Design, synthesis and structure-activity relationships of substituted oxazole-benzamide antibacterial inhibitors of FtsZ

Article abstract of DOI:10.1016/j.bmcl.2013.11.002

The design, synthesis and structure-activity relationships of a series of oxazole-benzamide inhibitors of the essential bacterial cell division protein FtsZ are described. Compounds had potent anti-staphylococcal activity and inhibited the cytokinesis of the clinically-significant bacterial pathogen Staphylococcus aureus. Selected analogues possessing a 5-halo oxazole also inhibited a strain of S. aureus harbouring the glycine-to-alanine amino acid substitution at residue 196 of FtsZ which conferred resistance to previously reported inhibitors in the series. Substitutions to the pseudo-benzylic carbon of the scaffold improved the pharmacokinetic properties by increasing metabolic stability and provided a mechanism for creating pro-drugs. Combining multiple substitutions based on the findings reported in this study has provided small-molecule inhibitors of FtsZ with enhanced in vitro and in vivo antibacterial efficacy.

Full text of DOI:10.1016/j.bmcl.2013.11.002

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis and structure–activity relationships of substituted
oxazole–benzamide antibacterial inhibitors of FtsZ
Neil R. Stokes a, , Nicola Baker a,à, James M. Bennett a, Pramod K. Chauhan c, Ian Collins a,
David T. Davies a,§, Maruti Gavade c, Dushyant Kumar c, Paul Lancett a, Rebecca Macdonald a,
Leanne MacLeod a,–, Anu Mahajan c, Jeffrey P. Mitchell b, Narendra Nayal c, Yashodanand Nandan Nayal c,
Gary R. W. Pitt b, Mahipal Singh c, Anju Yadav c, Anil Srivastava c, Lloyd G. Czaplewski a,k
,
David J. Haydon a,
⇑
a Biota Europe Ltd, Begbroke Science Park, Oxfordshire OX5 1PF, United Kingdom
b Biota Scientific Management Pty Ltd, 10/585 Blackburn Road, Notting Hill, VIC 3168, Australia
c Jubilant 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:
The design, synthesis and structure–activity relationships of a series of oxazole–benzamide inhibitors of
the essential bacterial cell division protein FtsZ are described. Compounds had potent anti-staphylococcal
activity and inhibited the cytokinesis of the clinically-significant bacterial pathogen Staphylococcus
aureus. Selected analogues possessing a 5-halo oxazole also inhibited a strain of S. aureus harbouring
the glycine-to-alanine amino acid substitution at residue 196 of FtsZ which conferred resistance to pre-
viously reported inhibitors in the series. Substitutions to the pseudo-benzylic carbon of the scaffold
improved the pharmacokinetic properties by increasing metabolic stability and provided a mechanism
for creating pro-drugs. Combining multiple substitutions based on the findings reported in this study
has provided small-molecule inhibitors of FtsZ with enhanced in vitro and in vivo antibacterial efficacy.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 2 October 2013
Revised 1 November 2013
Accepted 4 November 2013
Available online xxxx
Keywords:
Antibacterial
FtsZ
Cell division
3-Methoxybenzamide
MRSA
Small-molecule inhibitors of the essential and conserved pro-
karyotic cell division protein FtsZ offer promise as candidate new
antibacterial agents for the treatment of drug-resistant infec-
tions.1–6 In particular, synthetic analogues of 3-methoxybenzam-
ide7 1 (3-MBA, Fig. 1), such as PC190723 (2), have demonstrated
the potential to inhibit methicillin-resistant Staphylococcus aureus
in vitro and in relevant animal models of infection.8–11 Although
no FtsZ inhibitors have yet progressed through clinical develop-
ment, compound 2 is the best-characterised lead compound of this
novel class. These authors recently reported another substituted
derivative of 3-MBA, 3, and its succinate pro-drug, 4 (Fig. 1).12
Benzamides 3 and 4 exhibited a number of superior in vitro and
in vivo properties over 2. The most significant improvements were
whole-cell potency against the FtsZ G196A amino acid substitution
that renders S. aureus non-susceptible to 2, enhanced pharmacoki-
netic properties and improved in vivo efficacy.
Compounds 3 and 4 have two key structural differences relative
to 2: (i) the presence of a phenyl-substituted oxazole heterocycle
in place of the thiazolopyridine; and (ii) substitutions at the pseu-
do-benzylic position of the linker between the benzamide and oxa-
zole moieties. These novel improvements were discovered by
means of an iterative medicinal chemistry and screening effort.
This Letter describes the design and synthesis of a substantial
number of novel analogues of compounds 3 and 4 prepared during
this programme and reports their biological properties, along with
the structure–activity relationships (SAR) of the oxazole chemo-
type and the effects of substitutions on the benzamide linker. Syn-
thesized compounds were initially evaluated in minimal inhibitory
concentration (MIC) assays with S. aureus strain ATCC 29213, or a
derivative carrying the previously-characterised8 G196A amino
acid substitution in FtsZ, using the broth microdilution method
according to the guidelines of the Clinical and Laboratory Stan-
dards Institute.13 As with earlier studies the cell division inhibition
assay was used to monitor the on-target action of these ana-
logues.8–12 This assay provides a convenient confirmation of
⇑
Corresponding author. Tel.: +44 (0)1865 854700; fax: +44 (0)1865 854799.
E-mail address: d.haydon@biota.com.au (D.J. Haydon).
Present address: Redx Anti-Infectives Ltd, Alderley Park, Cheshire SK10 4TF,
United Kingdom.
à
Present address: College of Life Sciences, University of Dundee, Dundee DD1 5EH,
United Kingdom.
§
Present address: ANTABIO, Biostep, 31670 Labège, France.
Present address: TMO Renewables Ltd, Surrey Research Park, Guildford GU2 7YF,
–
United Kingdom.
k
Present address: Chemical Biology Ventures, Thame, Oxfordshire OX9 3EZ, United
Kingdom.
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.11.002
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

2
N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
O
case of the 4-chloro analogue (7), potency was increased fourfold.
H2
N
The greatest uplift in antibacterial potency was measured with the
4-methoxy analogue (8) for which an eightfold decrease in MIC
was observed.
O
A feature of the oxazole chemotype is the option for substitu-
tions at the C5 position of the heterocycle. In silico modeling using
the crystal structure of 2 bound to FtsZ8,19 indicated that when
compound 5 was docked into the putative binding site there was
space to extend substitutions from the C5 position (data not
shown). To this end, trisubstituted oxazoles 9–28 were prepared
as outlined in Scheme 1, combining substitutions to both the phe-
nyl moiety and the C5 position of the oxazole.20 Combining substi-
tutions in this way produced compounds with markedly enhanced
1 (3-MBA )
O
H N
2
F
F
O
S
N
potencies against wild-type S. aureus, with MICs as low as 0.03 lg/
N
mL recorded (Table 1). Again, the inhibition of cell division pheno-
type tracked potency. A wide range of alkyl and aryl substitutions
at the C5 position were tolerated and retained activity. Analogues
with amino (11) and nitro (12) groups were found to be inactive.
Unexpectedly, with certain combinations of substitutions, inhibi-
tion of the S. aureus G196A mutant was also observed. Compounds
10, 23, 25, and 26 that harbour halogen substitutions (specifically a
Cl
2 (PC190723)
O
H N
F
2
F
F
F
bromine or chlorine atom) at the C5 position had single-digit
lg/
F
mL MICs against the mutant strain and still acted by blocking cell
O
N
division.12
Selected compounds with potent in vitro anti-staphylococcal
activity were evaluated for their in vivo pharmacokinetic proper-
ties. The pharmacokinetic parameters for the representative
compounds 16 and 19 are shown in Table 2. With clearance rates
of P137 mL/min/kg and the area under the curve (AUC)
HO
O
Br
3
60.4 gꢀh/mL the pharmacokinetic properties of these compounds
l
O
were considered to be unsuitable for demonstrating efficacy in
animal models of infection. However, data with the non-oxazole
chemotypes depicted in Figure 2 indicated that substitutions at
the pseudo-benzylic carbon position could improve the pharmaco-
H2N
F
F
F
F
F
O
N
R
kinetic properties of benzamide derivatives. For example,
a
O
O
methyl-substituted analogue (30) of the oxadiazole-benzamide
showed improvements in its pharmacokinetic profile over the
unsubstituted oxadiazole- benzamide (29) (Fig. 3). The same effect
was also observed for the substituted benzothiazole-benzamide
(32) relative to the unsubstituted benzothiazole-benzamide (31)
(Fig. 3). This learning was then applied to the oxazole–benzamide
chemotype producing analogues with improved pharmacokinetic
profiles. The potentially labile pseudo-benzylic positions of 23,
25 and 26 were blocked by the introduction of a methyl group to
produce compounds 33, 34 and 35, respectively. The pharmacoki-
netic parameters for these three analogues are also presented in
Table 2. In each case compounds containing the methyl substitu-
tion showed good AUCs, half-lives and clearance rates. The
pharmacokinetic profile was even better in the case of the
hydroxymethyl-substituted analogue of 23, that is, compound 36
(Table 2).
The in vitro stability of compounds 16, 19, 35 and 36 was eval-
uated in a range of chemical and biological matrices. Data are pre-
sented in Table 3. All four compounds were stable over a 24 h
period in phosphate-buffered saline at pH 7.4. All of the com-
pounds were also stable in mouse plasma. In the presence of
mouse microsomes or mouse hepatocytes, the half-lives of com-
pounds 35 and 36, which have a methyl and hydroxymethyl sub-
stitution, respectively, at the carbon adjacent to the 2-oxazolyl
position, were higher than the half-lives of compounds 16 and
19, which have no substitution at the pseudo-benzylic position.
These data indicate that such substitutions improve metabolic sta-
bility, which is a critical pharmacokinetic parameter of compounds
for use in staphylococcal infections.
O
Br
O
HO
4
Figure 1. Chemical structures of 1 (3-methoxybenzamide), 2 (PC190723), 3 and 4.
whole-cell activity based on the well-understood phenotypic
response to chemical or genetic inhibition of FtsZ in cocci.14,15 In
each case the lowest concentration of compound at which inhibi-
tion of cell division in S. aureus was observed is reported. Selected
compounds were also evaluated for their in vivo pharmacokinetic
behaviour in mice as well as their stability in vitro in various chem-
ical and biological matrices.16,17
Our initial investigations into the oxazole series began with the
synthesis and characterization of phenyl oxazole 5. Synthesis of
this target commenced, with a Blümlein–Lewy oxazole-forming
reaction between 2-benzyloxyacetamide and phenacyl bromide
as shown in Scheme 1. Subsequent treatment with boron tribro-
mide resulted in concomitant debenzylation/bromination and the
resulting alkyl halide was then used to alkylate 2,6-difluoro-3-
hydroxybenzamide18 and yield benzamide 5. Three other
2,4-substituted oxazoles (6–8) were also prepared as outlined in
Scheme 1. Compound 5 inhibited wild-type S. aureus with an
MIC of 16
ing cell division at around the same concentration. This is consis-
tent with exerting its antibacterial effect through the
lg/mL (Table 1) and was also observed to begin inhibit-
5
disruption of FtsZ function. Small substituents at the 4-position
of the phenyl ring improved the antibacterial potency. For the
4-hydroxy analogue (6), potency was improved twofold. In the
Compound 36 and a variety of other pseudo-benzylic substi-
tuted examples were prepared by means illustrated in Scheme 2.
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
O
NH2
F
6: R = OH, Y = H
Br
(b)
(n)
N
N
8: R = OMe, Y = H
R = OMe, Y = Br
N
(e)
O
(f)
(c)
F
O
O
10: R = Cl, Y = Br
18: R = Cl, Y = nPrOH
23: R = CF3, Y = Br
24: R = CF3, Y = CN
26: R = OCF3, Y = Br
(o)
(p)
N
Br
O
R
Br
R
R
R
O
(d)
Y
Br
5: R = H, X = H
O
O
NH2
X
7: R = Cl, X = H
O
Br
(b)
(a)
(c)
9: R = Cl, X = Me
Bn
N
N
F
F
11: R = Cl, X = NH2
12: R = Cl, X = NO2
13: R = Cl, X = OMe
R = Cl, X = CH2OAc
14: R = Cl, X = CH2OH
15: R = OMe, X = OEt
O
O
(q)
(r)
N
O
R
X
X
R
R
R
O
X = H
X = CH3 (for 9, 14)
X = H (others)
(k)
X
X = NO2 (for 11, 12)
X = CH3
(m)
X = CH2OAc (for 14)
(c)
N
R'
O
R'
O
O
O
O
Cl
O
Bn
N
(g)
(h)
(j)
O
O
H2N
N
H
O
O
R
Bn
R
O
R'
R
R'
R
R' = Me (for 13)
R' = Et (for 15)
Scheme 1. Reagents and conditions: (a) 2-benzyloxyacetamide, DMF/130 °C, 6 h; (b) BBr3, dichloromethane (DCM)/ꢁ78 °C to 25 °C, 2 h; (c) 2,6-difluoro-3-hydroxyben-
zamide, K2CO3, DMF/rt, 6 h; (d) acetamide, DMF/130 °C, 6 h; (e) NBS, acetic acid/rt, 2 h; (f) azobisisobutyronitrile (AIBN), NBS, CCl4/reflux, 3 h; (g) 2-benzyloxyacetyl chloride,
tetraethylammonium (TEA), CH2Cl2/0 °C to 25 °C; (h) POCl3, benzene/reflux; (j) (i) H2, Pd/C, MeOH/rt/1 atm, 1 h, (ii) SOCl2, CH2Cl2; (k) HNO3, H2SO4; (m) (i) SeO2, xylene/
reflux, (ii) NaBH4, MeOH/reflux, 2 h, (iii) AcCl, TEA, CH2Cl2/0 °C to rt, 0.5 h; (n) Zn, AcOH/120 °C, 1 h; (o) (i) methyl acrylate, Pd(OAc)2, PPh3, dimethylacetamide (DMA) (Heck
coupling), (ii) H2, Pd/C, tetrahydrofuran (THF), (iii) lithium aluminium hydride (LAH), THF; (p) CuCN, pyridine, microwave heating; (q) SnCl2ꢀ2H2O, EtOH; (r) K2CO3, MeOH/rt
1 h.
Table 1
SAR of substitutions at the C4- and C5-positions of the oxazole chemotype
O
H2N
F
F
R4
O
N
O
R5
Compound
R4
R5
S. aureus ATCC 29213
Phenotypea
FtsZ G196A
MIC (lg/mL)
(lg/mL)
MIC (lg/mL)
2
5
6
7
8
9
N/A
Ph
N/A
H
H
H
H
Me
Br
NH2
NO2
OMe
CH2OH
OEt
Propyl
Allyl
3-OH propyl
Et
Cyclopropyl
Ph
Thiazol-5-yl
Br
CN
1
16
8
4
0.5
16
8
4
2
>64
>64
>64
>64
>64
>64
2
>64
>64
n.d.
> 64
> 64
n.d.
>64
>64
>64
>64
>64
n.d.
4
4-OH Ph
4-Cl Ph
4-OMe Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-OMe Ph
4-OMe Ph
4-OMe Ph
4-OMe Ph
4-CF3 Ph
4-CF3 Ph
2
0.5
0.06
>64
>64
0.5
8
0.06
0.03
0.06
2
0.12
0.12
0.12
2
0.12
2
0.5
0.03
> 64
> 64
0.5
8
0.03
0.015
0.06
2
0.12
0.12
0.12
2
0.12
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
>64
(continued on next page)
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

4
N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Table 1 (continued)
Compound
R4
R5
S. aureus ATCC 29213
FtsZ G196A
MIC (lg/mL)
Phenotypea
(
lg/mL)
MIC (lg/mL)
25
26
27
28
4-OCF3 Ph
4-OCF3 Ph
4-OCF3 Ph
4-OCF3 Ph
N/A
Cl
Br
I
CF3
N/A
N/A
0.12
0.06
0.06
0.5
2
0.12
0.03
0.06
0.5
N/A
N/A
8
2
>64
>64
n.d.
n.d.
Linezolid
Vancomycin
N/A
1
N/A, not applicable.
n.d., not determined.
a
Lowest concentration at which an inhibition of cell division phenotype was measured.
Table 2
In vivo pharmacokinetic parameters of compounds 16, 19, 33, 34, 35, and 36
following a 3 mg/kg IV dose in mice
Parameter (unit)
C0 g/mL)
AUC (
t1/2 (min)
CL (mL/min/kg)
16
19
33
34
35
36
(l
1.2
0.4
25
137
5.0
1.1
0.2
10
319
4.3
2.0
2.2
63
22
2.0
2.1
1.6
79
31
3.5
2.5
1.4
53
36
2.7
4.2
6.7
127
7
l
gꢀh/mL)
Volume of distribution (L/kg)
1.3
O
H2
N
F
Cl
F
O
N
N
O
29
O
Figure 3. Pharmacokinetic profiles of 29, 30, 31 and 32 in mice following a single IV
administration. Compound 29 (open squares) and compound 30 (solid squares)
were dosed individually at 3 mg/kg. Compound 31 (open circles) and compound 32
(solid circles) were dosed in a cassette at 2 mg/kg. Samples were taken after 5, 15,
30, 60, 120, 240 and 480 min and total plasma concentrations of the compounds
were determined by LC-MS/MS.
H
N
2
F
Cl
F
O
N
N
O
30
Table 3
In vitro stability of selected compounds in various chemical and biological matrices
O
H2
N
F
Parameter (unit)
16
19
35
36
Chemical stability
(% remaining after 24 h)
Plasma stability
(% remaining after 24 h)
Microsomal stability
(Half-life, min)
98
97
96
102
F
O
S
101
68.2
17.9
100
104
100
N
Cl
71.6
23.4
>120
101.2
>120
71.5
31
Hepatocyte stability
(Half-life, min)
O
H2
N
F
the gem dimethyl analogue, 37, was striking and pointed to the
likely requirement for a specific chiral configuration due to steric
hindrance at the ligand binding site.
F
O
S
N
Up to this stage, the pseudo benzylic-substituted analogues had
been prepared as racemic mixtures. Investigations were therefore
undertaken to determine whether one or both enantiomers were
active. To this end, a batch of compound 36 was resolved by chiral
chromatography resulting in pure (>99%) enantiomers 48 and 3.21
Circular Dichroism (CD) and Optical Rotatory Dispersion (ORD)
analyses revealed compound 48 to be levorotatory and compound
3 dextrorotatory.22 Each of the two separate enantiomers was then
tested for biological activity (Table 5) and the results indicated that
Cl
32
Figure 2. Chemical structures of 29, 30, 31 and 32.
The biological data indicated that a diverse range of substitutions
were tolerated, including extended alkyls, alcohols, amines and
five- and six-membered heterocycles (Table 4). The inactivity of
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
5
A
A
Br
N
A
O
NH2
F
33: A = Me, R = CF3
35: A = Me, R = OCF3
37: A = di-Me, R = Cl
38: A = Et, R = Cl
O
Br
O
A
F
O
N
O
H2N
(c)
(b)
R
O
N
(d)
A
N
(a)
O
Br
Br
R
O
Br
R
R
R
O
O
36: n = 1, X = OH, R = CF3
46: n = 1, X = OAc, R = CF3
39: n = 1, X = OH, R = Cl
42: n = 2, X = NMe2, R = CF3
(a)
(g)
( )n
H2N
N
O
O
O
Ac
Ac
F
NH2
O
F
NH2
O
( )n
( )n
Br
(h)
( )n
O
Y
F
( )n
Ac
F
( )n
41: n = 2, X = OH, R = CF3
O
O
N
(j,k)
X
N
N
47: n = 1, X =
, R = CF3
O
(b)
(c)
(d)
(e)
O
N
O
O
N N
43: n = 3, X = OH, R = CF3
44: n = 2, X = CO2H, R = CF3
(j)
Br
Br
O
O
R
Y = CO2Me
Y = CH2OH
Y = CH2OAc
Br
(f)
(g)
Br
R
R
R
R
Scheme 2. Reagents and conditions: (a) DMF/130 °C, 6 h; (b) NBS, acetic acid/rt, 2 h; (c) AIBN, NBS, CCl4/reflux, 3 h; (d) 2,6-difluoro-3-hydroxybenzamide, K2CO3, DMF/rt, 6 h;
(e) K2CO3, MeOH/rt, 1 h; (f) NaBH4, MeOH/reflux, 2 h; (g) AcCl, TEA, CH2Cl2/0 °C to rt, 0.5 h; (h) (i) PBr3, toluene/reflux, 0.5 h, (ii) NMe2, THF/rt, 16 h; (j) Jones Reagent, acetone/
0 °C to rt, 16 h; (k) (i) acetic hydrazide, EDC, HOBt, DMAP, DMF/rt, (ii) POCl3, toluene/reflux. Additionally, 5-chloro-oxazole 34 was prepared by the same route as shown for
bromide 35 using NCS in place of NBS. Diol 45 and pyridyl derivative 40 were prepared as described in WO2012/142671 A1, 2012 (see representative example 11, page 51).
Table 4
SAR of benzylic substitutions of the oxazole chemotype
O
H2N
F
F
R4
O
N
O
Rα
R5
a
Compound
R4
R5
R
S. aureus ATCC 29213
g/mL) (lg/mL)
Phenotypea
MIC (l
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
4-CF3 Ph
4-OCF3 Ph
4-OCF3 Ph
4-CF3 Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-Cl Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
Br
Cl
Me
Me
Me
CH2OH
Gem-dimethyl
Et
CH2OH
Pyrid-3-yl
0.06
0.12
0.12
0.25
>64
2
0.25
4
0.5
8
1
8
0.5
0.5
8
0.06
0.25
0.12
0.5
—
1
0.25
4
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
CH2CH2OH
0.5
16
1
CH2CH2N(Me)2
CH2CH2CH2OH
CH2CH2CO2H
CH2CH(OH)CH2OH
CH2OAc
8
0.5
0.5
16
(5-Methyl-1,3,4-oxadiazol-2-yl)methyl
a
Lowest concentration at which an inhibition of cell division phenotype was measured.
the anti-staphylococcal and cell division inhibitory activity resided
primarily with the (+) form. A stereoselective synthetic route
was subsequently developed using commercially available
N-Boc-(OBn)-serine as the chiral source (Scheme 3) and this served
to establish the absolute configuration of the required stereoiso-
mer. Key steps in the synthetic sequence included diazotization/
substitution of the amino acid with overall retention of configura-
tion and a subsequent Mitsunobu coupling with chiral inversion to
give the R enantiomer of the target product. This fortuitously
proved to be the desired more active configuration (3), as
confirmed by S. aureus susceptibility testing (Table 5). The less
Table 5
Biological activity of compound 36 and purified enantiomers 48 and 3
Compound
Stereo
assignment
S. aureus ATCC 29213
MIC
Phenotypea
g/mL) g/mL)
FtsZ G196A
MIC (lg/mL)
(l
(l
36
48
3
Racemic
(ꢁ)
0.25
16
0.5
16
8
64
4
(+)
0.12
0.25
a
Lowest concentration at which an inhibition of cell division phenotype was
measured.
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

6
N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
O
NH
O
NH
O
NH
2
2
2
O
O
H
N
F
F
F
F
F
F
HO
BOC
OH
O
(d)
(a)
(b)
(c)
O
O
O
O
O
O
O
O
Bn
Bn
–
+
OH
O
O
O
Cs
OH
O
Bn
NH
2
O
NH
O
NH
2
O
F
2
F
F
F
F
F
O
O
O
3
(e)
(g)
(f)
O
O
OH
N
N
CF
CF
3
3
CF
3
OH
O
OH
O
O
Br
Scheme 3. Reagents and conditions: (a) (i) TFA/rt, 1 h, (ii) NaNO2, water, acetate buffer (pH 4), acetone/5 °C to rt, 16 h, (iii) H2SO4, trimethylorthoformate, MeOH/reflux, 16 h;
(b) PPh3, diethyl azodicarboxylate (DEAD), 2,6-difluoro-3-hydroxybenzamide, TEA, THF/0 °C to rt, 48 h; (c) H2, Pd(OH)2/C, THF/rt/1 atm, 1 h; (d) Cs2CO3, water, acetonitrile/
55 °C, 4 h; (e) 2-bromo-1-(4-trifluoromethylphenyl)ethanone, DMF/rt, 1 h; (f) BF3ꢀEt2O, acetamide, xylenes/reflux, 2 h; (g) Br2, CH2Cl2/rt, 16 h.
active enantiomer (48) was prepared by a slight modification to
Table 7
the sequence, using Mitsunobu esterification with 4-nitrobenzoic
acid and then hydrolysis to invert the chirality of the secondary
alcohol immediately prior to step (b). This ultimately yielded the
S configuration of the final compound and a 128-fold reduction
of biological activity compared to the more active enantiomer 3
was observed by S. aureus susceptibility testing (Table 5).
In vitro kinetic solubility of selected ester and carbamate derivatives at pH 7.4
O
H2N
F
F
F
F
F
O
N
As well as providing enhanced metabolic stability, substitutions
at the pseudo-benzylic position also offered the prospect of
enhancing the solubility of the compounds either by increasing
the intrinsic solubility of the parent molecule or by providing a
handle to introduce pro-drug functionality via an ester linkage.
At neutral pH the kinetic solubility23 of compound 36 was
O
O
R
Br
O
Compound
R
Solubility (lg/mL)
50
51
52
53
54
55
56
CH2CH2CO2H
Piperazin-1-yl
NHCH2CO2H
200
200
800
200
400
100
800
100 lg/mL, compared to 12.5 lg/mL of the unsubstituted ana-
logue, compound 23 (Table 6). A number of other substituted ana-
logues also had improved solubility over compound 23 (Table 6).
To explore this opportunity, a library of derivatives encompassing
ester 50 and carbamates 51–56 was prepared by modification of
alcohol 36 according to methods described elsewhere.24 This li-
brary was evaluated for solubility (Table 7). Using this strategy,
improvements in the solubility were observed with various substi-
tutions, such as a succinate (50), piperazine (51), and amino tetra-
zole (56).
NHCH2CH2NHMe
2-Pyrrolidin-1-ylethylamino
2-(1H-Imidazol-4-yl)ethylamino
1H-Tetrazol-5-ylamino
Having established the significance of a halogen-substituted
oxazole for potency against the FtsZ G196A mutant, the impor-
tance of a substituted pseudo-benzylic position for metabolic sta-
bility and a pro-drug strategy for enhancing solubility, these
three chemical features were combined to produce the advanced
lead compound 3 and its succinate pro-drug 4. As reported previ-
ously, the improved in vitro properties of 3 and 4 translated into
enhanced in vivo pharmacokinetics and efficacy in the murine
thigh model of staphylococcal infection.12
In summary, the SAR of advanced benzamide-derived FtsZ
inhibitors has been presented. The activity of the oxazole–benzam-
ide chemotype has been described and the importance of 5-halo
substitutions to the oxazole group in driving potency against the
mutant encoding the FtsZ G196A amino acid substitution has been
reported. In parallel, the significance of substitutions to the pseu-
do-benzylic position for driving improvements in the pharmacoki-
netic properties by increasing metabolic stability and by providing
a means to enhance solubility has been established.
Table 6
In vitro kinetic solubility at pH 7.4 of selected compounds
O
H2N
F
F
R4
O
N
O
Rα
R5
a
Compound
R4
R5
R
Solubility (lg/mL)
23
36
43
44
45
46
49
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
4-CF3 Ph
Br
Br
Br
Br
Br
Br
Br
H
12.5
100
50
400
100
25
CH2OH
CH2CH2CH2OH
CH2CH2CO2H
CH2CH(OH)CH2OH
CH2OAc
Acknowledgements
This work was funded in part by an investment from The Well-
come Trust under the Seeding Drug Discovery Initiative. We thank
CH2OPO(OH)2
>800
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002

N. R. Stokes et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
7
0.1% v/v DMSO. Microsomes were added to a final protein concentration of
1 mg/mL and reactions were started by the addition of NADPH to
concentration of 1 mM. Reactions were stopped at 0, 10, 30 and 60 min by
the addition of 100 L of DMSO. For the hepatocyte stability assay, compounds
Steve Ruston and Simon Tucker for their encouragement during
the execution of this work. Circular Dichroism and Optical Rotatory
Dispersion spectroscopy were performed by Chiralabs Ltd (Beg-
broke, United Kingdom). Compounds are the subject of interna-
tional patent application numbers WO2007/107758 and
WO2012/142671. The authors have been paid for the work pre-
sented and some of the Biota authors are inventors on patents
and hold shares in Biota Pharmaceuticals Inc.
a
l
were tested in duplicate at 10 lM with a final DMSO concentration of 0.5% v/v
and incubated with 5 ꢂ 104 cryopreserved CD-1 mouse hepatocytes
(CellzDirect, Durham, NC) in a 100 lL volume. The assay was stopped by the
addition of 100 lL DMSO at seven time points up to 120 min. In all stability
experiments, samples were extracted using acetonitrile and centrifugation,
and the supernatants were analyzed by LC-MS.
18. 2,6-Difluoro-3-hydroxybenzamide was prepared from commercially available
2,6-difluoro-3-methoxybenzamide by treatment with BBr3 under nitrogen in
CH2Cl2 at ambient temperature for 48 h.
References and notes
19. Tan, C. M.; Therien, A. G.; Lu, J.; Lee, S. H.; Caron, A.; Gill, C. J.; Lebeau-Jacob, C.;
Benton-Perdomo, L.; Monteiro, J. M.; Pereira, P. M.; Elsen, N. L.; Wu, J.;
Deschamps, K.; Petcu, M.; Wong, S.; Daigneault, E.; Kramer, S.; Liang, L.;
Maxwell, E.; Claveau, D.; Vaillancourt, J.; Skorey, K.; Tam, J.; Wang, H.;
Meredith, T. C.; Sillaots, S.; Wang-Jarantow, L.; Ramtohul, Y.; Langlois, E.;
Landry, F.; Reid, J. C.; Parthasarathy, G.; Sharma, S.; Baryshnikova, A.; Lumb, K.
J.; Pinho, M. G.; Soisson, S. M.; Roemer, T. Sci. Transl. Med. 2012, 4, 126.
20. In addition to the steps shown explicitly, phenyl oxazole 21 was prepared via
Suzuki coupling of the appropriate oxazolyl bromide. Compounds 16, 17, 19
and 22 were prepared from the corresponding oxazolyl bromide by means of
Stille coupling and, where required, subsequent Pd-catalysed hydrogenation.
1. Vollmer, W. Appl. Microbiol. Biotechnol. 2006, 73, 37.
2. Lock, R. L.; Harry, E. J. Nat. Rev. Drug Disc. 2008, 7, 324.
3. Czaplewski, L. G.; Stokes, N. R.; Ruston, S.; Haydon, D. J. In Antibiotic Discovery
and Development; Dougherty, T. J., Pucci, M. J., Eds.; Springer: New York, 2012.
4. Awasthi, D.; Kumar, K.; Ojima, I. Expert Opin. Ther. Pat. 2011, 21, 657.
5. Schaffner-Barbero, C.; Martín-Fontecha, M.; Chacón, P.; Andreu, J. M. ACS Chem.
Biol. 2012, 17, 269.
6. Ma, S.; Ma, S. Chem. Med. Chem. 2012, 7, 1161.
7. Ohashi, Y.; Chijiiwa, Y.; Suzuki, K.; Takahashi, K.; Nanamiya, H.; Sato, T.;
Hosoya, Y.; Ochi, K.; Kawamura, F. J. Bacteriol. 1999, 181, 1348.
8. Haydon, D. J.; Stokes, N. R.; Ure, R.; Galbraith, G.; Bennett, J. M.; Brown, D. R.;
Baker, P. J.; Barynin, V. V.; Rice, D. W.; Sedelnikova, S. E.; Heal, J. R.; Sheridan, J.
M.; Aiwale, S. T.; Chauhan, P. K.; Srivastava, A.; Taneja, A.; Collins, I.; Errington,
J.; Czaplewski, L. G.; Science 2008, 321, 1673.
9. Czaplewski, L. G.; Collins, I.; Boyd, E. A.; Brown, D.; East, S. P.; Gardiner, M.;
Fletcher, R.; Haydon, D. J.; Henstock, V.; Ingram, P.; Jones, C.; Noula, C.;
Kennison, L.; Rockley, C.; Rose, V.; Thomaides-Brears, H. B.; Ure, R.; Whittaker,
M.; Stokes, N. R. Bioorg. Med. Chem. Lett. 2009, 19, 524.
10. Haydon, D. J.; Bennett, J. M.; Brown, D.; Collins, I.; Galbraith, G.; Lancett, P.;
Macdonald, R.; Stokes, N. R.; Chauhan, P. K.; Sutariya, J. K.; Nayal, N.; Srivastava,
A.; Beanland, J.; Hall, R.; Henstock, V.; Noula, C.; Rockley, C.; Czaplewski, L. J.
Med. Chem. 2010, 53, 3927.
11. Brown, D.R. Collins, I. Czaplewski, L.G. Haydon, D.J. International Patent,
publication WO2007/107758.
5-Chloro- and 5-iodo-oxazoles 25 and 27 were synthesised in
a manner
analogous to 26 using N-chlorosuccinimide (NCS) or N-iodosuccinimide (NIS)
in place of N-bromosuccinimide (NBS). Trifluoromethyl derivative 28 was
prepared by treatment of the appropriate oxazolyl iodide intermediate with
methyl 2,2-difluoro-2-(fluorosulfonyl)acetate and copper(I) iodide in HPMA
and N,N-dimethylformamide (DMF).
21. Enantiomers 48 and 3 were separated from a 1 g sample of the racemic
mixture 36 using a CHIRALPAKÒ AD-H, (250 ꢂ 30) mm, 5
lm 5 122 column
with a n-hexane/EtOH/MeOH (80:10:10 v/v/v) mobile phase.
22. For the CD analysis acetonitrile solutions of the compounds were prepared and
dispensed into pre-cleaned Quartz Suprasil cylindrical cells. CD spectra were
acquired on
a Jasco J720 spectrometer. Instrument parameters were:
wavelength range, 200–360 nm; spectral bandwidth, 1 nm; scan speed,
10 nm/min; response time, 4 s; accumulations, 4; data interval, 0.1 nm;
nominal cell length, 1.0 cm; temperature, 20 0.1 °C. In each case
a CD
12. Stokes, N. R.; Baker, N.; Bennett, J. M.; Berry, J.; Collins, I.; Czaplewski, L. G.;
Logan, A.; Macdonald, R.; MacLeod, L.; Peasley, H.; Mitchell, J. P.; Nayal, N.;
Yadav, A.; Srivastava, A.; Haydon, D. J. Antimicrob. Agents Chemother. 2013, 57,
317.
13. Clinical and Laboratory Standards Institute. CLSI Document M7–A7. Clinical
and Laboratory Standards Institute, Wayne, PA.
spectrum and UV absorption spectrum were simultaneously acquired. Spectra
were corrected by solvent baseline subtraction, followed by normalization for
molar concentration and pathlength. For the ORD analysis tetrahydrofuran
solutions of the compounds were prepared and dispensed into pre-cleaned
polarimetry cells. Optical rotations were determined using
a specially
converted J600 spectropolarimeter. Instrument parameters were: wavelength
range, 320–800 nm; spectral bandwidth, 1 nm; scan speed, 50 nm/min;
response time, 1 s; accumulations, 4; data interval, 0.1 nm; nominal cell
length, 1.0 cm; temperature, 20 0.1 °C. In each case an ORD spectrum and UV
absorption spectrum were simultaneously acquired. Spectra were corrected by
solvent baseline subtraction, followed by normalization for concentration and
pathlength.
14. Pinho, M. G.; Errington, J. Mol. Microbiol. 2003, 50, 871.
15. Stokes, N. R.; Sievers, J.; Barker, S.; Bennett, J. M.; Brown, D. R.; Collins, I.;
Errington, V. M.; Foulger, D.; Hall, M.; Halsey, R.; Johnson, H.; Rose, V.;
Thomaides, H. B.; Haydon, D. J.; Czaplewski, L. G.; Errington, J. J. Biol. Chem.
2005, 280, 39709.
16. Groups of healthy adult male Balb/c mice received compounds by intravenous
(IV) or oral (PO) administration at a dose level of 3 mg/kg. In all cases the dose
volume was 5 mL/kg. Compounds were formulated in 10% dimethyl sulfoxide
(DMSO), 40% tetraethylene glycol (TEG), and 20% 2-hydroxypropyl-b-
cyclodextrin (HPbCD) in water (final). At various time points, samples of
blood were collected for analysis. Three animals were used for each time point.
Total plasma concentrations of compounds were measured by LC-MS/MS and
pharmacokinetic parameters were calculated using Win-Nonlin.
23. The kinetic solubility of the compounds was measured using a turbidimetric
method. A series of doubling dilutions of compounds were prepared in neat
DMSO. Five microlitre samples were diluted twenty-fold into 95 lL volumes of
phosphate buffered saline (PBS, pH 7.4) in microtiter assay plates and allowed
to reach equilibrium at room temperature for 24 h. The absorbance within each
well of the plate was read spectrophotometrically at a wavelength of 620 nm. A
precipitate forms when the maximum aqueous solubility level is reached. The
value quoted is the highest concentration where the compound was in
solution, that is, where no measurable precipitate was present.
17. The chemical and plasma stability of
a 10 lM solution of compound in
phosphate buffered saline (PBS, pH 7.4) or neat mouse plasma (Sera
Laboratories International Ltd, Haywards Heath, UK) was determined
following 24 h incubation at 37 °C. The final concentration of DMSO was 5%
24. Haydon, D.J. Czaplewski, L.G. Stokes, N.R. Davies, D. Collins, I. Palmer, J.T.
Mitchell, J.P. Pitt G.R.W. Offermann, D. International Patent WO 2012/142671.
v/v. Microsomal stability was assessed at 10 lM with a final concentration of
Please cite this article in press as: Stokes, N. R.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.11.002