecules with the goals of improving oral bioavailability while retaining potent activity on
the MetRS target. The evolution of these compounds included changing the aminoquinolone
group to a fluorinated-imidazopyridine (e.g. 1614) that improved oral bioavailability in mice from
<10% for 1312 to ~40% for 1614 (14). Subsequent changes to the linker region reported
previously (13) and in this paper have further improved potency and pharmacological properties
of the series. The results of testing the MetRS inhibitors against recombinant S. aureus MetRS,
bacterial cultures, and mammalian cells are shown in Table 1. All compounds tested had IC50
values on the S. aureus MetRS below the level of sensitivity of the assay (25 nM). The MetRS
inhibitors have potent activity on a variety of Gram positive bacterial strains, but essentially no
activity on Gram negative bacteria (E. coli and Pseudomonas aeruginosa). Specifically, MIC
values below 0.3 µg/mL were measured against strains of S. aureus (including MSSA, MRSA,
and VISA), S. epidermidis, E. faecalis, and E. faecium (including VSE and VRE strains). The
compounds with the lowest MICs were 1717, 2093, and 2144 which were >10-times more
potent than the control drugs vancomycin or linezolid against many strains. These compounds
are the subject of further investigations discussed below. Higher MICs were seen against
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
S. pyogenes and no activity seen on S. pneumoniae. The selectivity on Staphylococci versus
5

117
118
mammalian cells (comparing MIC to CC50) was at least 35-fold for these three most potent
compounds.
119
Microbiological characterization of selected compounds.
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
1. Macromolecular synthesis assays: In order to verify that the compounds are acting by
the expected mechanism of action, radioisotope incorporation assays were performed
(Fig. 2). Incorporation of the amino acid (3H-Lysine) was inhibited by the MetRS
inhibitors (1717, 2093, and 2144), consistent with inhibition of protein synthesis. The
findings were similar to those seen with linezolid (Fig. 2, panel A) which is known to
inhibit protein synthesis by interfering with the bacterial ribosome (20). In contrast, the
MetRS inhibitors had less effect on both the incorporation of 3H-uridine (a measure of
RNA synthesis) and the incorporation of 3H-thymidine (a measure of DNA synthesis)
(21). Ciprofloxacin showed selective inhibition of DNA synthesis (panel B) consistent
with its mechanism as an inhibitor of DNA topoisomerases. Finally, rifampicin showed
selective inhibition of RNA synthesis (slightly above protein synthesis) (panel C)
consistent with its mechanism as an inhibitor of bacterial RNA polymerase.
2. Activity on permeable E. coli strains: The purpose of these experiments was to
determine if the non-susceptibility of Gram negative strains (e.g. E. coli ATCC 25922,
shown in Table 2) was due to inability of the MetRS inhibitors to penetrate the Gram-
negative cell wall. The mutant MB4902 is an outer membrane permeable E. coli strain
and showed no greater susceptibility to three MetRS inhibitors (1717, 2093, and 2144)
than to the wild-type E. coli strain (MB4827). Similarly, the efflux negative strain
MB5747 showed no increased susceptibility to the MetRS inhibitors, nor did the mutant
containing both mutations (MB5746). The control drug, novobiocin, had increased
activity in the hyperpermeable E. coli strains as has been previously reported (22, 23).
6

142
143
144
145
146
147
148
3. MIC/MBC: Measurements of minimum bactericidal concentrations (MBCs) were done
with the S. aureus strain ATCC 29213 (Table 3). The MBC is defined as the drug
concentration that reduces bacterial growth by ≥99.9%. Compounds exhibiting an
MBC/MIC ratio of ≤4 are generally considered bactericidal, while an MBC/MIC ratio >4 is
considered bacteriostatic (24). The data indicates that 1717, 2093, and 2144 have
bacteriostatic activity similar to linezolid.
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
4. Resistance frequency rates: The propensity for S. aureus (ATCC 29213) to develop
resistance to MetRS inhibitors was also studied (Table 4). This was done by plating high
numbers (3.8 x109 in Expt. 1 and 5.5 x 109 in Expt. 2) of S. aureus on TSA plates
impregnated with compound at concentrations of 4X or 8X the MIC and incubating for 72
h. The resistance frequency rates for 1717, 2093, and 2144 at 8X EC50 were in the
range of 2 x 10-8 to 4 x 10-9. These rates are comparable to test drug novobiocin, but
higher than the rates found for ciprofloxacin or linezolid.
5. Serum shift and protein binding assays. Serum shift assays were done to analyze the
impact of protein binding on the MICs (Table 5). The MIC shifts in the presence of 50%
human serum range from 16-fold to 128-fold for the MetRS inhibitors which is consistent
with high protein binding (e.g. 95.4% for 1717). Although the shifts are much higher than
the shift for vancomycin (2-fold), the absolute MICs for some compounds in serum (e.g.
1717 and 2144) are still comparable to that of vancomycin (in the range of 1-4 µg/mL).
Metabolic stability studies:
7

166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
Compounds were incubated with murine or human liver microsomes to evaluate stability to
hepatic metabolic enzymes (Table 6). Metabolism rates were similar between mouse and
human microsomes for individual MetRS inhibitors. Compound 2093 demonstrated the most
rapid metabolism (3.2 min in mouse microsomes) whereas compound 1962 demonstrated the
highest metabolic stability (27.0 min in mouse microsomes). The drug, linezolid, is more
metabolically stable with half-lives of >145 min in both mouse and human liver microsomes.
Pharmacokinetics studies:
Selected MetRS inhibitors (1614, 1717, 2093, and 2144) or linezolid were administered to mice
in single oral or IV doses, and tail blood was sampled at time intervals out to 24 h to assess
blood exposure (Table S2). The terminal half-life for the MetRS inhibitors in blood ranged from
17 - 58 minutes (compared to 35 min for linezolid). Clearance ranged from 22 - 63 mL/min/kg
(compared to 20 mL/min/kg for linezolid). The maximum blood concentration (Cmax) following
oral dosing ranged from 0.39 - 9.3 uM (compared to 13.7 uM for linezolid). The AUC in blood
following oral dosing ranged from 117 - 615 min*uM (compared to 1707 for linezolid). Finally,
the apparent oral bioavailability ranged from 24 - 46% (compared to 94% for linezolid).
Efficacy studies in mice:
Selected MetRS inhibitors were tested for in vivo efficacy in the neutropenic S. aureus
thigh infection model. Compounds 1717 and 2144 resulted in a ~3-4-log decrease in CFUs
compared to the vehicle group, similar to vancomycin and linezolid (Fig. 3). Note the drop is
below the stasis level which was determined by harvesting a group of mice at 1 hour post-
infection.
Protein sequence analysis. Using coordinates from the T.brucei MetRS complex with inhibitor
1312 (PDB# 4EG5), the residues in the binding site of the inhibitor were aligned for various
8

192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
species (Table 7). The S. aureus MetRS (UniProtKB – A0A0H2XID2, strain USA300) has
extremely high sequence conservation with the Trypanosoma brucei MetRS with 22 of 25
identical amino acids (and potentially 23 identical amino acids since position 456 could either be
a Leu or His, but is ambiguous in the model due to loop length). This confirms that many
inhibitors of the T. brucei MetRS will likely inhibit the S. aureus MetRS. We also compared the
MetRS from the human mitochondrial MetRS with the S. aureus sequence and identified four
different amino acid residues (at positions 249, 291, 470, and 471). Three of these changes
occur in pocket q that binds the quinolone moiety of 1312.
Discussion:
The essential enzyme, methionyl-tRNA synthetase, was targeted for antibiotic drug discovery
against Gram positive bacteria. The research capitalizes on progress to develop antimicrobial
agents against pathogenic protozoa including Trypanosoma brucei and Giardia intestinalis (10-
14, 25). In particular, challenges with poor oral bioavailability observed with early
aminoquinolone compounds such as 1312 (10, 22, 26) were significantly improved with the
fluoroimidazopyridine derivatives (1614-2144, Fig. 1) (14). Compounds of this scaffold were
optimized for activity against the T. brucei MetRS (a type 1 enzyme). The comparison of the
protein sequences of the MetRS enzymes of T. brucei and S. aureus shows identity of 22 of 25
amino acid residues in the inhibitor binding site (Table 7) suggesting that cross activity from
T. brucei to S. aureus was likely. In fact, all the compounds tested for inhibitory activity against
recombinant S. aureus MetRS enzyme (Table 1) demonstrated IC50 values below 25 nM, the
sensitivity limit of the assay. Further titration below this concentration was not possible with the
applied methods due to the need for 25 nM enzyme to give a suitable signal for measurement
(see Methods). The assays against live bacterial cultures demonstrated potency of the MetRS
inhibitors in the sub-microgram per milliliter range against Staphylococcus and Enterococcus
species, and lower potency against Streptococci (Table 1). Of the compounds with the “linear
9

218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
linker” structure (e.g., 1312-1717), compound 1717 was the most potent with a MIC of 0.16
µg/mL. This was the most potent compound against T. brucei cultures (14). Compounds with
the ring sytem in the linker region of the scaffold (e.g. 1962-2144) also had potent anti-
Staphylococcal activity including 2093 and 2144 with MICs of 0.04 and 0.02 µg/mL (Table 1).
Compound 2093 contains a single ring in the linker (an imidazole-2-one) whereas compound
2144 contains a fused imidazo[1,2-a]pyridine ring system in the linker region.
The spectrum of activity of the MetRS inhibitors was explored against ATCC strains of
pathogenic Gram positive and Gram negative bacteria. As was predicted, the antibiotic activity
is restricted to bacteria dependent upon the type 1 MetRS enzyme, i.e. Gram positive bacteria.
Gram negative bacteria (i.e., Escherichia coli and Pseudomonas aeruginosa), which are known
to contain the type 2 MetRS enzyme (17), were insensitive to all the tested compounds at the
highest concentration of 10 µg/mL. The selectivity for Gram positive organisms is potentially
advantageous in that the MetRS inhibitor developed as a drug will not add to resistance of non-
targeted Gram negative bacteria. Sensitive Gram positive strains were S. aureus, Enterococcus
faecium, Enterococcus faecalis, and Staphylococcus epidermidis. Furthermore, these included
drug-resistant strains such as MRSA, VISA, and VRE whose mechanisms of resistance to
semisynthetic penicillins and glycopeptide antibiotics are unrelated to the cellular processes
inhibited by the MetRS inhibitors. The MICs of the MetRS inhibitors to S. pyogenes (ATCC
19615) was higher than to S. aureus and Enterococcus strains which we ascribe to the need to
grow S. pyogenes in media containing lysed blood. We have already shown the effects of
plasma protein binding on MICs of the compounds (Table 5) and we suspect the addition of
laked horse blood to be similar. With the shift observed with blood, the MIC for 2144
(1.3 µg/mL) is about the same as MICs observed with vancomycin and linezolid (0.63 and
1.3 µg/mL). The gram positive coccus, Streptococcus pneumoniae (ATCC 49619), was
resistant to the MetRS inhibitors (MICs >10 µg/mL). This is consistent with previous reports that
~45% of S. pneumoniae strains are resistant to type 1 MetRS inhibitors due to the presence of a
10

244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
second (type 2) MetRS inhibitor in the genome (16). It is likely that MetRS inhibitors would need
to be used with caution for treatment of pneumonia or other clinical syndromes in which
S. pneumoniae is commonly found, at least until cultures rule out S. pneumoniae as the cause
of the infection. Future studies will investigate a broader collection of S. pneumoniae isolates to
assess the MIC range against this pathogen. The issue of a secondary MetRS gene has not
been described in other Gram positive bacteria, so this is unlikely to be a broader concern. We
expect that MetRS inhibitors will be active against many other bacteria containing the type 1
MetRS enzyme including species of Clostridia, Corynebacterium, Bacillus, Propionibacterium,
Actinomyces, and others (16). Various species of these are, of course, pathogenic in humans
and their susceptibility will be tested in future studies. An exception to the Gram positive rule
mentioned above is Brucella (a Gram negative rod) which is known to contain a type 1 MetRS
and is susceptible to MetRS inhibitors (27).
In order to address the question about target of action in living bacteria, macromolecular
synthesis assays were run with MetRS inhibitors and various control drugs (Fig. 2). As was
expected, MetRS inhibitors resulted in rapid dose-dependent decreases in uptake of
radiolabeled amino acid (Lys) consistent with disruption of protein synthesis. The changes were
similar to those seen with the protein synthesis inhibitor, linezolid. At the same time, RNA and
DNA synthesis was unaffected by MetRS inhibitors (2093 and 2144) until concentrations above
the MIC were used while the control drugs, rifampicin and ciprofloxacin, caused selective
inhibition of these pathways, respectively, in the anticipated manner. Compound 1717 showed
mild suppression (20-40%) of RNA and DNA suppression at the MIC which could suggest an
undefined secondary target or perhaps different binding kinetics to MetRS than for 2093 and
2144. These studies provide assurance that the compounds are likely to be mediating their
antibiotic effects through inhibiting the MetRS target in vivo.
In a similar vein, selected MetRS inhibitors were tested on strains of E. coli with defects
in cell wall permeability and/or efflux (Table 2). The purpose of these experiments was to show
11

270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
that resistance of E. coli was not due to exclusion of the MetRS inhibitors by the Gram negative
cell wall or efflux, but rather due to inherent resistance. The findings that the cell permeable
strains were resistant to the three most potent MetRS inhibitors (1717, 2093, and 2144) is
consistent with the understanding that E. coli contains a type 2 MetRS enzyme (17) which is not
inhibited by the compounds under development. Furthermore, it indicates that off target
mechanisms of action are not at play, at least with this species of bacteria.
Minimum bactericidal concentrations of 1717, 2093, and 2144 were determined against
the S. aureus ATCC 29213 strain. The MBC/MIC ratio was between 16 and 32 for these three
compounds. A ratio of 4 or less is considered bactericidal (28), thus the MetRS inhibitors would
be considered bacteriostatic against this strain of S. aureus. An MBC/MIC ratio of 64 was
observed with the clinical drug linezolid (known to be bacteriostatic), whereas the ratio for
vancomycin was 4, consistent with its bactericidal mechanism.
Resistance frequency rates of S. aureus (ATCC 29213) to MetRS inhibitors were
determined on agar plates containing MetRS inhibitors at concentrations 4 or 8-times times their
MICs. The resistance frequency rates for MetRS inhibitors (at 8X MIC) were between 2x10-8
and 4x10-9 which is higher than observed for ciprofloxacin and linezolid (Table 4). Resistance
frequency rates in the 10-6 to 10-9 range are indicative of a single drug target within the cell (28)
which is consistent with the understanding of the mechanism of action of these compounds.
Drugs such as rifampicin have even higher rates (2x10-7) (28), but are generally used in
combination with other drugs to avoid generating resistance. Vancomycin and linezolid are
known to have low resistance frequency rates (<10-11) and, along with this characteristic,
relatively little resistance (at least from Staphylococci) has developed in the clinic. Further
research will be necessary to find out if the rates of resistance to MetRS inhibitors are
problematic for their clinical development as monotherapy agents. If the risk for resistance
developing appears high, then developing the compounds with a partner antibiotic may be an
attractive option to mitigate the problem.
12

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
The MetRS inhibitors characterized in this report exhibit high protein binding properties
(95-99.9%). The low unbound concentration of compounds translates to substantial effects
when MICs are measured in the presence of serum (Table 5). Serum shifts ranging from 16-fold
to 128-fold were observed with the series of tested compounds. For perspective, vancomycin
only demonstrates about a 2-fold serum shift (Table 5) whereas fusidic acid is reported to have
97% protein binding and a 130-fold increase in MIC to S. aureus in the presence of 50% serum
(28). Due to the high potency of the MetRS inhibitors, the MICs of compounds 1717 and 2144
in the presence of 50% serum (1 and 4 µg/mL, respectively) are comparable to the MIC for
vancomycin (2 µg/mL).
Incubation of the MetRS inhibitors with mouse or human liver microsomes showed
variable rates of metabolism, although generally half-lives were relatively short (<20 min for
human microsomes and <10 min for mouse) (Table 6). The pharmacokinetic studies in mice
showed clearance values for MetRS inhibitors ranging from 18-63 mL/min/kg compared to a
value of 20 ml/min/kg for linezolid. The fact that clearance of MetRS inhibitors is similar to
linezolid despite the more rapid microsome metabolism is likely attributable to the high plasma
protein binding which can protect compounds from liver cytochrome P450 metabolism (29).
Oral bioavailability for the MetRS inhibitors ranged from 24-46% which is substantially higher
than the original aminoquinolone compounds such as 1312 (oral bioavailability <10%) (14), but
lower than linezolid (94%). This difference in oral bioavailability is probably responsible for the
higher AUC of linezolid (1707 min*µM) following oral dosing than observed with the MetRS
inhibitors (117 – 615 min*µM). As will be discussed with the efficacy results, the combined
properties of the compounds (particularly 1717 and 2144) appear to be sufficient to clear
bacteria from infected mice with similar efficiency as vancomycin or linezolid.
The results of the efficacy experiments are very encouraging towards the prospects of
developing MetRS inhibitors as antibiotics. A pilot experiment (not shown) and two independent
experiments showed the reproducibility of the S. aureus thigh infection model in mice made
13

322
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
neutropenic by cyclophosphamide pre-treatment. The model represents a soft tissue infection
which resembles the disease process (skin and skin structure infection) for which clinical
development of the compounds would initially be targeted. Both compounds 1717 and 2144
demonstrated significant reduction of bacterial load below the stasis level at least as effectively
as the comparator drugs vancomycin and linezolid. The approximately 2-log reduction below
stasis levels in a neutropenic mouse is noteworthy in light of the bacteriostatic activity observed
in vitro (Table 3). This shows that tissue levels at the site of infection were sufficient to
substantially reduce bacteria levels even in the absence of neutrophils. Many bacteriostatic
antibiotics (including linezolid) are widely and successfully used in the clinic, so the
bacteriostatic characteristic of the MetRS inhibitors may not be a significant liability.
Interestingly, compound 2093 was found to have weaker activity than the other MetRS inhibitors
evaluated. The explanation probably relates to the particularly high protein binding of this
compound that presumably reduce the levels of unbound compound below the threshold
needed to exceed the MBC at the site of infection. Future dose response experiments will help
us determine the relative potency of 1717 and 2144 compared to each other and additional
MetRS inhibitors under development.
To date, we have observed no apparent side-effects of the MetRS inhibitors given to
mice. In this study, uninfected mice received single dose (50 mg/kg) of compounds for PK
analysis had no acute reactions during the 24 hour observation period. In a previous
publication, compounds 1614 and 1717 were administered to mice infected with T. brucei for 10
days at 50 mg/kg PO twice-per-day with no deleterious effects on weight, grooming, or body
condition (14). Cytotoxicity against mammalian cells was low for the MetRS inhibitors (Table 1).
For example, the ratio of CC50 to MIC for compounds 2093 and 2144 were >500 demonstrating
a wide therapeutic window. A potential toxicity concern for the MetRS inhibitors is inhibition of
the mammalian mitochondrial MetRS enzyme which bears close homology to the S. aureus
MetRS (Table 7). Manifestations of this potential toxicity have not been evident with in vitro
14

348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
cytotoxicity testing (the 48 h assay against a lymphocyte and hepatocyte cell lines) nor in mice
as described above. Many antibiotics acting as protein synthesis inhibitors are known to inhibit
mitochondrial protein synthesis as an off-target effect (30, 31). These include widely used drugs
such as tetracycline, erythromycin, aminoglycosides, and linezolid. Instead of directly affecting
mitochondrial oxidative phosphorylation, these drugs interfere with mitochondrial biogenesis and
are relatively slow to result in clinical problems, often with tissue specific toxicity depending on
the particular drug. Linezolid for example is known to cause hematologic disturbances,
peripheral neuropathy, and metabolic acidosis when it is administered for more than a 28-day
period (32). The fact that these side effects are slow to manifest makes them more tolerable for
antibiotics since treatment course are typically relatively short (<10 days). Studies of the effects
on MetRS inhibitors on mammalian mitochondrial function will be part of future investigations.
In summary, with the aid of structure-based drug design (14) new MetRS inhibitors have
been developed with potent and broad spectrum activity against Gram positive bacteria.
Macromolecule labeling studies demonstrate the inhibition of protein synthesis, consistent with
the designed mechanism of action. As with other protein synthesis inhibitors such as
oxazolidinones, tetracyclines, and lincosamides, the MetRS inhibitors have bacteriostatic
properties against S. aureus in vitro. The compounds are highly protein bound which may help
sustain plasma levels in vivo by limiting availability to cytochrome P450 metabolism. At least
two MetRS inhibitors displayed activity in the neutropenic mouse thigh infection model
(comparable to linezolid) which indicates that the free fraction of compound is sufficient to inhibit
bacterial growth. In fact, the bacterial load decreased by 1-2 logs below the status level
indicating that the in vivo activity was not just bacteriostatic, but bactericidal. These studies and
previous reports (10, 14) have shown that the MetRS inhibitors have little in vitro toxicity and
appear well-tolerated when dosed to mice for up to 10 days. Additional preclinical toxicology
studies are planned to further investigate the potential for adverse effects from inhibition of the
mitochondrial MetRS or other off-target activities. In total, these MetRS inhibitors with oral
15

374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
bioavailability represent a class of compounds acting by a novel mechanism with excellent
potential for clinical development.
MATERIALS AND METHODS
Media and culture conditions. Mueller Hinton broth (MHB), cation adjusted Mueller Hinton
broth (CA-MHB), and Brain Heart Infusion broth (BHI) were purchased from Becton Dickinson
(Franklin Lakes, NJ). Tryptic soy agar (TSA) plates, TSA with 5% sheep blood plates, and CA-
MHB with 3% laked horse blood were purchased from Remel (San Diego, CA). MHB was used
to assay all Staphylococcus aureus strains. CA-MHB was used for Staphylococcus epidermidis,
Enterococcus faecalis, Enterococcus faecium, Escherichia coli, and Pseudomonas aeruginosa.
CA-MHB supplemented with 3% laked horse blood was used for Streptococcus pneumoniae
and Streptococcus pyogenes. Staphylococcus, Enterococcus, Escherichia, and Pseudomonas
strains were cultured at 37ºC with ambient air. Streptococcus strains were cultured in 37ºC with
5% CO2. (24, 33). Separate conditions for radiolabeled precursor uptake assays are described
below.
Compounds, reagents, and radiochemicals. The synthesis methods for the following
compounds have been previously described: 1312 (10), 1575 (14), 1614 (14), and 1717 (14).
The synthesis methods for the additional compounds are described in the supplementary
materials: 1962, 2062, 2093, 2114, and 2144. The following antibiotics were purchased
commercially: vancomycin (Sigma-Aldrich, St. Louis, MO), linezolid (Chem-Impex International,
Wood Dale, IL), rifampicin (Chem-Impex International, Wood Dale, IL), ciprofloxacin (Acros
Organics, Geel, Belgium), and novobiocin (Promega, Madison, WI). Ketoprofen was purchased
from Sigma Aldrich (St. Louis, MO). Human pooled serum was purchased from Thermo Fisher
16

400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
Scientific (Waltham, MA). Dulbecco’s phosphate buffered saline with calcium and magnesium
(dPBS) was purchased from Lonza (Basel, Switzerland). [Methyl-3H]-thymidine (2% EtOH, 69.7
Ci/mmol) and [5,6-3H]-uridine (sterile water, 60 Ci/mmol) were purchased from American
Radiolabeled Chemicals (St. Louis, MO). L-[4,5-3H(N)]-lysine (2% EtOH, 82.4 Ci/mmol) was
purchased from PerkinElmer (Waltham, MA). The logarithm of the partition coefficient values
(logP) were calculated with ChemAxon software (Cambridge, MA).
Production of recombinant S. aureus MetRSs. The SaMetRS gene (UniProtKB –
A0A0H2XID2) was PCR amplified (Sense 5’-
GGGTCCTGGTTCGGCTAAAGAAACATTCTATATAACAACCCCAATATAC-3’ and Antisense
5’-CTTGTTCGTGCTGTTTATTATTTAATCACTGCACCATTTGGAATTG-3’) from genomic DNA
isolated from S. aureus (ATCC 29213) cultures. The PCR product was then cloned into the
AVA0421 plasmid and sequence verified. The expression of recombinant protein was
performed as previously described (34). The N-terminal 6-His fusion proteins were purified by
nickel affinity chromatography followed by size exclusion gel chromatography (Superdex 75
26/60; GE Biosciences, Piscataway, NJ).
Enzyme assays. Inhibition of SaMetRS was measured using the ATP depletion assay as
previously described (34) with some modifications. Compounds were pre-incubated for 15
minutes at room temperature in a 96-well plate with 400 µg/mL bulk E. coli tRNA, 25 nM
SaMetRS, 0.1 U/mL pyrophosphatase, 0.2 mM spermine, 0.1 mg/mL bovine serum albumin, 2.5
mM dithiothreitol, 25 mM HEPES-KOH pH 7.9, 10 mM MgCl2, 50 mM KCl, and 2% DMSO.
Reagents were purchased from Sigma-Aldrich or Roche. The reaction was started with the
addition of 150 nM ATP and 20 µM L-methionine and after 120 minute incubation was stopped
by the addition of an equal volume (50 µL) of Kinase-Glo® (Promega). Percent inhibition = 100
X (test compound – AVG low control) / (AVG high control – AVG low control) where the low
17

426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
control is all reagents except the compound and the high control is all reagents except the
compounds and L-methionine. IC50 values were calculated by non-linear regression, sigmoidal-
dose response, in Prism 3.0.
Bacterial strains. Strains with ATCC designations were either obtained directly from the
American Type Culture Collection (Manassas, VA) or were kindly provided by the University of
Washington Clinical Microbiology laboratory. Escherichia coli permeability mutants (properties
shown in Table S1 with supplementary data) were provided as a gift from Dr. Katherine Young
at Merck (Rahway, NJ).
Macromolecular synthesis assays: Methods for measuring uptake of radiolabeled precursors
by S. aureus (ATCC strain 29213) were adapted from previous publications (21, 35, 36). For
these assays, bacteria were grown in defined media (DM): RPMI-1640 pH 7.3 ± 0.1 without
phenol red or L-glutamine (Lonza, Basel, Switzerland) supplemented with 4mM L-glutamine
(Lonza, Basel, Switzerland), 10mM HEPES (Lonza, Basel, Switzerland), and 1% (w/v) D-
glucose (Sigma Aldrich, St. Louis, MO). Fresh overnight cultures grown in DM at 37ºC were
diluted 1:50 in pre-warmed DM and grown at 37ºC with shaking (150 rpm) until reaching an
OD600 of 0.420 correlating to ~1*109 CFUs/mL in mid-log phase. Each compound was assayed
in quadruplicate with an 11-point three-fold serial dilution per radioisotope. A pre-warmed 96-
well V-bottom plate (Corning 3894; Corning, Corning, MA) containing 25 µL of 4X final
concentration of test compound was inoculated with 65 µL of mid-log phase bacteria (OD600 of
0.420). Both positive and negative control wells received 25 µL untreated DM and 65 µL of
inoculum at the same time. After one minute, 10 µL of radiolabeled precursor (10X final
concentration in DM) was added to samples and positive control wells. Final isotope
concentrations for assay of [3H]-lysine (protein), [3H]-thymidine (DNA), and [3H]-uridine (RNA)
were 10 µCi/mL, 2 µCi/mL, and 2µCi/mL respectively. The plates were incubated at 37ºC for 25
18

452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
minutes and terminated by the addition of 50 µL of 30% trichloroacetic acid (TCA)/70% ethanol
to all test and control wells. After termination, 10 µL of 10X radiolabeled precursor was added to
negative control wells. The negative control consisted of adding radiolabeled precursors after
termination of the bacterial incubation in order to represent background measurement of the
isotope. Plates were sealed with plate tape (Thermo Fischer Scientific, Waltham, MA) and
shaken at 250 rpm for one hour at room temperature. Aliquots of 125 µL were transferred from
the 96-well V-bottom plates to 96-well filter plates (Merck Millipore, Billerica, MA). To bind
macromolecules, the samples were passed through the filter membrane (0.45µM hydrophilic
Durapore PVDF membrane) with a vacuum manifold, then the filter was washed with 4 x 200 µL
10% TCA and 1 x 150 µL of 95% ethanol, and dried overnight in vacuum at room temperature.
25 µL Ultima Gold scintillation fluid (Perkin Elmer, Waltham, MA) was added to each well and
DPM was quantified using a MicroBeta2-2450 (Perkin Elmer, Waltham, MA) scintillation counter.
The percent incorporation was determined by subtracting each well by the average negative
background and dividing by the average positive incorporation x 100. Error bars represent SEM
between replicates. The assay was run twice with similar results.
Susceptibility testing. MIC determinations were performed in triplicate in 96-well round
bottom microtiter plates (Corning, Corning NY) as described by the Clinical and Laboratory
Standards Institute (CLSI) (24, 33). Serial two-fold dilutions of compounds were added to plates
in 50 µL volumes. An additional 50 µL of media containing bacterial cells (1x106 CFUs/mL) was
then added to each well. Maximum DMSO concentrations were 0.5%. Plates were incubated
at 37˚C for at least 18 h before reading the susceptibility result by optical absorbance (OD600
)
using a BioTek ELx800. The lowest concentration causing ≥90% growth inhibition compared to
the untreated control was recorded as the MIC (and also corresponded to the visual MIC).
MICs were measured at least twice and the higher value (if different) was recorded herein.
19

477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
MBC determinations (defined as the concentration killing 99.9% of the inoculum) were
performed according to published methods (36,37). Using glass tubes (16 x 100-mm), serial
twofold dilutions of 2x compound were generated from DMSO stocks in singlicate 1mL volumes.
Maximum DMSO concentrations were 0.5%. An additional 1 mL of media with 1*106 CFUs/mL
was added per sample. Each experiment’s inoculum was serially diluted and plated on TSA to
count competent cells. Cultures were incubated at 37ºC for at least 20 h and plated on TSA for
CFU determination. Additionally, each concentration was sampled after the 20 h incubation for
MIC determination as above.
Cytotoxicity testing on mammalian cells: Human cell lines were acquired from the American
Tissue Type Collection: CRL-1855 (lymphoblasts) and HepG2 (hepatocellular carcinoma
cells). The cultures were grown in RPMI medium with 10% fetal bovine serum, penicillin, and
streptomycin at 37°C with 5% CO2. In 96-well plates, the cells were exposed to serial dilutions
of compounds for 48 hours and toxicity was quantified using AlamarBlue (ThermoFisher
Scientific, Waltham, MA) (37). Assays were performed in quadruplicate and EC50 values were
calculated with non-linear regression methods using software by the Collaborative Drug
Database (Burlingame, CA. www.collaborativedrug.com).
Resistance frequency rates determination: The spontaneous resistance frequency rates to
test compounds was determined according to published methods (35, 38). Agar selection
plates were made by adding compound from DMSO stocks into molten Mueller Hinton agar in a
55ºC water bath. Each compound used four plates (P5981-100EA, 150 x 15mm; Sigma Aldrich,
St. Louis, MO) containing 4X or 8x the MIC of the compound. The final DMSO concentration
was <0.1% per plate. Plates were dried in a sterile hood for 30 minutes prior to overnight
storage at 4ºC, and pre-warmed in the 37ºC incubator for 1 hour prior to assay.
20

502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
A fresh overnight culture was diluted 1:50 in MHB and grown at 37ºC with shaking
(150 rpm) until reaching an OD600 of 0.4 correlating to ~2x109 CFUs/mL. Approximately 3 mL
for a total of 6*109 CFUs were distributed onto 4 plates for each compound. Plates were
incubated at 37ºC for 72 h prior to counting of colonies. The starting inoculum was also serially
diluted and plated to quantify initial bacterial load. The resistance frequency was determined as
the number of compound-resistant colonies divided by the total colonies plated.
Serum shift assays: To assess the role of protein binding on compound susceptibility, MIC
determinations were performed in triplicate in the presence and absence of 50% human serum
(39, 40). Serial twofold dilutions of 2x compound were generated in MHB and 50 microliters
were aliquoted onto 96 well plates with a DMSO limit of 0.5%. Bacteria were adjusted to 1*108
CFUs/mL in MHB, then separately further diluted 1:100 in MHB and 100% heat deactivated filter
sterilized pooled human serum. Fifty microliters were added to each well of the corresponding
plates, and the plates were incubated at 37ºC for ~20 h. The lowest concentration causing
≥90% growth inhibition was recorded as the MIC. The median value of three independent
assays is reported.
Protein binding assays: Compound binding to mouse plasma proteins was determined using
96-well equilibrium dialyzer plates (SDIS 9610EN, Nest Group, Inc.). Mouse plasma
(MSEPLLIHP-SW-F, BioreclamationIVT, Westbury, NY) containing compound (final
concentration 1 µM) was added to a donor chamber as a 150 µL volume. The buffer solution
(0.2 mM phosphate buffer, 150 µL) was added to the reciprocal acceptor chamber. Each
compound was tested in triplicate. To prepare calibration solution for compound quantifications,
blank wells were prepared containing only mouse plasma in a donor well and buffer solution in
its acceptor well. The equilibrium dialysis was carried out by rocking the plate for 22 hours in
37 °C. Once equilibrium was reached, the plasma and buffer solution from both wells were
21

528
529
530
531
532
533
534
535
536
carefully removed for further analysis with liquid chromatography-tandem mass spectrometry.
Plasma solution and internal standard were mixed in the presence of 80% acetonitrile. After
centrifuging the solution, the supernatant was transferred to an insert. Similarly, the buffer
solution from the acceptor side was prepared containing 40% acetonitrile. Calibration standards
for donor and acceptor sides were prepared with compound concentrations of 50 nM, 100 nM,
250 nM, 500 nM, and 1µM. The compound concentrations from each well were calculated from
the calibration curves using Microsoft Excel. The percentage of the test compound bound was
determined as follows:
% Free = (Concentration buffer chamber/Concentration plasma chamber) × 100%
537
% Bound = 100% - % Free
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
Metabolic stability. Liver microsome stability assays were done by contract research
laboratory, Wuxi AppTec Co. (Hubei, China). Briefly,compounds at 1 µM concentration were
incubated in singlet with human or CD-1 mouse liver microsomes for 6 time points (0, 5, 10, 20,
30, and 60 min). Loss of parent compound was quantified by liquid chromatography/tandem
mass spectrometry. The measured half-lives for control compounds (testosterone, diclofenac,
and propafenone) were within the expected ranges.
Murine pharmacokinetics studies. Non-fasted female Swiss Webster mice (6-8 weeks) in
groups of thee were administered compounds by oral gavage (10 mg/kg PO) or by retro-orbital
injection (5 mg/kg IV). MetRS inhibitors were dosed in vehicle consisting of 7% Tween 80, 3%
ethanol, and 0.9% saline. Linezolid was dosed in vehicle consisting of 0.5% methylcellulose
(cP400), 0.5% Tween 80, and 0.9% saline. Time points for the blood collections were as
follows. Oral: 30, 60, 120, 240, 360, 480, and 1440 min. IV: 5, 15, 30, 60, 240, 360, 480, and
1440 min. Tail blood was collected in heparinized capillary tubes and 20 uL was spotted onto
Whatman FTA DMPK-C Cards (GE, Faairfield, CT). The whole blood samples were extracted
22

553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
with acetonitrile for measurement of compound concentrations by liquid
chromatography/tandem mass spectrometry. Pharmacokinetic parameters were calculated
using Phoenix WinNonlin Version 6.3 software (Certara, Princeton, NJ).
Murine thigh infection model. Animal studies were approved by the Institutional Animal Care
and Use Committee at the University of Washington, Seattle. The methods were based on
published literature (41-44). Female specific pathogen free CD1 mice were obtained from
Charles River (Wilmington, MA) weighing 23-27 grams and allowed at least 3 days to acclimate
prior to study. Mice had access to food and water ad libitum. Neutropenia was induced by
administering cyclophosphamide monohydrate (Sigma Aldrich C7397; St. Louis, MO) via IP
injection 4 days (at 150 mg/kg) and 1 day (at 100 mg/kg) prior to infection. Neutropenic status
was confirmed by neutrophil count < 100 cells/mm3. Overnight culture of S. aureus (ATCC strain
29213) was diluted 1:100 in MHB and incubated until reaching mid-log phase (OD600 < 0.750).
The inoculum was prepared by pelleting log-phase culture and re-suspending in sterile dPBS.
The culture was adjusted to OD600 of 0.200 and diluted 1:100 in sterile dPBS correlating to an
inoculum of ~5*105 CFU/50 µL. The mice were infected by an intramuscular injection of 50 µL in
the right posterior thigh while under isoflurane gas anesthesia. At 1 h post infection, one vehicle
group was sacrificed for determination of initial inoculum (status level of infection). Mice were
dosed at 2 and 12 h post-infection with test compounds as follows. Vancomycin was given at
100 mg/kg SC in 100 µL in a 0.9% saline solution. Linezolid was administered at 75 mg/kg PO
in 200ul of 0.5% methylcellulose, cP400 (Sigma Aldrich, St. Louis, MO), 0.5% Tween80 (Sigma
Aldrich, St. Louis, MO) in distilled water (45). MetRS inhibitors were administered at 75 mg/kg
PO in 200µL of vehicle containing 60% Phosal 53 MCT (Lipoid, Ludwigshafen Germany), 30%
PEG400 (Sigma Aldrich, St. Louis, MO), and 10% EtOH. Mice were sacrificed at 24 h post-
infection; the thigh muscle was sterilely removed, weighed, homogenized in 5mLs dPBS,
23

578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
serially diluted, plated on tryptic soy agar in duplicate, and incubated overnight at 37ºC.
Colonies were counted to quantify the bacterial load in CFUs per gram of thigh tissue.
Sequence alignments. Global pairwise amino acid sequence alignments were generated with
NCBI alignment tool, CLUSTAL omega (46).
Acknowledgments
The research was supported by a grant from the Comotion Innovation Fund at University of
Washington and by the National Institutes of Health grant AI097177. The authors wish to
acknowledge Jennifer McCullar and Lynn Silver for consulting on the project direction and
experimental design. The authors would like to thank Nicole Duster and Stephen Nakazawa
Hewitt for cloning and purifying the SaMetRS, and Uyen Nguyen for technical assistance with
the microbiology work.
24

594
595
596
597
598
599
600
601
Figure legends.
Figure 1. Chemical structures of MetRS inhibitors
Figure 2. Macromolecular synthesis experiments. Incorporation of radiolabeled precursors into
S. aureus (ATCC 29213) in 30 minute incubations in the presence of established antibiotics or
MetRS inhibitors. The dashed vertical line indicates the MIC.
602
603
604
605
Figure. 3. Efficacy of MetRS inhibitors in neutropenic mouse S. aureus thigh infection model.
Error bars are SEMs. Stasis level was determined from untreated mice sacrificed 1 h post-
infection (p.i.).
25

606
607
TABLE 1. Assay results of representative MetRS inhibitors. The second column shows the IC50
values against recombinant S. aureus MetRS enzyme.
MetRS
S. aureus
(enzyme)
IC50
S. aureus
MSSA
(ATCC
29213)
S. aureus
MSSA
(ATCC
19636)
S. aureus
MRSA
(ATCC
43300)
S. aureus
MRSA
(ATCC
33591)
S. aureus
VISA
(ATCC
700699)
E. faecium
VRE
(ATCC
51559)
E. faecalis
(ATCC
29212)
E. faecium
(ATCC
19434)
Cmpd
Name
(μM)*
MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml) MIC (μg/ml)
MIC (μg/ml) MIC (μg/ml)
1312
1575
1614
1717
1962
2062
2093
2114
2144
VAN
CIP
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
<0.025
>10.0
2.5
10
ND
ND
5
ND
ND
5
ND
ND
10
0.31
1.3
1.3
0.08
5
0.16
0.16
0.63
0.63
0.16
1.3
>10
5
2.5
2.5
0.63
0.04
1.3
0.16
2.5
0.08
ND
0.16
5
0.08
ND
ND
0.31
ND
0.04
1.3
0.16
ND
ND
0.31
ND
0.04
10
0.08
0.04
0.31
0.02
1.3
ND
ND
0.31
ND
0.04
1.3
0.16
2.5
1.3
0.16
0.63
0.02
5
0.31
0.08
0.08
0.02
0.63
5
0.31
0.08
0.16
0.02
>10
5
0.31
ND
0.08
1.3
>10.0
0.16
2.5
0.08
1.3
0.31
1.3
>10
1.3
1.3
2.5
LNZ
608
609
>10.0
2.5
1.3
TABLE 1 (cont.)
S. epi-
S. epi-
S.
P.
dermidis
(ATCC
49134)
dermidis
(ATCC
12228)
S. pyogenes
(ATCC
19615)
MIC (μg/ml)
pneumoniae
(ATCC
49619) MIC
(μg/ml)
E. coli
(ATCC
25922)
aeruginosa
(ATCC
Mamm. cells
Mamm. cells
(HepG2)
Cmpd
Name
27853) MIC (CRL8155)
(μg/ml) CC50 (μg/ml) CC50 (μg/ml)
MIC (μg/ml)
MIC (μg/ml)
MIC (μg/ml)
1312
1575
1614
1717
1962
2062
2093
2114
2144
VAN
CIP
ND
ND
ND
ND
ND
>10
>10
2.5
ND
>10
>10
>20
ND
ND
>8
12
>8
>18
>18
10
>10
>10
>10
ND
>10
10
5
>10
>10
ND
ND
ND
ND
ND
ND
0.31
ND
15
0.31
ND
0.16
ND
5.8
ND
ND
1.3
>20
>21
>22
>21
>11
>145
>33
>34
>20
>21
>22
>21
>11
>145
>33
>34
1.3
ND
ND
ND
1.3
0.63
1.3
>20
ND
ND
1.3
ND
1.3
ND
0.31
2.5
0.16
2.5
>20
0.31
1.3
ND
0.63
0.63
1.3
ND
0.31
0.63
0.16
0.63
0.04
ND
LNZ
610
611
612
613
614
1.3
*Lower limit of detection for assay is 0.025 M. (All the lis te d inhibitors ha ve low- or subnanomolar
activity on the SaMetRS enzyme). Abbreviations: CIP (ciprofloxacin), LNZ (linezolid), MSSA (methicillin
sensitive Staphylococcus aureus), MRSA (methicillin resistant Staphylococcus aureus), ND (not done),
VAN (vancomycin), VISA (vancomycin intermediate Staphylococcus aureus).
615
26

616
617
TABLE 2. MICs (µg/mL) of selected MetRS inhibitors against E. coli strains with increased
permeability.
Compound
1717
2093
2144
CIP
MB4827 (wild-type)
MB4902 (lpxC)
MB5747 (tolC)
MB5746 (lpxC, tolC)
>20
>20
>20
0.016
160
>20
>20
>20
0.008
20
>20
>20
>20
0.008
0.5
>20
>20
>20
0.008
0.5
NOV
618
619
TABLE 3. MICs and MBCs against S. aureus (ATCC 29213)
Molecule
name
MIC
MBC
MBC/
Published
mechanism (48)
(µg/mL) (µg/mL) MIC ratio
1717
2093
2144
LNZ
VAN
NOV
CIP
0.16
0.06
0.010
2.5
5
1.3
0.31
160
5
32
16
32
64
4
Static
Cidal
Static
Cidal
1.3
0.16
0.31
5
32
4
1.3
620
621
622
TABLE 4. Frequency of spontaneous resistance for S. aureus (ATCC 29213) (two independent
experiments and average)
Resistance
Resistance
frequency
(Expt. 1)
Resistance
frequency
(Expt. 2)
frequency
(average or
greater #)
Compound
(fold over MIC)
NOV (4)
NOV (8)
CIP (4)
ND
1.0 x 10-7
1.6 x 10-8
7.6 x 10-9
<1.8 x 10-10
<1.8 x 10-10
<1.8 x 10-10
2.6 x 10-8
5.43 x 10-9
1.5 x 10-7
2.6 x 10-8
8.1 x 10-8
2.45 x 10-8
1.0 x 10-7
1.6 x 10-8
8.9 x 10-9
<2.6 x 10-10
<2.6 x 10-10
<2.6 x 10-10
1.6 x 10-8
4.16 x 10-9
9.8 x 10-8
2.1 x 10-8
5.0 x 10-8
1.38 x 10-8
ND
1.0 x 10-8
<2.6 x 10-10
<2.6 x 10-10
<2.6 x 10-10
5.3 x 10-9
2.89 x 10-9
5.0 x 10-8
1.6 x 10-8
2.0 x 10-8
3.15 x 10-9
CIP (8)
LNZ (4)
LNZ (8)
1717 (4)
1717 (8)
2093 (4)
2093 (8)
2144 (4)
2144 (8)
623
624
27

625
626
627
TABLE 5. MIC values for S. aureus (ATCC 29213) grown in MHB or MHB with 50% human
serum. The table also shows experimental or published values for plasma protein binding
(reference in parentheses).
Median MIC
(no serum)
µg/mL
Median MIC
(+50% serum)
µg/mL
Fold
change of
MIC
% binding to
mouse
plasma
% binding to
human
plasma
#
expts
3
Compounds
1
2
2
VAN
25 (49)
ND
55*
95*
3
3
3
3
3
3
3
3
0.13
2
16
2
128
1
NOV
LNZ
39.5
ND
31*
0.13
2
0.25
32
1
2
CIP
20-40*
ND
16
16
64
128
64
1614
1717
2069
2093
2144
96.5
97.6
99.6
99.9
98.4
0.06
0.25
0.06
0.06
95.4
ND
16
8
ND
4
ND
628
629
630
*Data from Drugbank (www.drugbank.ca)
631
TABLE 6. Liver microsome stability half-lives.
Microsome
stability:
Microsome
stability:
Molecule
Name
Human T1/2
(minutes)
Mouse T1/2
(minutes)
1614
9.8
6.6
10.3
27.0
8.4
1717
1962
10.3
25.9
12.6
8.3
2062
2093
3.2
2114
15.2
16.4
>145
17.4
10.7
7.1
7.7
2144
6.5
Linezolid
Testosterone
Diclofenac
Propafenone
>145
3.1
82.8
1.7
632
633
28

634
TABLE 7. Inhibitor binding site amino acid residues for MetRS enzymes
Seq. No. a
Pocket b
T.brucei
247 248 249 250 287 289 290 291 292 456 460 461
b
b
b
l
q
q
q
q
q
q
q
q
Pro Ile
Pro Ile
Pro Ile
Tyr Tyr Asp His Gly Gln Lys Leu Ala Ile
S. aureus
Tyr Tyr Asp His Gly Gln Lys
Phe Tyr Asp His Gly Leu Lys
¶
¶
Gly Val/Ile*
Gly Ile
H.sapiens -
mitochondrial
635
Seq. No. a
Pocket b
T.brucei
470 471 472 473 474 476 477 478 480 481 519 522 523
q
q
q
q
b
q
b
b
b
b
b
b
b
Cys Val Tyr Val Trp Asp Ala Leu Asn Tyr Ile
Val Val Tyr Val Trp Asp Ala Leu Asn Tyr Ile
Phe His
Phe His
Phe His
S. aureus
H.sapiens/mito Thr Ile
Tyr Val Trp Asp Ala Leu Asn Tyr Ile
636
637
638
a Sequence numbers refer to the T.brucei sequence; b l = linker zone, b = benzyl pocket, q = quinolone
pocket (13). * This residue is Val in all S.aureus sequences in UNIPROT except for A0A033UAT9 (strain
C0673), where it is an Ile. ¶ Due to a different loop length it is unclear what the equivalent residue is.
639
640
29

641
642
References
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
1.
2.
Infectious Diseases Society of A. 2010. The 10 x '20 Initiative: pursuing a global commitment to
develop 10 new antibacterial drugs by 2020. Clin Infect Dis 50:1081-3.
Nakama T, Nureki O, Yokoyama S. 2001. Structural basis for the recognition of isoleucyl-
adenylate and an antibiotic, mupirocin, by isoleucyl-tRNA synthetase. J Biol Chem 276:47387-
47393.
3.
O'Dwyer K, Spivak AT, Ingraham K, Min S, Holmes DJ, Jakielaszek C, Rittenhouse S, Kwan AL, Livi
GP, Sathe G, Thomas E, Van HS, Miller LA, Twynholm M, Tomayko J, Dalessandro M, Caltabiano
M, Scangarella-Oman NE, Brown JR. 2015. Bacterial resistance to leucyl-tRNA synthetase
inhibitor GSK2251052 develops during treatment of complicated urinary tract infections.
Antimicrob Agents Chemother 59:289-298.
Jarvest RL, Berge JM, Berry V, Boyd HF, Brown MJ, Elder JS, Forrest AK, Fosberry AP, Gentry DR,
Hibbs MJ, Jaworski DD, O'Hanlon PJ, Pope AJ, Rittenhouse S, Sheppard RJ, Slater-Radosti C,
Worby A. 2002. Nanomolar inhibitors of Staphylococcus aureus methionyl tRNA synthetase with
potent antibacterial activity against gram-positive pathogens. J Med Chem 45:1959-1962.
Jarvest RL, Berge JM, Brown P, Houge-Frydrych CS, O'Hanlon PJ, McNair DJ, Pope AJ,
Rittenhouse S. 2003. Conformational restriction of methionyl tRNA synthetase inhibitors leading
to analogues with potent inhibition and excellent gram-positive antibacterial activity. Bioorg
Med Chem Lett 13:1265-1268.
Jarvest RL, Berge JM, Brown MJ, Brown P, Elder JS, Forrest AK, Houge-Frydrych CS, O'Hanlon PJ,
McNair DJ, Rittenhouse S, Sheppard RJ. 2003. Optimisation of aryl substitution leading to potent
methionyl tRNA synthetase inhibitors with excellent gram-positive antibacterial activity. Bioorg
Med Chem Lett 13:665-668.
Jarvest RL, Armstrong SA, Berge JM, Brown P, Elder JS, Brown MJ, Copley RC, Forrest AK,
Hamprecht DW, O'Hanlon PJ, Mitchell DJ, Rittenhouse S, Witty DR. 2004. Definition of the
heterocyclic pharmacophore of bacterial methionyl tRNA synthetase inhibitors: potent
antibacterially active non-quinolone analogues. Bioorg Med Chem Lett 14:3937-3941.
Critchley IA, Ochsner UA. 2008. Recent advances in the preclinical evaluation of the topical
antibacterial agent REP8839. Curr Opin Chem Biol 12:409-417.
Critchley IA, Green LS, Young CL, Bullard JM, Evans RJ, Price M, Jarvis TC, Guiles JW, Janjic N,
Ochsner UA. 2009. Spectrum of activity and mode of action of REP3123, a new antibiotic to treat
Clostridium difficile infections. J Antimicrob Chemother 63:954-963.
Shibata S, Gillespie JR, Kelley AM, Napuli AJ, Zhang Z, Kovzun KV, Pefley RM, Lam J, Zucker FH,
Van Voorhis WC, Merritt EA, Hol WG, Verlinde CL, Fan E, Buckner FS. 2011. Selective Inhibitors
of Methionyl-tRNA Synthetase Have Potent Activity against Trypanosoma brucei Infection in
Mice. Antimicrob Agents Chemother 55:1982-1989.
Shibata S, Gillespie JR, Ranade RM, Koh CY, Kim JE, Laydbak JU, Zucker FH, Hol WG, Verlinde CL,
Buckner FS, Fan E. 2012. Urea-based inhibitors of Trypanosoma brucei methionyl-tRNA
synthetase: selectivity and in vivo characterization. J Med Chem 55:6342-6351.
Ranade RM, Zhang Z, Gillespie JR, Shibata S, Verlinde CL, Hol WG, Fan E, Buckner FS. 2015.
Inhibitors of Methionyl-tRNA Synthetase Have Potent Activity against Giardia intestinalis
Trophozoites. Antimicrob Agents Chemother 59:7128-7131.
Huang W, Zhang Z, Barros-Alvarez X, Koh CY, Ranade RM, Gillespie JR, Creason SA, Shibata S,
Verlinde CL, Hol WG, Buckner FS, Fan E. 2016. Structure-guided design of novel Trypanosoma
brucei Methionyl-tRNA synthetase inhibitors. Eur J Med Chem 124:1081-1092.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
30