A novel 6-benzyl ether benzoxaborole is active against mycobacterium tuberculosis in vitro

Article abstract of DOI:10.1128/AAC.01205-17

We identified a novel 6-benzyl ether benzoxaborole with potent activity against Mycobacterium tuberculosis. The compound had an MIC of 2 μM in liquid medium. The compound was also able to prevent growth on solid medium at 0.8 μM and was active against intracellular bacteria (50% inhibitory concentration [IC₅₀] 3.6 μM) without cytotoxicity against eukaryotic cells (IC₅₀ 100 μM). We isolated resistant mutants (MIC ≥ 100 μM), which had mutations in Rv1683, Rv3068c, and Rv0047c.

Full text of DOI:10.1128/AAC.01205-17

AAC Accepted Manuscript Posted Online 3 July 2017
Antimicrob. Agents Chemother. doi:10.1128/AAC.01205-17
Copyright © 2017 Patel et al.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license.
1
2
A novel 6-benzyl ether benzoxaborole is active against Mycobacterium tuberculosis in
vitro
3
4
Authors:
5
6
7
Nipul Patel1, Theresa O’Malley1, Yong-Kang Zhang2, Yi Xia2, Bjorn Sunde1, Lindsay
Flint1, Aaron Korkegian1, Thomas R. Ioerger3, Jim Sacchettini3, M. R. K. Alley2, Tanya
Parish1*
8
9
10 1 TB Discovery Research, Infectious Disease Research Institute, Seattle, WA, USA
11 2 Anacor Pharmaceuticals, Palo Alto, CA, USA
12 3 Texas A&M University, College Station, TX, USA
13
14
15
16 *Corresponding author:
17 tanya.parish@idri.org
18 Tel +1 206 858 6074
19 Fax +1 206 361 3678
20

21 Abstract
22 We identified a novel 6-benzyl ether benzoxaborole with potent activity against
23 Mycobacterium tuberculosis. The compound had a minimum inhibitory concentration of
24 2 µM in liquid medium. The compound was also able to prevent growth on solid medium
25 at 0.8 µM and was active against intracellular bacteria (IC50 = 3.6 µM) without
26 cytotoxicity against eukaryotic cells (IC50 >100 µM). We isolated resistant mutants (MIC
27 100 µM), which had mutations in Rv1683, Rv3068c and Rv0047c.
28

29
30
Tuberculosis (TB) remains a serious global health problem, with an increase in
31 the reported incidence of new infections combined with increasing levels of drug
32 resistance (1). We are interested in finding both new molecules with anti-tubercular
33 activity and also to determine the mode of resistance to new agents, and/or their
34 molecular targets.
35
In screening the Anacor boron library we identified a member of the 6-
36 benzylether benzoxaborole class, 6-(benzyloxy)-4,7-dimethylbenzo[c][1,2]oxaborol-
37 1(3H)-ol (Figure 1 and Supplementary Information), with good in vitro activity against
38 Mycobacterium tuberculosis under aerobic conditions. Briefly, we tested the compound
39 in DMSO as two-fold serial dilutions against M. tuberculosis H37Rv (ATCC 25618) for 5
40 days in Middlebrook 7H9 medium supplemented with 10% OADC (oleic acid, albumin,
41 dextrose, catalase) and 0.05% w/v Tween-80. Growth was monitored by OD590; the
42 minimum inhibitory concentration (MIC) was determined by fitting the growth inhibition
43 curve using the Levenberg Marquardt algorithm . MIC was defined as the concentration
44 required to inhibit growth by 90% (2). The compound had an MIC of 2.0  0.24 µM
45 (n=6).
46
The cytotoxicity of the compound was determined in HepG2 cells were cultured
47 in DMEM, 10% fetal bovine serum (FBS), and 1X penicillin-streptomycin solution (100
48 U/mL). Cells were exposed to compounds for 2 days at 37°C, 5% CO2 (final DMSO
49 concentration of 1%). Cell viability was measured using CellTiter-Glo® Reagent
50 (Promega) and determining relative luminescent units (RLU). Inhibition curves were
51 fitted using the Levenberg–Marquardt algorithm and used to calculate IC50 as the

52 concentration required to reduce cell viability by 50%. We tested the compound using
53 either glucose or galactose as the carbon source and the IC50 was >100 µM (n=2) under
54 both conditions. Therefore, we tested the compound for activity against intracellular
55 bacteria using a luminescent strain of M. tuberculosis (3). THP-1 cells were infected
56 overnight with M. tuberculosis at an MOI of 1 in complete RPMI (RPMI-1640, 10% FBS,
57 2 mM Corning glutagro, 1 mM sodium pyruvate). Extracellular bacteria were removed
58 by washing and infected cells were seeded at 4 x 104 cells per well in 96-well plates
59 containing compounds. Compounds were tested as a 10-point, 3-fold dilution series
60 (0.5% DMSO). Infected cells were incubated for 3 days in a humidified atmosphere of
61 37°C, 5% CO2. Relative luminescent units (RLU) were used as a measure of bacterial
62 viability. Growth inhibition curves were fitted using the Levenberg–Marquardt algorithm;
63 the IC50 and IC90 were defined as the compound concentrations that produced 50% and
64 90% inhibition of intracellular growth respectively. The IC50 and IC90 were 3.6  0.07 and
65 22  12 µM respectively (n=2).
66
We tested the ability of the compound to prevent growth on solid medium. We
67 plated aerobically-cultured M. tuberculosis onto Middlebrook 7H10 plus 10% OADC
68 containing compounds (4). Plates were incubated for 3-4 weeks at 37°C and growth
69 recorded. The MIC99 under these conditions was 5 µM; we plated M. tuberculosis
70 H37Rv onto solid medium containing 5X or 10X the MIC and isolated colonies isolated
71 after 3-6 weeks. Clones were tested for resistance in liquid and solid medium. Four
72 isolates with high-level resistance were confirmed with an MIC 100 µM. DNA isolated
73 from these mutants was subjected to whole genome sequencing (5). Several single

74 nucleotide polymorphisms were identified (Table 1) and confirmed by PCR amplification
75 and sequencing.
76
Two of the four strains had mutations in Rv1683, while the other two had
77 mutations in Rv0047c and Rv3068c. The mutations in Rv0047c would result in a
78 premature stop codon, while the mutations in Rv3068c would result in a threonine to
79 alanine change. The Rv0047c gene is located upstream of ino1, which is involved in
80 phosphatidylinositol metabolism and is required for growth on inositol (6). Rv0047c is
81 proposed to be co-transcribed with ino1 suggesting a link with inositol metabolism.
82 Therefore, we determined if additional of inositol had any effect on the compound
83 activity, but we saw no shift in MIC (range 5.4-5.9 M with 6.25-100 M inositol). We
84 also tested L-histidine supplementation, but saw no difference (range 3.2-3.8 M with
85 10-100 M inositol). Since the mutation in Rv0047c was linked to a mutation in
86 Rv3068c in both strains with the same nonsynonymous substitution it is possible the
87 two strains are siblings. The Rv3068c gene encodes a non-essential enzyme, PgmA, a
88 putative phosphoglucomutase involved in glucose metabolism.
89
Rv1638 encodes a possible bifunctional protein involved in catabolism and
90 anabolism of triglycerides (7). In Mycobacterium bovis, BCG1721 (homolog of Rv1683)
91 is responsible for accumulation and breakdown of triglycerides (TGs) stored as lipid
92 droplets (LDs) (7). Several studies have shown TGs as carbon source utilized by M.
93 tuberculosis in non-replicating persistence phase (8) and the buildup of TGs has been
94 correlated with drug tolerance (9). It is not clear if the mutations we see would affect the
95 enzymatic activity of the protein, or if the mutations might be in an enzyme binding site.
96 However, it is of note that Rv1683 is one of three esterases active in the normoxia,

97 hypoxia and resuscitation phases of growth underlining its importance (10). Future work
98 should help to elucidate if one of these is the true target, or if there are physiological
99 changes which result in resistance.
100
In summary, we have identified a novel compound with efficacy against M.
101 tuberculosis in both solid and liquid medium, as well as active against intracellular
102 bacteria, but with no cytotoxicity, thus the profile of this compound is encouraging for
103 future development. We have identified two routes to resistance to this compound in
104 Rv1683, or Rv0047c and Rv3068c.
105
We thank James Ahn, Dean Thompson, James Johnson, Douglas Joerss,
106 Catherine Shelton, Lina Castro and Yulia Ovechkina for technical assistance.
107
108 Funding
109 This research was supported with funding from the Bill and Melinda Gates Foundation
110 and by NIAID of the National Institutes of Health under award number R01AI099188.
111 The content is solely the responsibility of the authors and does not necessarily
112 represent the official views of the National Institutes of Health.

113 References
114 1. WHO.
115 http://www.who.int/tb/publications/global_report/en/
Global
tuberculosis
report
2016.
116 2. Ollinger J, Bailey MA, Moraski GC, Casey A, Florio S, Alling T, Miller MJ, Parish T.
117 2013. A dual read-out assay to evlaute the potency of compounds active against
118 Mycobacterium tuberculosis. PLOS One. 8: e60531.
119 3. Andreu N, Zelmer A, Fletcher T, Elkington PT, Ward TH, Ripoll J, Parish T, Bancroft
120 GJ, Schaible U, Robertson DR, Wiles S. 2010. Optimisation of bioluminescent
121 reporters for use with mycobacteria. PLOS One, 5: e10777.
122 4. Sirgel FA, Wild IJ, van Helden PD.
2009. Measuring minimum inhibitory
123 concentrations in mycobacteria. Methods Mol Biol, 465:173-86
124 5. Ioerger TR, O'Malley T, Liao R, Hickey MJ, Mohaideen N, Murphy KC, Boshoff GI,
125 Mizrahi V, Rubin EJ, Sassetti CM, Barry CE, Sherman DR, Parish T, Sacchettini JC.
126 2013. Identification of new drug targets and resistance mechanisms in Mycobacterium
127 tuberculosis. PLOS One 8: e75245
128 6. Movahedzaedh F, Smith DA, Norman RA, Dinadayala P, Murray-Rust J, Russell DG,
129 Kendall SL, Rison SC, McAlister MS, Bancroft GJ, McDonald NQ, Daffe M, Av-Gay Y,
130 Stoker NG. 2004. The Mycobacterium tuberculosis ino1 gene is essential for growth and
131 virulence. Mol Microbiol. 51: 2003-1014.
132 7. Low KL, Shui G, Natter K. 2010. Lipid droplet-associated proteins are involved in the
133 biosynthesis and hydrolysis of triacylglycerol in Mycobacterium bovis bacillus Calmette-
134 Guérin. J Biol Chem. 285: 21662-21670.

135 8. Russell, D.G. 2003. Phagosomes, fatty acids and tuberculosis. Nat Cell Biol. 5: 776-
136 778.
137 9. Deb C, Lee CM, Dubey VS, Daniel J, Abomoelak B, Sirakova TD, Pawar S, Rogers
138 L, Kolattukudy PE. 2009. A novel in vitro multiple-stress dormancy model for
139 Mycobacterium tuberculosis generates a lipid-loaded, drug-tolerant, dormant pathogen.
140 PLOS One. 4: e6077.
141 10. Tallman KR, Levine SR, Beatty KE. 2016. Small-molecule probes reveal esterases
142 with persistent activity in dormant and reactivating Mycobacterium tuberculosis. ACS
143 Infect Dis. 2: 936-944.
144
145
146
147

148
Mutant isolate
MIC99 (µM)
Rv0047c
Rv3068c
Rv1683
RM1
RM2
RM3
RM4
100
>100
>100
100
wt
wt
L341P
E128*
wt
T351A
wt
wt
M200I A201T
wt
E128*
T351A
149
150 Table 1. Profile of resistant mutants
151 Resistant mutants were isolated on solid medium. MIC99 was calculated on solid
152 medium (4). The SNPs listed in the table were identified by whole genome sequencing
153 and confirmatory PCR/sequence in each strain.
154

155 (A)
156
157 (B)
158
159
160 Figure 1. (A) Structure of 6-benzyl ether (B) Synthetic pathway for compound: a)
161 chloromethyl ethyl ether, DIPEA, DCM, rt, overnight ; b) n-butyl lithium, DMF, THF,
162 18oC, 1.5 h; c) HCl, THF, rt, overnight; d) sodium cyanoborohydride, THF, rt, 3 h; e)
163 Phosphorus oxychloride, DMF, rt, overnight; f) benzyl bromide, NaHCO3, KI, AcCN,
164 80oC, overnight; g) Triflic anhydride, triethylamine, DCM, rt, 3 H; h) 5,5,5',5'-tetramethyl-
165 2,2'-bi(1,3,2-dioxaborinane), PdCl2(dppf)2, potassium acetate, 1,4-dioxane, 90oC,
166 overnight; j) sodium borohydride, THF, rt, 3 h, and then HCl, water, overnight.
167
168