Discovery and in vivo evaluation of alcohol-containing benzothiazoles as potent dual-targeting bacterial DNA supercoiling inhibitors

Article abstract of DOI:10.1016/j.bmcl.2014.07.037

A series of dual-targeting, alcohol-containing benzothiazoles has been identified with superior antibacterial activity and drug-like properties. Early lead benzothiazoles containing carboxylic acid moieties showed efficacy in a well-established in vivo model, but inferior drug-like properties demanded modifications of functionality capable of demonstrating superior efficacy. Eliminating the acid group in favor of hydrophilic alcohol moieties at C ⁵, as well as incorporating solubilizing groups at the C⁷ position of the core ring provided potent, broad-spectrum Gram-positive antibacterial activity, lower protein binding, and markedly improved efficacy in vivo.

Full text of DOI:10.1016/j.bmcl.2014.07.037

Bioorganic & Medicinal Chemistry Letters 24 (2014) 4215–4222
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery and in vivo evaluation of alcohol-containing
benzothiazoles as potent dual-targeting bacterial DNA
supercoiling inhibitors
a
b
b
a
James T. Palmer a, , Lorraine C. Axford , Stephanie Barker , James M. Bennett , Michael Blair ,
Ian Collins b, David T. Davies b, Leigh Ford a, Carlie T. Gannon a, Paul Lancett b, Alastair Logan b,
Christopher J. Lunniss a, Craig J. Morton a, Daniel A. Offermann a, Gary R. W. Pitt a, B. Narasinga Rao c,
Amit K. Singh c, Tarun Shukla c, Anil Srivastava c, Neil R. Stokes b, Helena B. Thomaides-Brears b,
Anju Yadav c, David J. Haydon b
⇑
a Biota Scientific Management, 10/585 Blackburn Road, Notting Hill, Victoria 3168, Australia
b Biota Europe Ltd, Begbroke Science Park, Begbroke, Oxfordshire OX51PF, United Kingdom
c Jubliant Chemsys Ltd, B-34, Sector 58, Noida 201301, India
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of dual-targeting, alcohol-containing benzothiazoles has been identified with superior antibac-
terial activity and drug-like properties. Early lead benzothiazoles containing carboxylic acid moieties
showed efficacy in a well-established in vivo model, but inferior drug-like properties demanded modifi-
cations of functionality capable of demonstrating superior efficacy. Eliminating the acid group in favor of
hydrophilic alcohol moieties at C5, as well as incorporating solubilizing groups at the C7 position of the
core ring provided potent, broad-spectrum Gram-positive antibacterial activity, lower protein binding,
and markedly improved efficacy in vivo.
Received 16 May 2014
Revised 10 July 2014
Accepted 14 July 2014
Available online 19 July 2014
Keywords:
DNA gyrase
Topoisomerase
Antibacterial
Synthesis
Ó 2014 Elsevier Ltd. All rights reserved.
In vivo
Dual-targeting
Increasing emergence of drug resistance in pathogenic bacteria
has resulted in higher human mortality and increased healthcare
costs.1,2 We therefore targeted a new class of agents with no
cross-resistance to existing therapies. Recently, we reported the
discovery of a new class of dual-targeting, benzothiazole-contain-
ing small molecule inhibitors of bacterial DNA gyrase and topoiso-
merase IV.3 Bacterial type II topoisomerases form heterotetramers
consisting of two subunits: DNA gyrase forms an A2B2 complex
comprised of GyrA and GyrB, and topoisomerase IV forms a C2E2
complex made up of ParC and ParE monomers.3 In addition, the
GyrB and ParE subunits are functionally similar with an ATPase
active site within their N-terminal domains. The C-terminal
domains of GyrB and ParE, by contrast, interact with the GyrA
and ParC subunits of the heterotetramers, and also form the DNA
binding grooves necessary for supercoiling and decatenation.4
Highly conserved among bacterial species, these enzymes are ideal
for broad spectrum inhibitor approaches. Fluoroquinolones inter-
act with the DNA binding grooves of the GyrA and ParC subunits,
but the similarity between the ATP binding regions of GyrB and
ParE suggests a motif for simultaneous targeting.5 GyrB/ParE inhib-
itors offer the advantage of no pre-existing resistance and a low
propensity for the development of resistance.
A number of groups have recently described ATP site-binding
GyrB/ParE inhibitor series.6–10 We recently reported the discovery
of isonipecotic acid-containing benzothiazole ureas with enhanced
solubility and pharmacokinetic properties,11,12 along with microbio-
logical assessment of well-characterized examples.13 To establish the
potential for further development of this chemotype, we evaluated
compounds in a neutropenic mouse model of MSSA Staphylococcus
aureus infection.14 In this acute model, which was not designed to
establish PK/PD parameters, compounds showed a dose-dependent
drop in colony-forming units (CFUs) versus controls (Fig. 1).
Compounds 1 and 2 (MIC versus MSSA type strain S. aureus ATC
29213 of 0.06 and 0.12 lg/mL, respectively) were dosed
intravenously at 2, 4, and 6 h after inoculation. Assessment of bac-
terial load 24 h post infection showed the compounds to have been
efficacious. Although the lower doses of 3 Â 30 mg/kg afforded
⇑
Corresponding author. Tel.: +61 3 9915 3700; fax: +61 3 9915 3701.
E-mail address: j.palmer@biota.com.au (J.T. Palmer).
http://dx.doi.org/10.1016/j.bmcl.2014.07.037
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

4216
J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
Figure 1. Efficacy of isonipecotic acids in neutropenic mouse thigh model.
Eliminate carboxylate
Install alcohol
HO
HO2C
R1 R2
N
N
N
N
HO
O
O
O
N
N
H
N
S
N
N
H
N
N
S
N
H
N
S
N
H
N
H
N
H
N
N
N
Additional property-
enhancing groups
R3
3
1
Figure 2. Evolution of alcohol-containing benzothiazole ureas.
O
M = Li or Mg
N
N
n-BuLi/ PhMe
N
N
N
Br
Br
M
Br
I
Br
or i-PrMgCl/
Et2O/THF
HO
HO
52%
O
N
6
7
5
N
N
H
55%
i) IBCF/NMM/THF, -10 oC
O
N
N
Br
Br
ii) NaBH4/H2O, 0 o
30%
C
HO
HO
N
N
8
2
1
R
R
OH
B
O
N
N
HO
O
N
S
HO
H
N
N
N
H
N
S
H
6 / 7 / 8
N
H
Pd(dppf)Cl2.DCM,
N
Cs2CO3, aq. dioxane,
R1, R2 = CH3, 3
R1 = H, R2 = CH3, 9
R1, R2 = H, 10
N
85 oC, 50-75%
4
Scheme 1. Synthesis of alcohol-containing C5-substituted benzothiazoles.
modest reductions in CFUs (<3 logs) the higher doses of
3 Â 100 mg/kg demonstrated greater than 4 log drops in CFUs,
comparable to a single dose of vancomycin at 30 mg/kg. While cal-
culating CFU reductions from an early time point is commonly
used, it is not improper to calculate reductions from the matched
control time point; when a compound has a static or slow cidal
mode of action, calculating reductions in CFUs to the matched con-
trols is more informative.6,14
focused on reducing plasma protein binding, while retaining solu-
bility sufficient to permit formulation within acceptable pH ranges
for hospital-based intravenous administration. Lastly, we required
demonstration of efficacy by both intravenous and oral routes, lay-
ing the groundwork for both inpatient and outpatient treatments.
To this end, we eliminated the carboxylic acid functionality in
favor of an alcohol-substituted pyrimidine (R1R2COH) attached to
the C5 position of the benzothiazole (Fig. 2). This moiety provided
an alternative hydrophilic, yet neutral group suitable for prodrug-
ging as needed. At the same time, we recognized a further
Armed with these encouraging proof-of-concept data, we sought
to improve beneficial physical properties further. In particular, we

J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
4217
Table 1
Antibacterial (MIC) activity of alcohol-containing benzothiazole compounds
O
C5
N
H
N
S
N
H
S. aureus
S. pyogenes
H. influenzae
49247a
S. aureus T173 N
S. pyogenes A53S
(ParE)a,b
S. aureus 29213 + 50% horse
Entry
29213a
51339a
(GyrB)a,b
seruma (shift)
N
C5
OH
N
O
0.06
0.06
0.06
0.12
4
1
0.25
0.25
0.5
1
4 (64Â)
N
1
3
N
N
N
0.25 (4Â)
OH
N
0.03
0.03
0.06
0.06
0.5
0.25
0.25
1
0.12 (4Â)
0.06 (2Â)
9
HO
HO
N
N
0.25
0.5
10
N
a
l
g/mL.
b
See Refs. 11,13 for methodology.
R2
R3
N
HO
O
R2
4
i) n-BuLi/ PhMe
N
R3
N
N
H
N
N
Br
N
I
Br
O
ii)
H
S
HO
N
Pd(dppf)Cl2.DCM,
N
R2
R3
Cs2CO3, aq. dioxane,
85 oC, 50-75%
N
11-20
Scheme 2. Preparation of tertiary alcohol-containing benzothiazole ureas.
Table 2
Biochemical, antibacterial activity of tertiary alcohol-based benzothiazole ureas
R2
R3
N
HO
O
N
N
H
N
S
N
H
S. aureus GyrB S. aureus
S. pyogenes
H. influenza S. aureus
S. pyogenes
S. aureus 29213 + 50%
Entry
ATPasea
29213a
51339a
49247a
T173 N (GyrB)a A53S (ParE)a
horse seruma
N
R2, R3
(CH3)2
0.006
0.017
0.005
0.054
0.005
0.005
0.006
0.011
0.010
0.029
0.006
0.06
0.25
0.03
8
0.12
0.5
0.12
8
1
8
0.25
1
1
8
0.25
4
3
11
12
13
14
15
16
17
18
19
20
(CH3CH2)2
–(CH2)4–
–(CH2)5–
1
0.25
2
0.5
16
> 16
1
> 16
> 16
2
–(CH2CH2OCH2CH2)–
–(CH2CH2SCH2CH2)–
0.06
0.12
0.25
1
0.12
0.5
0.25
1
0.25
0.25
1
0.12
1
2
4
–(CH2CH2SO2CH2CH2)–
–(CH2CH2N(CH3)CH2CH2)–
–(CH2CH2NHCH2CH2)–
–(CH2NHCH2)–
1
2
0.5
0.5
4
8
8
8
8
2
> 16
> 16
0.5
> 16
> 16
0.5
> 16
> 16
2
> 16
0.06
8
> 16
0.12
–(CH2OCH2)–
0.12
a
lg/mL.
opportunity to reduce protein binding by substituting on the C7
heterocycle (R3).
Preparation of 3, and secondary and primary alcohol analogs 9
and 10 respectively, utilized our established core synthesis method
to couple boronic acid 43 with an appropriately functionalized
heterocycle. Thus, as depicted in Scheme 1, metalation of 2-iodo-
5-bromopyridine gave 5, followed by acetone quenching, delivered
the tertiary alcohol 6. Palladium-catalyzed cross-coupling of 6 with
4 yielded 3 in good yields. Likewise, quenching of the anionic 5
with acetaldehyde delivered secondary alcohol 7. Sodium

4218
J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
R2
of target pathogens, compared directly with carboxylic acid 1,
(Table 1).
Notably, the antibacterial activity remained consistent irrespec-
N
I
HO
O
R2
i) n-BuLi/ PhMe
N
N
N
4
N
H
N
S
N
N
N
Br
N
H
tive of the distal C5 substituent, whether acidic or alcoholic. Mod-
eling of the series within the active site of the Escherichia coli
enzymes suggested neither moiety would interfere with binding
in the ATPase pockets of either GyrB or ParE.15 The characterization
of first-step mutants (S. aureus GyrB T173 N and Streptococcus
pyogenes ParE A53S) indicated that the primary target is GyrB in
O
ii)
HO
Pd(dppf)Cl2.DCM,
Cs2CO3, aq. dioxane,
85 oC, 50-75%
R2
H
21-26
Br
N
N
4
H
O
N
OH
N
O
B
I
O
O
NBS, NaOH
O
N
N
N
N
N
N
Pd(dppf)Cl2.DCM,
Cs2CO3, aq. dioxane,
85 oC, 79%
µwave,
120 o
aq. dioxane, 12%
C
S. aureus and ParE in S. pyogenes.13 The retention of sub-
lg/mL
MIC values against these strains indicate that the compounds also
inhibit the secondary targets, ParE and GyrB respectively, reflecting
potential for dual targeting. Activity (4 lg/mL) against a double
Br
Br
Br
Br
27
28
29
EtO
P
O
HO
N
Br
N
EtO
4
O
i) MsCl, Et3N
ii) LiBr, THF
(EtO)3P, MeCN
µwave, 100 o
N
N
Pd(dppf)Cl2.DCM,
N
N
C
Cs2CO3, aq. dioxane,
mutant of S. aureus 29213 (GyrB T173N, ParE T167N) for 3 and 9
further illustrate the dual targeting potential of representative
compounds. In addition, the serum shift observed when com-
pounds were assessed for anti-staphylococcal activity in the pres-
ence of 50% horse serum was greatly diminished, illustrating the
improvement in protein binding properties.
To explore the scope and limitations of the alcohol-containing
substituent, we prepared an additional series of C5 alcohols. Lithi-
ation of 2-iodo-5-bromopyrimidine, followed by quenching by the
relevant ketone, afforded tertiary alcohol-containing C5 moieties
suitable for Pd-mediated cross-coupling with boronic acid 4
(Scheme 2).
We observed subtle steric constraints upon assaying this series
against the primary pathogen panel (Table 2). Consistently, the ser-
ies proved to show good on-target potency against DNA gyrase
ATPase, as measured in a malachite green assay format,16 with
all examples tested displaying double digit ng/mL potency or
better.
85 oC, 50-75%
Br
Br
30
8
Br
31
R4
N
N
(EtO) P(O)O
2
O
O
N
N
N
N
N
N
S
NaOMe, R4CHO
THF
H
H
N
H
N
H
S
N
N
N
trans olefins
32
OH
OH
N
R4
O
OH N
O
N
N
N
S
OsO4, NMO
OH N
H
N
N
H
H
N
H
acetone/THF,
pyridine (cat)
S
N
OsO4, NMO,
acetone/THF,
pyridine (cat)
N
33-36
37
N
F
F
O
B–
K+
4
N
F
I
N
N
N
H
N
H
Br
N
N
S
N
Pd(dppf)Cl2.DCM,
Cs2CO3, aq. dioxane,
Pd(dppf)Cl2.DCM,
N
38
85 oC, 50-75%
Br
Cs2CO3, aq. dioxane,
cis olefin
85 oC, 50-75%
Introduction of bulk at the tertiary, pseudobenzylic position off
the pyrimidine ring (diethyl, 11, and cyclohexyl ring system, 13)
resulted in a loss of activity, particularly against the ParE mutant
strain of S. pyogenes. For 13, both biochemical and antibacterial
activities were significantly reduced compared to other members
of the series. Introduction of a heteroatom into the alicyclic 6-
membered ring (14–18), however, increased on-target ATPase
inhibition, suggesting a subtle conformational role, possibly a flat-
tening of the ring. Bacterial membrane penetration was reduced in
the piperazine containing examples (17, 18), the effect more pro-
nounced for the secondary amine. Contraction of the ring to an
azetidine (19), leaving the free NH group more exposed, reduced
antibacterial activity further. However, the physically constrained
oxetane analog 20 displayed strong potency across the entire
OH
R5
N
HO
N
4
i) MeOOC(CH2)2COCl,
NaH, DMF, 16%
O
N
N
N
OH N
N
N
S
Pd(dppf)Cl2.DCM,
N
H
H
ii) OsO4, NMO,
acetone/THF,
pyridine (cat)
Cs2CO3, aq. dioxane,
Br
Br
6
85 oC, 50-75%
39, R5 = C(CH3)=CH2
N
41
OH
40, R5
=
HO
Scheme 3. Preparation of secondary alcohol, diol-containing benzothiazole ureas.
borohydride-mediated reduction of the mixed anhydride of 5-bro-
mopyrimidine-2-carboxylic acid and isobutyl chloroformate, in the
presence of N-methylmorpholine, afforded primary alcohol 8.
SAR for this series illustrated that alcohol-containing com-
pounds had potent antibacterial activity against a primary panel
Table 3A
Biochemical, antibacterial activity of secondary alcohol-based benzothiazole ureas
R2
N
HO
O
N
N
H
N
S
N
H
S. aureus GyrB S. aureus
S. pyogenes
H. influenzae S. aureus
S. pyogenes
S. aureus 29213 + 50%
Entry
ATPasea
29213a
51339a
49247a
T173 N (GyrB)a A53S (ParE)a
horse seruma
N
R2
Et–
0.006
0.007
0.012
0.008
0.006
0.011
0.06
0.03
0.12
0.06
0.12
0.12
0.06
0.06
0.12
0.06
0.12
0.12
1
1
0.25
0.25
0.25
0.25
2
1
2
2
1
2
2
0.25
0.25
2
21
22
23
24
25
26
n-Pr–
tert-Bu–
cyclo-Pr–
HOCH2–
16
0.5
0.5
2
0.25
0.5
O(CH2CH2)2NCH2–
1
0.25
a
lg/mL.

J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
4219
Table 3B
Biochemical, antibacterial activity of diol-containing benzothiazole ureas
R4
N
O
N
N
H
N
S
N
H
S. aureus GyrB S. aureus
S. pyogenes
H. influenzae
S. aureus
S. pyogenes
S. aureus 29213 + 50%
Entry
ATPasea
29213a
51339a
49247a
T173 N (GyrB)a
A53S (ParE)a
horse seruma
N
R4
OH
0.007
0.004
0.005
0.009
0.25
0.06
0.06
0.5
0.12
0.12
0.12
0.12
1
1
1
1
1
2
0.25
0.25
0.25
0.5
33
34
35
36
OH
OH
0.25
0.5
2
0.5
0.25
4
F HC
2
OH
OH
F C
3
OH
OH
OH
O
OH
0.004
0.006
0.12
0.12
0.12
0.12
0.5
0.5
0.5
1
1
2
0.25
0.25
37
42
F C
3
OH
OH
HO
a
lg/mL.
Br
Br
Br
R2
R2
R2
N
N
KOH/DMSO
or K2CO3/MeCN
N
HN
NMM, NaCNBH3,
THF, 45-60%
N
HN
+
R1
+
R1
Br
R1
O
43
O
(CH2O)n,
MeOH,
MeNH2,
NMM
I
N
H
N
S
O
N
H
I
N
N
N
S
Br
44
N
Br
45
N
HO
N
O
N
HO
N
N
N
HO
N
N
S
Pd(PPh3)4,K3PO4,
B(Pin)2, KOAc,
PCy3, Pd2(dba)3
N
O
dioxane/MeOH,85 oC
N
Br
Dioxane, 85 oC
47
B
O
Br
6
46
N
N
O
i) B(gly)2, KOAc,
HO
N
O
N
H
HO
N
Pd(dppf)Cl2.DCM,
toluene, 130 oC, µwave
N
N
N
HCl (g)
N
S
N
H
N
S
MeOH/DCM
ii) 43, Pd(dppf)Cl2.DCM
Cs2CO3, aq. dioxane
R2
N
N
R2
N
N
49-59
R1
48
R1
Scheme 4. Preparation of C7-modified benzothiazole ureas.
panel, with only a modest two-fold shift in MIC in the presence of
50% horse serum.
Discovery of this possible steric constraint around the pyrimi-
dine C5 appendage prompted us to explore additional secondary
alcohol and diol-containing molecules with further solubility
enhancement potential. While this approach introduced latent
complications of additional chiral centers, we chose initially to
work within the racemic series (Scheme 3). For simple secondary
alcohols, the building blocks were available through aldehyde
quenching of 5-bromo-2-lithiopyrimidine as illustrated above.
Suzuki coupling with 4 once again delivered final products 21–25
in good yields. The morpholine derivative 26 came from Suzuki

4220
J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
Table 4
Further profiling of C7-modified benzothiazole ureas
N
O
HO
N
N
H
N
S
S. aureus GyrB
S. aureus
S. pyogenes
51339a
H. influenzae
S. aureus 29213 + 50% horse Mouse PPB
N
H
Entry
ATPasea
29213a
49247a
seruma
(FU)b (%)
C7
C7
N
N
0.006
0.012
0.005
0.008
0.06
0.5
0.12
0.5
1
4
1
2
0.25
0.5
5.1
6.7
6.3
6.1
3
49
50
51
N
N
N
N
0.06
0.12
0.06
0.06
0.25
0.25
N
N
N
F
F
0.007
0.006
0.06
0.12
0.03
0.12
> 16
0.25
0.25
4.1
5.6
52
53
F
F
N
H
2
N
N
0.009
0.007
0.008
0.011
0.004
0.011
0.25
0.5
0.5
0.12
1
0.25
0.25
0.25
0.12
0.25
0.12
4
4
8
2
2
4
0.5
0.5
0.25
0.25
1
9.0
19.4
19.4
11.0
22.0
22.3
54
55
56
57
58
59
N
O
N
N
HO
N
N
HO
O
N
N
N
N
HO
N
0.5
0.5
HO
N
a
l
g/mL.
b
See footnote 19 for methodology.
vinylation of the pyrimidine (27), epoxide formation (28), opening
to building block 29, and finally coupling with the core. For diols
derived from trans-olefins, the building blocks came from primary
alcohol 8, which we converted to the bromomethylpyridine bro-
mide 30 via the unisolated mesylate. Arbuzov replacement with
triethyl phosphite gave Wadsworth–Horner–Emmons precursor
31, which underwent Suzuki coupling with 4 to give 32. Olefin-
ation with aldehydes afforded trans-olefins, which underwent cis-
dihydroxylation17 in the presence of OsO4 to give the final products
33–36. We prepared trans-diol 37 first by assembling the cis-ole-
fin-based building block via Suzuki reaction between 2-iodo-5-
bromopyridine and the trifluoroborate of 2-propene, coupling 38
with 4, and dihydroxylation of the cis-olefin. Finally, we made ter-
tiary diol 41 from 8 in a multi-step procedure, first by conversion
to its methoxysuccinyl ester, which underwent E2 elimination
under reaction conditions to form 39 as a by-product. Dihydroxy-
lation and coupling with 4 as before completed the synthesis.
Tables 3A and 3B illustrate the profile of compounds within this
subset of compounds. Extension of R2 to longer chain alkyl and
small cycloalkyl groups (21, 22, 24) had virtually no effect on the
potency of the series towards the enzyme. Echoing our earlier
observations for bulkier rings in the tertiary series, the large t-butyl
moiety (23) gave a substantial drop in activity against Haemophilus
influenzae, for reasons not readily apparent, although we also
observed a concomitant drop in activity against S. aureus in the
presence of serum. The loss of permeability of the lipooligosaccha-
ride-containing outer membrane of this Gram-negative bacterium
is one possible explanation. The morpholine group 26, designed to
enhance hydrophilicity, was reasonably well tolerated, as was diol
25.
Extension of the diol-containing series is illustrated in Table 3B.
Once again, the introduction of small groups did not hamper on-
target or antimicrobial activity significantly, but diol 36 once again
illustrated the relative intolerance of larger rings. Particularly

J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
4221
encouraging, however, was the high level of potency seen in the
presence of serum, suggesting that protein binding within the ser-
ies was reduced or rapidly reversible. Lastly, while the examples in
Table 3B all reflected racemic compounds, there appeared to be lit-
tle, if any effect on diol geometry; the cis-diol 35 (resulting from
the trans-olefin) was within a doubling dilution of its trans-diol
analog 37. Further exploration of the chiral versions of the alcohol
series will be the subject of a future report.
We next turned our attention to the C7 portion of the benzothi-
azole core. Since the alcohol moiety was well-tolerated, we chose
the gem-dimethylcarbinol moiety of 3 as the starting point for
our exploration. By incorporating further property enhancing
groups (e.g., mildly basic moieties capable of forming salts,
Fig. 2) designed not to interfere with binding, we targeted
improved solubility and lower intrinsic plasma protein binding.
Preparation of the C7-modifications followed
a convergent
Figure 3. Improved efficacy in the neutropenic mouse thigh model.
approach (Scheme 4) wherein the C7 moieties arose either from
reductive aminations of 2-bromo-4-formylpyridine or amine dis-
placement of 2-bromo-4-bromomethylpyridine to yield building
blocks 43. We prepared the triazone 4518 in near quantitative
yields by incubating 44 with paraformaldehyde and methanolic
methylamine under basic conditions, which then smoothly cou-
pled with boronate 46, available directly from precursor 6. C7 bro-
mide 47 underwent a two-step conversion, first to the unisolated
glycolatoboron ester under microwave conditions, and then subse-
quent Suzuki coupling with building blocks 43 to yield the fully
elaborated triazones, represented as 48. Release of the final prod-
ucts 49–59 came via HCl deprotection of the triazones in DCM/
methanol.
We further assessed examples in this series in an ultracentrifu-
gation-based protein binding assay19 in the presence of mouse
plasma, chosen to inform the potential for subsequent animal effi-
cacy studies. The introduction of basic groups significantly
increased the unbound fraction (FU), while maintaining high on-
target potency and low MICs against Gram-positive targets
(Table 4).
We were pleased to observe that not only was 3 effective in this
model by both oral and intravenous routes, but it showed compa-
rable efficacy to linezolid and superior efficacy than novobiocin.
Moreover, the study was conducted with a single dose of 3, as com-
pared to our earlier experiment requiring multiple dosing of car-
boxylic acids 1 and 2. Pharmacokinetic parameters for 3 in the
mouse20 revealed a bioavailability of 34%, consistent with the
observed drops in CFU values for the two routes of administration.
In conclusion, introduction of alcohol-containing moieties at
the C5 position of the benzothiazole series, along with modifica-
tions at the C7 position designed to enhance hydrophilicity and
reduce protein binding, yielded significant enhancements to an
established core with potent Gram-positive antibacterial activity.
The observation of proof-of-concept efficacy, at least as good as
the marketed standard of care, in a gold-standard, high hurdle,
model of infection via both intravenous and oral dosing, suggests
that this series is suitable for further enhancement towards clinical
development for both hospitalized and outpatient groups. Further
details on this series will be published in due course.
Notably, while we observed consistently good activity against
the GyrB ATPase of S. aureus and low MIC values against both S.
aureus and S. pyogenes, we saw a drop-off of activity against H.
References and notes
influenzae (values ranging from 1 to >16 lg/mL) thus suggesting
either a change in permeability properties or efflux. We also
observed a limited reduction in activity against the S. pyogenes
1. Magiorakos, A.-P.; Srinivasan, A.; Carey, R. B.; Carmeli, Y.; Falagas, M. E.; Giske,
C. G.; Harbarth, S.; Hindler, J. F.; Kahlmeter, G.; Olsson-Liljequist, B.; Paterson,
D. L.; Rice, L. B.; Stelling, J.; Struelens, M. J.; Vatopoulos, A.; Weber, J. T.; Monnet,
D. L. Clin. Microbiol. Infect. 2012, 18, 268.
2. Silver, L. L. Clin. Microbiol. Rev. 2011, 24, 71.
3. Haydon, D.J.; Czaplewski, L.G.; Palmer, N.G.; Mitchell, F.R.; Atherall, J.F.; Steele,
C.R.; Ladduwahetty, T., U.S. Patent # 7,977,340.
A53S (ParE) mutant strain, but still observed single-digit
potency within the series. Activity against the S. aureus T173N
(GyrB) mutant strain remained at single digit g/mL or less, the
lg/mL
l
4. a) Maxwell, A. Nat. Struct. Biol. 1996, 3, 109; b) Wang, J. C. Quart. Rev. Biophys.
1998, 31, 107; c) Drlica, K.; Zhao, X. Microbiol. Mol. Biol. Rev. 1997, 61, 377; (d)
Laponogov, I.; Veselkov, D. A.; Sohi, M. K.; Pan, X.-S.; Achari, A.; Yang, C.;
Ferrara, J. D.; Fisher, L. M.; Sanderson, M. R. PLoS One 2007, 2, e301. http://
dx.doi.org/10.1371/journal.pone.0000301.
5. Collin, F.; Karkare, S.; Maxwell, A. Appl. Microbiol. Biotechnol. 2011, 92, 479.
6. Charifson, P. S.; Grillot, A.-L.; Grossman, T. H.; Parsons, J. D.; Badia, M.; Bellon,
S.; Deininger, D. D.; Drumm, J. E.; Gross, C. H.; LeTiran, A.; Liao, Y.; Mani, N.;
Nicolau, D. O.; Perola, E.; Ronkin, A.; Shannon, D.; Swenson, L. L.; Tang, Q.;
Tessier, P. R.; Tian, S.-K.; Trudeau, M.; Wang, T.; Wei, Y.; Zhang, H.; Stamos, D. J.
Med. Chem. 2008, 51, 5243.
7. East, S. P.; White, C. B.; Barker, O.; Barker, S.; Bennett, J. M.; Brown, D.; Boyd, E.
A.; Brennan, C.; Chowdhury, C.; Collins, I.; Convers-Reignier, E.; Dymock, B. W.;
Fletcher, R.; Haydon, D. J.; Gardiner, M.; Hatcher, S.; Ingram, P.; Lancett, P.;
Mortenson, P.; Papadopoulos, K.; Smee, C.; Thomaides-Brears, H. B.; Tye, H.;
Workman, J.; Czaplewski, L. G. Bioorg. Med. Chem. Lett. 2009, 19, 894.
8. Starr, J. T.; Sciotti, R. J.; Hanna, D. L.; Huband, M. D.; Mullins, L. M.; Cai, H.; Gage,
J. W.; Lockard, M.; Rauckhorst, M. R.; Owen, R. M.; Lall, M. S.; Tomilo, M.; Chen,
H.; McCurdy, S. P.; Barbachyn, M. R. Bioorg. Med. Chem. Lett. 2009, 19, 5302.
9. Tari, L. W.; Trzoss, M.; Bensen, D. C.; Li, X.; Chen, Z.; Lam, T.; Creighton, C. J.;
Cunningham, M. L.; Kwan, B.; Stidham, M.; Shaw, K. J.; Lightstone, F. C.; Wong,
S. E.; Nguyen, T. B.; Nix, J.; Finn, J. Bioorg. Med. Chem. Lett. 2013, 23, 1529.
10. Tari, L. W.; Li, X.; Trzoss, M.; Bensen, D. C.; Chen, Z.; Lam, T.; Zhang, J.; Lee, S. J.;
Hough, G.; Phillipson, D.; Akers-Rodriguez, S.; Cunningham, M. L.; Kwan, B. P.;
Nelson, K. J.; Castellano, A.; Locke, J. B.; Brown-Driver, V.; Murphy, T. M.; Ong,
V. S.; Pillar, C. M.; Shinabarger, D. L.; Nix, J.; Lightstone, F. C.; Wong, S. E.;
hallmark of dual-targeting, indicating a low overall propensity
for developing resistance. Gratifyingly, we achieved an increase
in free fraction within the series; the installation of 5-membered
rings further substituted with ether or alcohol containing moieties
(54–56) increased the unbound fraction from 9 to 19%, although
this enhancement was partially offset by a moderate increase in
MIC. Azetidine-containing examples 57–59 enhanced this property
even further, giving us optimism for further in vivo testing with a
regimen wherein lower or less frequent dosing would be required.
To assess the efficacy of our compounds, we initially chose 3 as
a proof-of-concept example, with a moderate free fraction (5.1%
unbound) and a strong antibacterial profile, in the aforementioned
neutropenic mouse thigh model. Again, mice inoculated with
ꢀ2 Â 105 CFUs of S. aureus (ATCC 19636) were given a single dose
of test compound at 30 mg/kg (intravenously) and 100 mg/kg (oral
gavage), using identical formulation conditions for both routes. We
included two control compounds, novobiocin, with a similar mech-
anism of action to our series, and linezolid, as the current standard
of care in Gram-positive infections. Figure 3 indicates the results in
this Letter.

4222
J. T. Palmer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 4215–4222
Nguyen, T. B.; Shaw, K. J.; Finn, J. PLoS One 2013, 8(12), e84409. http://
dx.doi.org/10.1371/journal.pone.0084409.
containing the compounds. For the S. aureus topoisomerase IV ATPase assay,
the final assay composition was 25 nM topoisomerase IV enzyme (C2E2
complex), 0.01 mg/mL DNA, 50 mM Tris-HCl pH 7.5, 5 mM magnesium
chloride, 0.35 M potassium glutamate, 0.05 mg/mL BSA and 5 mM DTT,
1 mM ATP and 1% DMSO solution containing the compounds. The reactions,
11. Axford, L. C.; Agarwal, P. K.; Anderson, K. H.; Andrau, L. N.; Atherall, J.; Barker,
S.; Bennett, J. M.; Blair, M.; Collins, I.; Czaplewski, L. G.; Davies, D. T.; Gannon, C.
T.; Kumar, D.; Lancett, P.; Logan, A.; Lunniss, C. J.; Mitchell, D. R.; Offermann, D.
A.; Palmer, J. T.; Palmer, N.; Pitt, G. R. W.; Pommier, S.; Price, D.; Rao, B. N.;
Saxena, R.; Shukla, T.; Singh, A. K.; Singh, M.; Srivastava, A.; Steele, C.; Stokes, N.
R.; Thomaides-Brears, H. B.; Tyndall, E. M.; Watson, D.; Haydon, D. J. Bioorg.
Med. Chem. Lett. 2013, 23, 6598.
12. Palmer, J.T.; Lunniss, C.L.; Offermann, D.A.; Axford, L.C.; Blair, M.; Mitchell, D.;
Palmer, N.; Steele, C.; Atherall, J.; Watson, D.; Haydon, D.; Czaplewski, L.;
Davies, D.; Collins, I.; Tyndall, E.M.; Andrau, L.; Pitt, G.R.W.; PCT Int. Appl.
WO2012/045124.
in a volume of 25 lL, were started by the addition of the ATP and allowed to
incubate at 30 °C for 60 min. Reactions were stopped by adding 0.2 mL
malachite green solution (0.034% malachite green, 10 mM ammonium
molybdate, 1 M HCl, 3.4% ethanol, 0.01% tween 20). Color was allowed to
develop for 5 min and the absorbance at 600 nm was measured
spectrophotometrically. The half-maximum inhibitory concentration (IC50
)
values were determined from the absorbance readings using no compound and
no enzyme controls. The assay sensitivity limits of this assay preclude precise
13. Stokes, N. R.; Thomaides-Brears, H. B.; Barker, S.; Bennett, J. M.; Berry, J.;
Collins, I.; Czaplewski, L. G.; Gamble, V.; Lancett, P.; Logan, A.; Lunniss, C. J.;
Peasley, H.; Pommier, S.; Price, D.; Smee, C.; Haydon, D. J. Antimicrob. Agents
Chemother. 2013, 57, 5977.
values <0.005 lg/mL.
17. VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1973, 1976, 17.
18. Knapp, S.; Hale, J. J.; Bastos, M.; Gibson, F. Tetrahedron Lett. 1990, 31, 2109.
19. Protein binding was measured using a centrifugation method and neat plasma.
14. Weiss, W. J.; Murphy, T.; Lenoy, E.; Young, M. Antimicrob. Agents Chemother.
2004, 48, 1708. Thighs of female CD-1 mice (18–22 g) rendered neutropenic by
the intraperitoneal administration of cyclophosphamide (150 mg/kg at day-4
and 100 mg/kg at day -1) were inoculated with approximately 2 Â 105 CFUS of
MSSA type strain S. aureus (ATCC 19636) prepared from a fresh overnight
culture. Compounds were administered at two hour intervals (2, 4, 6 h post
inoculation) and CFUs enumerated at 24 h post dosing by harvesting the
thighs, homogenizing in saline on ice, and plating serial dilutions onto charcoal
containing plates for growth overnight and colony counting. MIC values for S.
aureus type strains ATCC 29213 and ATCC 19636 have been consistent within
this compound series.
15. (a) Unpublished results; (b) PDB ID: 3FV5; Wei, L.; Letiran, A. Design and
synthesis of novel C7-derived aminobenzimidazole ureas: bacterial gyrase/
topoisomerase IV inhibitors with potent Gram-positive antibacterial activity.
16. ATPase assay: enzymes were purchased from Inspiralis Ltd. (Norwich, United
Kingdom). For the S. aureus DNA gyrase ATPase assay, the final assay
composition was 10 nM DNA gyrase enzyme (A2B2 complex), 0.08 mg/mL
DNA, 40 mM HEPES pH 7.6, 10 mM magnesium acetate, 0.5 M potassium
glutamate, 0.01 mg/mL BSA and 2 mM DTT, 1 mM ATP and 5% DMSO solution
Compounds were tested at 10 lM (1% DMSO), and allowed to equilibrate at
37 °C for 60 min. Two aliquots of each sample were subjected to
ultracentrifugation (250,000g for 4.2 h at 37 °C to separate plasma proteins
from plasma water. 50 lL supernatant (plasma water) was sampled per tube to
determine the free compound concentration, compared to samples that had
not been centrifuged but were subject to the same incubation period.
Compound degradation was assessed by testing samples that had not been
centrifuged and kept at À20 °C. Diltiazem was used as a stability control and
has plasma instability over the incubation period. Supernatants were analyzed
using LCMS.
20. Swiss male albino mice were assessed via intravenous dose formulation (10%
DMSO, 40% TEG, 10% HPbCD), administered as a single bolus dose through tail
vein at dose level of 3 mg/kg, and via oral dose formulation (10% DMSO, 40%
TEG, 10% HPbCD) by oral gavage needle at dose level of 3 mg/kg. Dose volumes
were 5 mL/kg. Sampling time points were pre-dose, 0.08, 0.25, 0.5, 1, 2, 4, and
8 h. Calculated pharmacokinetic parameters were as follows: IV dosing:
C
0 = 7.4
lg/mL; AUC (0–8 h) = 11.9
lg h/mL; t1/2 = 68 min; CL = 0.5 mL/min/
kg; Vd = 0.4 L/kg. PO dosing: Cmax = 2.1
l
g/mL; AUC (0–8 h) = 2.1 g h/mL; t1/2
l
=
97 min; F = 34%.