Synthesis and SAR of novel parenteral anti-pseudomonal cephalosporins: Discovery of FR264205

Article abstract of DOI:10.1016/j.bmcl.2008.07.085

We describe herein the synthesis and biological evaluation of a series of novel cephalosporins with potent activity against Pseudomonas aeruginosa. Introduction of various amino groups to the 4-position of a 3-amino-2-methylpyrazole cephalosporin 3-side c

Full text of DOI:10.1016/j.bmcl.2008.07.085

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4849–4852
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR of novel parenteral anti-pseudomonal cephalosporins:
Discovery of FR264205
a
a
a
a
a
Ayako Toda a, , Hidenori Ohki , Toshio Yamanaka , Kenji Murano , Shinya Okuda , Kohji Kawabata ,
*
Kazuo Hatano b, Keiji Matsuda a, Keiji Misumi c, Kenji Itoh c, Kenji Satoh c, Satoshi Inoue c
a Chemistry Research Labs, Astellas Pharma Inc., 2-1-6 Kashima, Yodogawa-ku, Osaka 532-8514, Japan
b Pharmacology Research Labs, Astellas Pharma Inc., 2-1-6 Kashima, Yodogawa-ku, Osaka 532-8514, Japan
c Institute for Medical Research, Wakunaga Pharmaceutical Co. Ltd., 1624 Shimokotachi, Koda-cho, Akitakata-shi, Hiroshima 739-1195, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe herein the synthesis and biological evaluation of a series of novel cephalosporins with potent
activity against Pseudomonas aeruginosa. Introduction of various amino groups to the 4-position of a 3-
amino-2-methylpyrazole cephalosporin 3-side chain resulted in enhanced MIC values against multiple
Pseudomonas aeruginosa strains and ultimately led to the discovery of FR264205 (15) with excellent
anti-bacterial activity and weak convulsion effect by direct intracerebroventricular injection assay.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 2 May 2008
Revised 16 July 2008
Accepted 21 July 2008
Available online 24 July 2008
Keywords:
FR264205
Cephalosporin
Pseudomonas aeruginosa
SAR
Pseudomonas aeruginosa is a major nosocomial pathogen and is
responsible for many serious infections in patients who are immu-
nocompromised due to treatment for malignant tumors, neutrope-
nia or resulting from organ transplantation.1 Such infections are
difficult to cure, especially with the ongoing emergence of strains
resistant to the currently available clinically used agents, which of-
ten leads to prolongation of the treatment period. Against this bac-
terium, several members of the carbapenem class of antibacterials
have been effective, but their use has induced resistance due to de-
creased penetration through the outer membrane resulting from
loss of the D2 porin.2 Certain fluoroquinolones also show potent
anti-pseudomonal activity, but some strains have acquired resis-
tance due to mutations of the drug target and/or due to induction
of active efflux pumps.3 In terms of general resistance trends,
whilst much recent research has been focused on tackling the
problems of resistant Gram-positive pathogens such as MRSA
and VRE, only few efforts have been directed at identifying im-
proved agents for P. aeruginosa infections.
zole ring at the 3-position of the cephalosporin nucleus, the dis-
tinctive chemical feature of FK518 is the presence of
a
dimethylacetic acid group in place of the methyl group on the
oxime moiety and a thiadiazole instead of thiazole at the 7-posi-
tion. These modifications were presumed to be responsible for
the improved efficacy of FK518 against P. aeruginosa species.
In the research described herein, we report our efforts aimed to-
ward the discovery of novel cephalosporin derivatives with potent
activity against Class C (AmpC) b-lactamase-producing P. aerugin-
osa strains and a low convulsion-inducing effect, and the discovery
of FR264205 as a novel agent with significantly improved activity
and good potential as a new anti-pseudomonal agent.7
In general terms, a strategy to improve the anti-pseudomonal
activity of cephem derivatives requires either, or all of, the follow-
ing: (1) maintenance of high penicillin-binding protein (PBP) affin-
ity, (2) increased outer membrane permeability, and (3) improved
stability to AmpC b-lactamase. Our previous study showed that the
PBP affinity of FK518 was very high, especially toward PBP3 which
is important for anti-bacterial activity, presumably due to the con-
tribution of the dimethylacetic acid moiety on the oxime at the 7-
position.6 This speculation was supported by the PBP affinity data
of CAZ8 (which also has dimethylacetic acid moiety) which is high-
er compared to FK037 and CZOP (with a 7-position methoxyimino
moiety). Therefore, we selected this structure for our investigations
and our attention was focused on the following strategy.
Among the marketed cephalosporins, Ceftazidime (CAZ) shows
the best anti-pseudomonal activity, but displays very little activity
against AmpC b-lactamase-producing strains which have recently
been increasing and are becoming a significant problem.4 During
the course of our research on fourth-generation broad-spectrum
cephalosporins, we previously discovered FK037 (cefoselis)5 and
FK518 (Fig. 1).6 Although both compounds possess an aminopyra-
The outer membrane of P. aeruginosa represents a formidable
barrier to entry of antibiotics, and the relatively large side chain
at the 7-position was presumed to be unfavorable for outer-mem-
* Corresponding author. Tel.: +81 3 3244 3011.
E-mail address: ayako.toda@jp.astellas.com (A. Toda).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.085

4850
A. Toda et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4849–4852
CH3
O
N
H
HSO4
N
S
N
O
N
N
H2N
NH2
OH
S
N
O
O
O
CO2H
N
N
Cl
CO2H
H
FK037
N
S
N
N
N
H2N
NH2
OH
S
N
O
CO2H
FK518
Figure 1. Structures of FK037 (cefoselis) and FK518.
1. DIEA, DMF, (NaI)
R4
O
O
CO2
N
O
O
CO2H
N
O
O
CO2H
N
H
N
N
H
N
S
H
R4
R3
N
or
R3
N
S
N
N
S
N
R2
R1
N
N
N
S
R1
N
N
X
H2N
N
S
N
N
N
O
H2N
H2N
S
O
N
R2
O
CO2
Y
CH3CO2H
2. TFA, anisole
CO2
CO2
1
X; Cl or I
Y; -CHPh2 or p-methoxybenzyl
Scheme 1. Synthesis of novel cephalosporins.
Table 1
Anti-pseudomonal activities and permeability (differential ratio) of pyridiniumcephems
R:
N
N
N
O
O
CO2H
N
N
H
2
3
5
NH2
H2N
N
S
N
N
N
R
H2N
S
O
NH
CO2
HN
4
NH2
Compound
MIC (lg/mL)
Calculated pKa (amine)
FP2056a
FP1380b
meanc
MIC50
Permeabilityd
c
2
3
4
5
1
1
1
1
1
32
32
8
4
16
2.1
2
1
1
1
1
NT
32
16
8
—
1.71
1.43
1.36
1.51
Non-basic
7.04
8.51
FK518
64
Non-basic
NT, not tested.
a
Class A TEM b-lactamase-induced clinical isolate.
Class C AmpC(ld) b-lactamase-induced clinical isolate.
Fifty-four clinical isolates.
b
c
d
MICDEAE-Dex(ꢀ)/MICDEAE-Dex(+)
.
brane permeability. Our attention was initially focused on the
observation that the outer-membrane permeability of carbapen-
ems is good, and is most likely due to the small molecular size
and the presence of a basic side chain.9 We aimed to apply the
latter observation to cephems by introducing various basic side-
chains to the 3-position of the cephem nucleus. Initially, we
introduced basic substituents to a pyridine ring to obtain a rapid
validation of our hypothesis regarding outer membrane
permeability. Basicity (pKa) of side chains was calculated by ACD/
PhysChem Batch Ver9 (ACD/Labs). Compounds 2–5 were synthe-
sized according to the methods outlined in scheme 1. A chloro-
methyl (or iodomethyl) cephem derivative 1 was coupled with a
pyridine derivative, followed by global deprotection and purifica-
tion to afford the final cephem derivatives.
Table 2
Anti-pseudomonal activities of 2-modified pyrazoliocephems
R:
N
N
NH2
N
NH2
O
O
CO2H
N
N
N
H
6
8
7
CH3
N
S
F
N
N
N
R
N
H2N
NH
N
S
O
N
NH
CO2
(sulfate)
MIC ( g/mL)
(sulfate)
9
Compound
l
FP2056a
FP1380b
Meanc
MIC50
c
6
7
8
9
1
1
1
1
1
32
8
16
4
1.24
1.74
1.47
1.43
1.51
1
1
1
1
1
The in vitro antibacterial activity and outer-membrane perme-
ability of compounds 2–5 is shown in Table 1. MICs were deter-
mined according to the CLSI method,10 permeability was
determined by differential assay from the ratio of MIC values in
FK518
16
a
Class A TEM b-lactamase-induced clinical isolate.
Class C AmpC(ld) b-lactamase-induced clinical isolate.
Fifty-four clinical isolates.
l
g/mL DEAE-Dextran.11 When
b
c
the presence and absence of 100
an amino group is added to the 3-position of the pyridine side

A. Toda et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4849–4852
4851
Table 3
Anti-pseudomonal activities and convulsion-inducing effect of 4-modified pyrazoliocephems
NH
NH2
R:
NH2
NH2
N
H
NH2
O
O
CO2H
N
N
N
N
N
H
N
N
S
10
CH3
11
N
CH3
N
R
H2N
NH2
S
O
N
N
CO2
NH2
(sulfate)
CH3
12
Compound
MIC (lg/mL)
ConvulsiondED50
(
lg/head)
Calculated pKa (amine)
FP2056a
FP7380b
Meanc
MIC50
c
10
11
12
FK518
CAZ
CZOP
0.5
0.5
0.5
1
2
2
8
1
1
16
128
32
0.91
0.66
0.77
1.51
3.7
0.5
0.5
0.5
1
2
2
39.3
4.69
80.1
46.9
71
7.27
10.66
9.66
Non-basic
—
2.7
230
—
a
Class A TEM b-lactamase-induced clinical isolate.
Class C AmpC(ld) b-lactamase-induced clinical isolate.
Fifty-four clinical isolates
b
c
d
Convulsion-inducing effect: ED50 (lg/head, mouse intracerebroventricular injection).
chain (compound 3), an improvement in activity relative to the
non-substituted pyridine analog 2 was observed: mean MIC:1.71
vs 2.1 lg/mL, MIC50 1 lg/mL vs 2 lg/mL (MIC50 = concentration
The structure of the substituent at the 2-position of the 30-
aminopyrazolium group affected MICs against the Class C b-lacta-
mase-producing isolate, whilst MICs against the ClassA b-lacta-
inhibiting 50% of strains). Larger substituents such as a guanidino
group (compound 4) or an aminoethyl group (compound 5) were
tolerated and they showed improved activity against P. aeruginosa.
As we had speculated, the presence of a basic side chain on the 3-
position substituent appears to improve permeability. The perme-
ability was improved as the basicity increased, with the best activ-
ity being observed for the 4-aminoethyl-substituted derivative 5,
with a calculated pKa of 8.51 (Table 1). Although the MICs of the
pyridine analogs 2–5 against P. aeruginosa FP2056, a Class A b-lac-
mase isolate were all the same (1
small substituent at this position showed weaker activity against
Class C b-lactamase-producing strains than FK518 (32 and 16 g/
mL, respectively). This indicated that compound 6 was less stable
to Class C (AmpC) b-lactamase. Relatively large substituents at this
position (7–9) had a good effect on activity against the Class C b-
lg/mL). Compound 6 having a
l
lactamase-producing isolate (MICs: 8, 16, and 4 lg/mL, respec-
tively). It is notable that modification of this part of the 3-position
side chain can alter stability toward Class C b-lactamase. Therefore
steric effects at this position appear to be important for stability.
Although compound 6 only had a moderate MIC against the Class
tamase-inducer clinical isolate, were the same (1 lg/mL), those
against P. aeruginosa. FP1380, a Class C b-lactamase inducer clinical
isolate, were improved in parallel with the improved outer-mem-
brane permeability, and this improved activity against Class C
b-lactamase-producing strains led to an excellent mean MIC value
C-producing isolate, it had the best mean MIC (1.24 lg/mL). There-
fore, we selected the 2-methylpyrazole group for further modifica-
tion, and next introduced various basic side chains at the 4-
position of the pyrazole ring (Table 3).
(1.36 lg/mL; 54 clinically isolated strains).
The observed improvement in activity for the pyridine deriva-
tives led us to speculate that further modification of the pyrazole
moiety of FK518 may lead to clinically significant levels of anti-
pseudomonal activity, since the intrinsic antibacterial activity of
FK518 is superior to the pyridine derivatives. First, we introduced
various substituents to the 2-position of the pyrazole ring to deter-
mine the most suitable structure for further investigation and opti-
mization (Table 2). Compounds (6–9) were prepared as shown in
Scheme 1 and evaluated for antibacterial activity.
Compound 10 was obtained as indicated in our previous publi-
cation.12 Compounds 11 and 12 were obtained as outlined in
Scheme 1. Compound 10, in which the aminomethyl group
(pKa = 7.27) is added to the 4-position of 6, showed significantly
improved anti-pseudomonal activity. MIC against the Class C b-lac-
tamase-producing strain was 8
32 g/mL for 6. The guanidino derivative 11 was even more potent,
with an MIC of 1 g/mL, presumably as a consequence of the high
pKa value (10.66). However, this compound (11) had very strong
lg/mL for 10, as compared to
l
l
OPh
O
OPh
O
NH2
NHBoc
HN
HN
H2N
Et3N, DMF
1. NaNO2, cHCl, H2O
2. 10%-Pd/C, H2, H2SO4
ClCO2Ph, NaOH
dioxane-H2O
N
NH2
N
TrCl, Et3N, THF
N
NH2
HN
N
N
N
NH2
NHTr
N
N
CH3
95%
86%
CH3
16
78%
73%
CH3
CH3
17
18
19
O
CO2tBu
N
NHBoc
H2N
NH2
H
O
O
CO2H
N
S
N
HN
N
HSO4
N
N
H
HN
HN
N
S
N
O
Cl
O
N
O
S
O
1. TFA, anisole, CH2Cl2
2. H2SO4, recryst
N
N
21
N
BSU, KI, DMAc
NHTr
H2N
NH2
CO2PMB
78%
N
S
N
O
CH3
CO2H
CH3
58%
20
15 (FR264205)
Scheme 2. Synthesis of FR264205 (15).

4852
A. Toda et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4849–4852
Table 4
Anti-pseudomonal activities and convulsion-inducing effect of 4-modified pyrazoliocephems
H
R:
N
NH2
NH2
O
N
N
NH2
13
O
O
CO2H
N
NH2
N
N
N
H
CH3
CH3
14
N
S
H
N
H
N
N
NH2
N
R
H2N
S
O
O
N
NH2
N
CO2
CH3
(sulfate)
15
Compound
MIC (lg/mL)
Convulsiond ED50
(l
g/head)
Calculated pKa (amine)
FP2056a
FP1380b
Meanc
MIC50
c
13
14
15
FK518
1
2
2
2
0.93
1.07
0.88 (0.65 )
1.51
0.5
0.5
0.5 (0.5 )
1
122
>200
428
8.55
7.18
7.95
Non-basic
0.5
0.5
1
*
*
16
46.9
a
Class A TEM b-lactamase-induced clinical isolate.
b
c
Class C AmpC(ld) b-lactamase-induced clinical isolate.
Fifty-four clinical isolates (*196 clinical isolates).
d
Convulsion-inducing effect: ED50 (lg/head, mouse intracerebroventricular injection).
convulsion-inducing effect (ED50; 4.69
l
g/head, mouse intracere-
ing P. aeruginosa strains. Furthermore, while convulsion-inducing
activity (due to CNS effects) was comparable to the marketed
cephems, we required an analog with significantly reduced poten-
tial, and discovered that conformational restriction of the 4-posi-
tion substituent on the pyrazolium ring reduced this potential
dramatically and led to the discovery of FR264205 which is a
promising derivative with excellent anti-bacterial activity suitable
for further evaluation.
broventricular injection). Next we altered this guanidine part to
an amino group to attempt to weaken the convulsion-inducing
effect by controlling basicity. As we predicted, the convulsion ef-
fect of 12 was relatively low, but not as low as that of CZOP. Inter-
estingly, the activity of 11 against the Class C b-lactamase-
producing strain remained strong (MIC: 1 lg/mL). Further study
altering the length of the alkyl chain at the 4-position of the
pyrazolium ring showed that an aminopropyl group (12) had the
best profile against this strain (data not shown). A 3D-structure
based study and enzyme inhibitory study suggested steric effects
were important for stability to Class C (AmpC) b-lactamase.12
Next, we investigated the effect of this side chain on anti-
pseudomonal activity and convulsion-inducing effect. Sterically-
restricted derivatives 13-15 were synthesized and evaluated. Com-
pound 15 was synthesized as shown in Scheme 2. While antibacte-
rial activities of these compounds were maintained, a significant
reduction in the convulsion-inducing effect was observed (Table
4). Compound 13 having the E-form olefin showed a lower convul-
sion-inducing effect than non-conformationally restricted deriva-
tive 12. The glycine analog (14) also showed a weak convulsion-
inducing effect, presumably due the further reduced pKa of the
amino group (pKa = 7.18), but this derivative also showed the
weakest mean MIC of the derivatives prepared. The optimal com-
pound obtained was compound 15 (FR264205). This analog
showed the best balance of MIC against the Class C b-lactamase-
producing strain, mean MIC against 54 clinically isolated strains,
and the weakest convulsion-inducing effect in mice, showing sig-
nificantly weaker activity than the marketed cephems CAZ and
CZOP.
Acknowledgment
We are grateful to Dr. David Barrett, Chemistry Research Labo-
ratories, Astellas Pharma, Inc., for advice and support and for assis-
tance in the preparation of this manuscript.
References and notes
1. Bodey, GP.; Bolivar, R.; Fainstein, V.; Jadeja, L. Rev. Infect. Diseases 1983, 5,
279.
2. Minami, S.; Araki, H.; Yasuda, T.; Akama, M.; Iyobe, S.; Mitsuhashi, S.Int. J. Exp.
Clin. Chemother. 1993, 6, 21.
3. Oh, H.; Stenhoff, J.; Jalal, S.; Wretlind, B. Microb. Drug Resist.-Mech. Epidemiol.
Disease 2003, 9, 323.
4. Cheng, K.; Smyth, RL.; Govan, JRW.; Doherty, C.; Winstanley, C.; Denning, N.;
Heaf, DP.; van Saene, H.; Hart, PA. Lancet 1996, 348, 639.
5. Mine, Y.; Watanabe, Y.; Sakamoto, H.; Hatano, K.; Kuno, K.; Higashi, Y.;
Kamimula, T.; Matsumoto, Y.; Tawara, S.; Matsumoto, F. J. Antibiot. 1993, 46,
71.
6. Mine, Y.; Sakamoto, H.; Kamimura, T.; Watanabe, Y.; Matsumoto, Y.; 30th
Interscience Conference on Antimicrobial Agents and Chemotherapy Atlanta,
GA, 1990; Abstr. No. 455.
7. Takeda, S.; Nakai, T.; Wakai, Y.; Ikeda, F.; Hatano, K. Antimicrob. Agents
Chemother. 2007, 51, 826.
8. Nakao, M.; Noji, Y.; Iwahi, T.; Yamazaki, T. J. Antimicrob. Chemother. 1992, 29,
509.
9. Sumita, Y.; Yamaga, H.; Sunagawa, M. J. Antibiot. 1995, 48, 89.
10. Methods for Dilution and Antimicrobial Susceptibility Tests for Bacteria that
Grow Aerobically, Approved M7-A7. Clinical and Laboratory Standards
Institute, 2001.
11. Hancock, RE.; Wang, PG. Antimicrob. Agents Chemother. 1984, 26, 48.
12. Murano, K.; Yamanaka, T.; Toda, A.; Ohki, H.; Okuda, S.; Kawabata, K.; Hatano,
K.; Takeda, S.; Akamatsu, A.; Misumi, K.; Itoh, K.; Inoue, S.; Takagi, T. Bioorg.
Med. Chem. 2007, 16, 2261.
In summary, the synthesis and anti-pseudomonal activities of a
series of novel cephalosporin derivatives have been explored based
on rationally improving activity by increasing outer-membrane
permeability by introduction of various basic amino substituents
to the 3-position substituent. Introduction of amino groups to
the 4-position of a 3-amino-2-methylpyrazole cephalosporin re-
sulted in improved MIC values against Class C b-lactamase-induc-