Evaluation of WO2012177707 and WO2012097269: Vertex's phosphate prodrugs of gyrase and topoisomerase antibacterial agents

Article abstract of DOI:10.1517/13543776.2013.820707

The two patent applications describe two novel compounds in the benzimidazole class of GyrB/ParE antibacterial agents and multiple phosphate prodrugs derived from these compounds. The new benzimidazole compounds have excellent antibacterial activity on Gram-positive strains. But like previous benzimidazoles, they have limited solubility and are highly protein bound. The phosphate prodrugs offer a drug substance with high aqueous solubility that should aid both intravenous and oral formulations. The potential utility of the prodrugs was demonstrated in efficacy studies.

Full text of DOI:10.1517/13543776.2013.820707

Patent Evaluation
Evaluation of WO2012177707
and WO2012097269: Vertex’s
phosphate prodrugs of gyrase
and topoisomerase antibacterial
agents
1. Introduction
2. Chemistry
3. Biology
4. Expert opinion
John Finn
Trius Therapeutics, San Diego, CA, USA
The two patent applications describe two novel compounds in the benzimid-
azole class of GyrB/ParE antibacterial agents and multiple phosphate pro-
drugs derived from these compounds. The new benzimidazole compounds
have excellent antibacterial activity on Gram-positive strains. But like previous
benzimidazoles, they have limited solubility and are highly protein bound.
The phosphate prodrugs offer a drug substance with high aqueous solubility
that should aid both intravenous and oral formulations. The potential utility
of the prodrugs was demonstrated in efficacy studies.
Keywords: antibacterial, DNA gyrase, prodrugs, topoisomerase IV
Expert Opin. Ther. Patents (2013) 23(9):1233-1237
1. Introduction
The emergence of bacterial resistance to the major classes of antibacterial drugs is
causing significant public health problems [1-3]. To address this challenge, research-
ers have actively pursued the discovery of new antibacterial classes that act on novel
targets for the past two decades [4]. This effort has proven to be difficult and the
majority of these programs failed to identify promising candidates [5].
One of the few programs to show progress has been the discovery of novel clas-
ses of antibacterial agents that target the GyrB and ParE subunits of bacterial
DNA gyrase and topoisomerase IV [6]. Bacterial DNA gyrase and topoisomerase
IV both form heterodimeric complexes and are validated antibacterial targets
inhibited by the fluoroquinolone (FQ) antibiotics [7]. The GyrB/ParE ATPase
sites are distal from the FQ-binding site and their value as novel antibacterial tar-
gets has long been appreciated. Inhibition of GyrB is the mechanism of action of
the natural product, novobiocin, discovered in the 1950s [8]. Roche in the late 90s
pioneered the use of structure-based drug design (SBDD) to engineer GyrB-based
antibacterial agents [9,10]. This effort encouraged other groups to target GyrB
using SBDD [11].
In 2001, the first Vertex patent application disclosing the benzimidazole series of
antibacterial agents was published [12]. Details of the potency of the two lead com-
pounds in this series, VRT-125853 and VRT-752586, were published [13-15]. These
compounds are notable for their high enzymatic and antibacterial potency on
Gram-positive strains (VRT-752586 Minimum Inhibitory Concentration (MIC)
values on Staphylococcus aureus range between 0.06 and 0.25 µg/ml). In addition,
the Vertex team designed these compounds to inhibit both GyrB and ParE, a fea-
ture missing in previous bacterial gyrase inhibitors. This design strategy led to com-
pounds with a low frequency for the emergence of new resistance [16].
10.1517/13543776.2013.820707 © 2013 Informa UK, Ltd. ISSN 1354-3776, e-ISSN 1744-7674
All rights reserved: reproduction in whole or in part not permitted
1233

J. Finn
Despite impressive antibacterial potency, compounds
(the racemic versions of 12 and 23) can be separated on
this column.
like VRT-752586 suffer from high protein binding and
presumably low aqueous solubility. The high protein bind-
ing results in significant increases in the MIC values when
the compounds are tested in the presence of serum. For
example, VRT-752586 has its MIC value against S. aureus
increased by 16-fold in the presence of 50% human serum.
The two recent patent applications that are the subject of
this report, WO2012/177707 and WO2012/097269,
detail new antibacterial members of the benzimidazaole
series Formula I [17,18]. While these compounds contain
the potentially solubilizing hydroxyl and cyclic ether func-
tionalities, the main advantage of these new compounds is
the ability to be converted to phosphate prodrugs that
have high aqueous solubility and potentially advantageous
pharmacokinetics.
2.2 WO2012177707
Claims phosphate diesters of Formula I where:
R is fluorine;
X is -- PO(OH)-- OR1, -- PO(O-M+)-- OR1;
M+ is a pharmaceutically acceptable monovalent cation;
and R1 is (C1-- C20)-alkyl, (C2-- C20)-alkenyl, (CH2CH2O)nCH3,
or CH2CH2R2, where n is an integer 1 -- 5 and the key R2
group is a 5 -- 6-membered heterocyclic aliphatic ring
system; such as a morpholine.
The application describes the synthesis of three phosphate
diester prodrugs. Scheme 2 shows the preparation of these
three compounds (X, Y, and Z) from compound 23.
F
3. Biology
N
O
N
N
O
The WO2012/097269 application provides extensive details
regarding the in vitro and in vivo activities of the lead com-
pounds. This includes enzymology, microbiology, pharmaco-
kinetics, efficacy, and safety.
O
O
NH
NH
N
N
N
NH
NH
NH
NH
VRT-125853
VRT-752586
The two compounds have similar antibacterial spectrum
with activity concentrated on Gram-positive strains. Data
are provided on the antibacterial activity of 13 and 23A
against both individual strains and modest size MIC90
panels. Table 1 lists the antibacterial potency against some
of the key strains in these panels. In general, the fluorinated
analog 23A is 4 -- 8-fold more potent than compound 13.
While there is limited spectrum on Gram-negative strains,
compound 23A is active on the respiratory tract pathogens
Haemophilus influenzae and Moraxella catarrhalis. The
MIC values for 23A against S. aureus (ATCC 29213) are
shifted eightfold in 50% human serum, indicating extensive
protein binding.
The alcohols 13 and 23A and the prodrug W are efficacious
in multiple animal infection models when given orally. For
example, compound W gave a 2.25 log cfu reduction (vs con-
trol) at 10 mg/kg in the S. aureus rat kidney infection model.
The performance of 13 and 23A were similar indicating that
there is some bioavailability for both the prodrug and the
active moiety.
Compound 23A was well tolerated in toxicity studies. Oral
administration of 23A to rats for 7 days (twice daily dosing)
demonstrated an NOAEL of 600 mg/kg/day (the highest
dose tested). Oral administration of 23A to monkeys for
7 days (once-daily dosing) demonstrated an NOAEL of
200 mg/kg/day (the highest dose tested). No data are given
for the tolerability of the prodrug W.
PK studies demonstrate that compound W is rapidly con-
verted to 23A when administered either orally or intrave-
nously. Oral bioavailability appears to be high (> 80%) in
monkey but not in rat. No data were given for the PK
O
R
X
O
O
R
N
HO
N
O
NH
N
O
NH
N
N
NH
NH
N
NH
NH
Formula I
R = H, F;
X = H, PO3-2, PO3H2, PO(O )OR′
12 R = H
23 R = F
-
2. Chemistry
2.1 WO2012097269
Claims compounds of Formula I where:
R is hydrogen or fluorine;
X
is hydrogen, -- PO(OH)2; -- PO(OH)O-M+, -- PO
(O-)2*2M+, -- PO(O-)2*D2+;
M+ is a pharmaceutical acceptable monovalent cation; and
D2+ is a pharmaceutical acceptable divalent cation; or a
pharmaceutically acceptable salt.
The application describes the synthesis of two novel benzimi-
dazoles (Formula I R is H or F, and X is H) that have good
antibacterial activity. These compounds were prepared as race-
mates or as optical isomers of high enantiomeric purity. The
resolution was accomplished by chromatography on a column
packed with a chiral stationary phase (CHIRALPAKÒICÒ) to
provide the more active (R) isomer. Scheme 1 shows the prep-
aration of compounds 13 and 23A and their conversion to the
phosphate prodrugs 1B and W. In this scheme, the chiral reso-
lution is conducted at an early stage with the 2-aryl-tetrahydro-
furan intermediate. Experimental details also demonstrate that
the stereoisomers of the final benzimidazole antibacterial agent
1234
Expert Opin. Ther. Patents (2013) 23(9)

Evaluation of WO2012177707 and WO2012097269
R
R
O
R
O
H Pd/C
2
+
R
O
NH
O
2
NO
Br
Pd(OAc)
2
NO
2
2
NO
2
Br
Br
R
R
R
O
a) (CF CO) O
3
2
NBS
ChiralPak
SFC
O N
2
b) NH NO
4
3
NH
CF
O
NH
O
NH
2
2
O
37% yield
95.5% ee
3
OH
N
N
OH
Br
N
NaOH
N
R
O
B
O
R
H
Pd/C
2
O N
2
NH
O
2
O N
2
Pd(PPh )
3 4
O
R
NH
Na CO
2
2
3
O
S
O
O
O
R
N
N
N
N
HO
HO
EtHN
NH
N
NHEt
NH
O
NH
2
N
H
N
NH
NH
2
12 R = H; 13 R = H mesylate salt
23 R = F; 23A R = F mesylate salt
-
+
O M
O
N
-
+
O
N
OBn
OBn
P
O M
P
O
O
N
N
-
a) H Pd/C; HO
2
a) (iPr) NP(OBn)
2
2
R
O
R
b) mCPBA
+
b) H
+-
N
N
c) M OH
NH
O
O
NH
HN
HN
HN
HN
O
1B R = H, M+ = Na+
W R = F, M+ = Na+
Scheme 1. Preparation of the active antibacterial agents 13 and 23A and prodrugs 1B and W.
of 23A (the compound used in the toxicity studies) when
dosed as the parent not the prodrug.
The prodrug strategy resulted in significantly greater
solubility. Table 2 lists the solubility of the different compounds
from both patent applications.
solubility of the new benzimidazole 23 still appears to be
low (reported at < 1 µg/ml), the prodrug strategy may have
addressed the problem of developing effective formulations.
There is ample precedent for phosphate prodrugs providing
a soluble drug substance when the parent molecule has low
solubility [19-21]. The PK data show that rats dosed with W
rapidly cleave the phosphate and generate the parent 23 in
concentrations well above the MIC values (Cmax of 16 µg/ml
when W is given orally at 300 mg/kg and a Cmax of 9 µg/ml
when W is given intravenously at 5 mg/kg). The fact that
prodrug W demonstrated efficacy in animal models sug-
gests that even with significant protein binding, there is
4. Expert opinion
It is likely that the poor solubility of the initial benzimid-
azole leads like VRT-752586 led to problems developing
either an intravenous or oral formulation. While the
Expert Opin. Ther. Patents (2013) 23(9)
1235

J. Finn
R
O
ROH
DIPEA
Cl
O
P
P
N
CN
N
CN
O
R = (CH2)2O(CH2)2OCH3,
R = (CH2)15CH3,
R = (CH2)2
O
N
R
+
O
NH
4
P
-
N
O
CN
O
O
a)
O
O
F
O
P
F
R′
N
N
HO
N
N
O
O
NH
NH
O
b) mCPBA
N
NH
NH
N
NH
NH
c) NH OH
4
X R = (CH2)2O(CH2)2OCH3,
Y R = (CH2)15CH3,
N
O
Z R = (CH2)2
Scheme 2. Preparation of phosphate diesters X, Y, and Z.
Table 1. Antibacterial potency of 13 and 23A against important bacterial pathogens.
13
23A
Bacteria
N
MIC range (mg/ml)
MIC90 (mg/ml)
MIC range (mg/ml)
MIC90 (mg/ml)
Staphylococcus aureus
Enterococcus faecalis
Streptococcus pneumoniae
Haemophilus influenzae
Moraxella catarrhalis
Escherichia coli
Klebsiella pneumoniae
Acinetobacter baumannii
Pseudomonas aeruginosa
67
34
67
55
26
12
12
12
12
0.03 -- 0.5
0.03 -- 0.25
0.015 -- 0.06
0.25 -- 8
0.015 -- 0.12
> 8 -- > 8
> 8 -- > 8
> 8 -- > 8
> 8 -- > 8
0.25
0.25
0.06
2
0.12
> 8
> 8
> 8
> 8
0.008 -- 0.06
0.015 -- 0.12
0.008 -- 0.03
0.06 -- 2
0.004 -- 0.03
2 -- > 8
2 -- > 8
4 -- > 8
> 8-- > 8
0.03
0.06
0.015
1
0.03
> 8
> 8
> 8
> 8
Table 2. Compound solubility.
The second prodrug application is interesting. In contrast
to phosphate monoesters that are rapidly cleaved by phospha-
tases in vivo, phosphate diesters convert much more slowly.
Thus, prodrugs X, Y, and Z are expected to result in a lower
Compound
Solid form
pH
Solubility
(mg/ml)
C
max at a given dose (either intravenously or orally) while still
23
W
X
Y
Z
Crystalline
Crystalline
Crystalline
Crystalline
Amorphous
> 3.0
4.39
4.41
3.01
4.35
< 0.001
0.25
12.74
< 0.001
> 41
generating a similar area under the curve to that of
prodrug W. These slow-release prodrugs could be very useful
especially when toxicity is driven by Cmax (which is likely for
solubility-limited compounds). Of the three diester prodrugs,
the polyether- and amine-containing prodrugs X and Z have
useful aqueous solubility.
enough 23 free drug concentration to be effective. Addi-
tional filings around compound W (a process patent and a
solid form) indicate that Vertex has seriously considered
advancing W into clinical trials [22,23].
Overall, these applications indicate that the Vertex team
has made significant progress. Currently, the only drug class
that is active on drug-resistant Gram-positive strains like
methicillin-resistant S. aureus available in intravenous and
1236
Expert Opin. Ther. Patents (2013) 23(9)

Evaluation of WO2012177707 and WO2012097269
oral formulations are the oxazolidinones. Advancement of a
benzimidazole drug would provide clinicians with a new
drug class for resistant Gram-positive infections and would
be a valuable addition to antibacterial therapy.
Declaration of interest
The author states no conflict of interest and has received no
payment in preparation of this manuscript.
Bibliography
10.
Lubbers T, Angehrn P, Gmunder H,
et al. Design, synthesis and
18. Vertex Pharmaceuticals. Phosphate esters
Papers of special note have been highlighted as
either of interest (ꢀ) or of considerable interest
(ꢀꢀ) to readers.
of gyrase and topoisomerase inhibitors
2012. WO2012177707
structure-activity relationship studies of
ATP analogs as DNA gyrase inhibitors.
Bioorg Med Chem Lett 2000;10:821-6
19. Rautio J, Kumpulainen H, Heimbach T,
et al. Prodrugs: design and clinical
applications. Nat Rev Drug Dis
2008;7:255-70
1.
Arias C, Murray B. Antibiotic-resistant
bugs in the 21st century -- a clinical
super challenge. N Engl J Med
2009;360:439-43
11.
Oblak M, Kotnik M, Solmajer T.
Discovery and development of ATPase
inhibitors of DNA gyrase as antibacterial
agents. Curr Med Chem
.
A prodrug review.
2.
Klein E, Smith D, Laxminarayan R.
Hospitalizations and deaths caused by
methicillin-resistant Staphylococcus
aureus, United States 1999-2005.
Emerg Infect Dis 2007;13:1840-6
20. Im WB, Choi SH, Park JY, et al.
Discovery of torezolid as a novel
5-hydroxymethyl-oxazolidinone
antibacterial agent. Eur J Med Chem
2011;46:1027-39
2007;14:2033-47
12.
13.
Vertex Pharmaceuticals.Gyrase. Inhibitors
and Uses Thereof 2012.
WO2002060879
3.
Boucher H, Talbot G, Bradley J, et al.
Bad bugs, no drugs: no ESKAPE! An
update from the infectious diseases
society of America. Clin Infect Dis
2009;48:1-12
.
An example of a phosphate prodrug
Mani N, Gross C, Parsons J, et al. In
vitro characterization of the antibacterial
spectrum of novel bacterial type II
topoisomerase inhibitors of the
aminobenzimidazole class.
strategy is used to provide a soluble
drug substance for an oxazolidinone
antibacterial agent of
limited solubility.
4.
5.
Silver LL. Challenges of antibacterial
discovery. Clin Microbiol Rev
2011;24:71-109
Antimicrob Agents Chemother
2006;50:1228-37
21. Baker W, Cai S, Dimitroff M, et al.
A prodrug approach toward the
..
A paper that details the antibacterial
properties of the benzimidazole series.
development of water soluble
Payne D, Gwynn M, Holmes D, et al.
Drugs for bad bugs: confronting the
challenges of antibacterial discovery.
Nat Rev Drug Dis 2007;6:29-40
fluoroquinolones and structure-activity
relationships of quinoline-3-carboxylic
acids. J Med Chem 2004;47:4693-709
14.
Grossman T, Bartels D, Mullin S, et al.
Dual targeting of GyrB and ParE by a
novel Aminobenzimidazole class of
antibacterial compounds.
.
An example of a phosphate prodrug
strategy is used to provide a soluble
drug substance for an fluoroquinolone
antibacterial agent of
6.
Collin F, Karkare S, Maxwell A.
Exploiting bacterial DNA gyrase as a
drug target: current state and
perspectives. Appl Microbiol Biotechnol
2011;92:479-97
Antimicrob Agents Chemother
2007;51:657-66
limited solubility.
.
A paper that details the advantages of
a dual-targeting antibacterial agent.
22. Vertex Pharmaceuticals. Process for
making gyrase and topoisomerase IV
inhibitors 2012. WO2012097274
.
A review DNA gyrase as an
antibacterial target
15.
Charifson P, Grillot A, Grossman T,
et al. Novel dual-targeting benzimidazole
urea inhibitors of DNA gyrase and
topoisomerase IV possessing potent
antibacterial activity: intelligent design
and evolution through the judicious use
of structure-guided design and
7.
Sissi C, Palumbo M. In front of and
behind the replication fork: bacterial type
IIa topoisomerases. Cell Mol Life Sci
2010;67:2001-24
23. Vertex Pharmaceuticals. Solid forms of
gyrase inhibitor (R)-1-ethyl-3-[6-fluoro-
5-[ 2-(1-hydroxy-1-methyl-ethyl)
pyrimidin-5-yl] -7-(tetrahydrofuran-2-yl)-
1H-benzimidazol-2-yl]urea
8.
9.
Lambert HP, O’Grady FW. editors.
Coumarins. Churchill Livingstone;
Edinburgh: 1992
structure-activity relationships.
2012. WO2012097273
J Med Chem 2008;51:5243-63
Affiliation
John Finn
16.
17.
Silver L. Multi-targeting by
Boehm H, Boehringer M, Bur D, et al.
Novel inhibitors of DNA gyrase: 3D
structure biased needle screening hit
validation by biophysical methods, and
3D guided optimization. A promising
alternative to random screening.
monotherapeutic antibacterials. Nat Rev
Drug Dis 2007;6:41-55
Trius Therapeutics,
6310 Nancy Ridge Dr,
San Diego, CA 92121, USA
Tel: +1 858 452 0370;
E-mail: jfinn@triusrx.com
Vertex Pharmaceuticals. Pyrimidine
gyrase and topoisomerase IV inhibitors
2012. WO2012097269
J Med Chem 2000;43:2664-74
.
One of the first applications of
fragment screening in drug discovery
Expert Opin. Ther. Patents (2013) 23(9)
1237