Structural requirements for the stability of novel cephalosporins to AmpC β-lactamase based on 3D-structure

Article abstract of DOI:10.1016/j.bmc.2007.11.074

AmpC β-lactamase is one of the leading causes of Pseudomonas aeruginosa (P. aeruginosa) resistance to cephalosporins. FR259647 is a cephalosporin having a novel pyrazolium substituent at the 3-position and exhibits excellent activity (MIC = 1 μg/mL) again

Full text of DOI:10.1016/j.bmc.2007.11.074

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 2261–2275
Structural requirements for the stability of novel cephalosporins
to AmpC b-lactamase based on 3D-structure
b
b
b,
b
Kenji Murano,a,* Toshio Yamanaka, Ayako Toda, Hidenori Ohki, Shinya Okuda,
Kohji Kawabata,b Kazuo Hatano,c Shinobu Takeda,c Hisashi Akamatsu,d
Kenji Itoh,d Keiji Misumi,d Satoshi Inoued and Tatsuya Takagie
*
aLead Discovery Research Labs, Astellas Pharma Inc., 21 Miyukigaoka, Tsukuba, Ibaraki 305-8585, Japan
bChemistry Research Labs, Astellas Pharma Inc., 2-1-6 Kashima Yodogawa-ku, Osaka 532-8514, Japan
cPharmacology Research Labs, Astellas Pharma Inc., 2-1-6 Kashima Yodogawa-ku, Osaka 532-8514, Japan
dInstitute for Medical Research, Wakunaga Pharmaceutical Co., Ltd, 1624 Shimokotachi, Koda-cho,
Akitakata-shi, Hiroshima 739-1195, Japan
eGraduate School of Pharmaceutical Sciences Osaka University, 1-6 Yamada-oka, Suita, Osaka 565-0871, Japan
Received 24 October 2007; revised 22 November 2007; accepted 27 November 2007
Available online 3 December 2007
Abstract—AmpC b-lactamase is one of the leading causes of Pseudomonas aeruginosa (P. aeruginosa) resistance to cephalosporins.
FR259647 is a cephalosporin having a novel pyrazolium substituent at the 3-position and exhibits excellent activity (MIC = 1 lg/
mL) against the AmpC b-lactamase overproducing P. aeruginosa FP1380 strain in comparison with the third-generation cephalo-
sporins FK518 [Abstracts of Papers, 30th Interscience Conference on Antimicrobial Agents and Chemotherapy, Atlanta, GA, Octo-
ber 21–24, 1990, Abs. 454; Abstracts of Papers, 30th Interscience Conference on Antimicrobial Agents and Chemotherapy, Atlanta,
GA, October 21–24, 1990, Abs. 455; Abstracts of Papers, 30th Interscience Conference on Antimicrobial Agents and Chemotherapy,
Atlanta, GA, October 21–24, 1990, Abs. 456; Abstracts of Papers, 30th Interscience Conference on Antimicrobial Agents and Che-
motherapy, Atlanta, GA, October 21–24, 1990, Abs. 457] (MIC = 16 lg/mL) and ceftazidime (CAZ) (MIC = 128 lg/mL). The sta-
bility of FR259647 and FK518 to AmpC b-lactamase was evaluated using MIC assays against both the P. aeruginosa PAO1 strain
and a PAO1 mutant strain overproducing AmpC b-lactamase as a differential assay, which indicates that the main difference derives
from their stability to AmpC b-lactamase. A structural analysis using computer simulations indicated that the difference in stability
may be due to steric hindrance of the 3-position substituents causing differential affinity. This steric hindrance may disturb entry of
the cephalosporins into the binding pocket. We predicted the possibility of inhibition of entry as a potential means of enhancing
stability by conformational analysis. In order to validate this speculation, novel FR259647 derivatives 4–9 were designed, calculated,
synthesized, and evaluated. As a result, we demonstrated that their probability of entry correlated with the MIC ratio of the mutant
strain to the parent strain and supports the validity of our model.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
clinical issue because it often leads to high mortality.2
P. aeruginosa has an intrinsic resistance to many antimi-
crobial agents because of its low outer membrane per-
meability, multiple efflux pumps, and chromosomal
AmpC b-lactamase.3 Only a few antimicrobial agents,
such as amikacin, imipenem, ceftazidime (CAZ) (1),
and ciprofloxacin, are used for treatment of P. aerugin-
osa infections, however, these antimicrobials often have
low activity because P. aeruginosa can acquire various
resistant mechanisms.4,5 Recently, the emergence of
resistant strains against these antimicrobials is a grow-
ing threat to human health. Accordingly, it is an urgent
issue to generate effective antibacterial drugs against the
resistant strains.
Pseudomonas aeruginosa (P. aeruginosa) is a ubiquitous
versatile environmental bacterium that is a leading cause
of opportunistic human infections.1 Infection of im-
mune-suppressed patients, such as elderly or hospital-
ized patients, with this pathogen is now an important
Keywords: AmpC b-lactamase; Conformational studies; Cephalospo-
rin; Pseudomonas aeruginosa.
*
Corresponding authors. Tel.: +81 29 863 7128; fax: +81 29 852 5387
(K.M.); tel.: +81 6 6390 1286; fax: +81 6 6304 5414 (H.O.); e-mail
addresses: kenji.murano@jp.astellas.com; hidenori.ohki@jp.astellas.
com
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.11.074

2262
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
Pseudomonas aeruginosa is naturally susceptible to the
third-generation cephalosporin CAZ.6 However, it is
known that CAZ has low activity against P. aeruginosa
that overproduces AmpC b-lactamase, which is the
primary mechanism of P. aeruginosa resistance to ceph-
alosporins.2 P. aeruginosa produces an inducible, chro-
mosomally encoded AmpC b-lactamase and this
enzyme is usually expressed at very low levels.7 CAZ
has effective antibacterial activity against P. aeruginosa
producing the enzyme at this level because it works as
a suicide inhibitor against the enzyme.8 However, CAZ
has decreased antibacterial activity against P. aeruginosa
overproducing the enzyme, because it is labile to the
overproduced enzyme.9
b-lactamase, which could be due to the slight structural
difference between FR259647 and FK518. This differ-
ence gave us a clue to analyze the structural require-
ments for the stability of FR259647 derivatives to
AmpC b-lactamase.
AmpC b-lactamase has strong activity for the hydrolysis
of cephalosporins, deactivating them.14 AmpC b-lacta-
mase is a serine-dependent enzyme and its catalytic reac-
tion mechanism consists of two steps, acylation of the
active site and hydrolytic deacylation.14,15 Recently,
two X-ray crystal complex structures of AmpC b-lacta-
mase from Escherichia coli (E. coli) with cephalosporins,
CAZ and cephalothin, have been reported and revealed
snapshots of the 3D-structure, such as the complex of
AmpC b-lactamase with intact cephalothin15,16 and the
acyl-enzyme intermediate,8,15 in its hydrolysis mech-
anism.
In order to generate promising anti-P. aeruginosa
agents, research in our group has been directed toward
the development of novel cephalosporins. Our group re-
ported that the cephalosporins FK518 (2)10a–d and
FR259647 (3)11 had 16- and 128-fold improved antibac-
terial activity (MIC) against AmpC b-lactamase over-
producing clinical isolates of P. aeruginosa FP1380
strains in comparison with CAZ, respectively (Table
1). Furthermore, the stability of these cephalosporins
to AmpC b-lactamase was evaluated using MIC assays
against both the P. aeruginosa PAO1 strain and a
PAO1 ampD-defective mutant (DampD/PAO1) strain
overproducing AmpC b-lactamase as a differential assay
and was indicated as the DampD/PAO1:PAO1 strain
MIC ratio.7,12,13 The DampD/PAO1 strain was con-
structed by the disruption of the ampD gene, which sug-
gested that the total amount of penicillin-binding
protein (PBP) and the affinity of the cephalosporins to
PBPs in the mutant strain were the same as those of
the parent strain.7,12 Their MIC ratios indicated that
FR259647 was more stable to AmpC b-lactamase than
FK518 and CAZ (Table 1). These data suggested that
improvement of the antipseudomonal activity of
FR259647 may be due to increased stability to AmpC
So far, the reported structure-based drug design research
using the 3D-structure of AmpC b-lactamase has been
mainly directed toward the design of inhibitors of the
enzyme, not cephalosporins which are stable to the enzy-
me.17a–f In this paper, we attempted to analyze the
structural requirements for the stability to AmpC b-lac-
tamase based on the structural difference in the 3-side
chain of FR259647 and FK518, considering the 3D-
structure of the 3-side chain binding pocket of AmpC
b-lactamase. The results led to the speculation that steric
hindrance of the 3-side chain of cephalosporin nucleus
against AmpC b-lactamase can disturb the entry of
cephalosporins into the 3-side chain binding pocket.
We predicted the possibility of entry by conformational
analysis and considered the relationship between the
probabilities of entry and the MIC ratios of the mutant
strain to the parent strain. From this consideration, we
proposed structural requirements for the stability of
cephalosporins having a pyrazolium ring at the 3-posi-
tion to AmpC b-lactamase, and verified the validity of
Table 1. MICs (lg/mL) of FR259647, FK518, and CAZ against P. aeruginosa FP1380, PAO1, and DampD/PAO1 strains and probability of entry
through the gate of the 3-side chain binding pocket by conformational analysis
R+ =
ceftazidime (CAZ) (1) X = CH,
N+
CO2H
O
N
H
7
N
S
4
N
N+
+R
-
3
FK518 (2)
X = N, R+ =
(HCl salt)
X
O
N
H2N
NH
3
N
2
S
O
(CH2)2OH
CO2
(CH2)3NH2
4
N+
FR259647 (3)
X = N, R+ =
(H2SO4 salt)
NH
3
N
2
Me
Compounds
MIC (lg/mL)
Calculation
Probability of entryb (%)
FP1380
PAO1
DampD/PAO1
MIC ratioa
FR259647 (3)
FK518 (2)
CAZ (1)
1
16
0.5
0.5
2
0.5
4
1
8
0.10
100
N.D.c
128
64
32
a MIC ratio of DampD/PAO1 to PAO1 strain.
b Vide Section 2.2.
c No data.

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2263
this proposal by design and synthesis of novel FR259647
derivatives using the probability of entry as an index.
Scheme 1. Therefore, this indicates that the difference
in the stability between them is due to FR259647 being
only poorly recognized by the enzyme in comparison
with FK518. So we decided to focus only on Step 1
as the key factor for understanding the stability
(Scheme 1).
2. Results and discussion
2.1. Consideration of the stability of FR259647 and
FK518 to AmpC b-lactamase
2.2. Construction of model of the structural requirements
for stability to AmpC b-lactamase
In the reaction pathway for AmpC b-lactamase, as
shown in Scheme 1, a cephalosporin first binds to the en-
zyme to form a precovalent first encounter complex
(complex A). Following nucleophillic attack by the cat-
alytic serine, the enzyme and b-lactam are covalently
bound in an acyl-enzyme complex (intermediate B). A
water molecule then attacks the carboxyl carbon of the
intermediate B, hydrolyzing the covalent enzyme–ligand
bond and generating a hydrolyzed product and the in-
tact enzyme (complex C). Finally, the product leaves,
and the active enzyme is regenerated. Therefore, the sta-
bility of cephalosporin to AmpC b-lactamase is deter-
mined by the following two factors; formation ability
of the complex A, and the thermodynamic stability of
intermediate B. The first factor indicates that com-
pounds which are poorly recognized by the enzyme
should be stable to the enzyme, and the second factor
indicates that compounds converted to a stable interme-
diate B inhibit the enzyme because they can work as a
suicide inhibitor, like CAZ.8 The only structural differ-
ence between FR259647 and FK518 is the substituent
on the 3-pyrazolium ring moiety, as shown in Table 1,
which induces the difference in stability to AmpC b-lac-
tamase. Because FR259647 and FK518 have the same 7-
side chain, the intermediate B derived from them by
reaction with Ser90 in the active center of AmpC b-lac-
tamase should be exactly the same structure, as shown in
In order to investigate the difference in recognition abil-
ity between FR259647 and FK518 by AmpC b-lacta-
mase, the 3D-structure of AmpC b-lactamase from
P. aeruginosa was constructed from the X-ray structure
of AmpC b-lactamase from E. coli (PDB entry 1KVL)15
by the homology modeling method using the FAMS
(Full Automatic Modeling System) program.18 The
modeling structure showed that cephalosporin binding
pocket has a cleft structure and a tunnel structure
bridged by Ala312-Pro322 over the 3-side chain binding
site in the cleft (Fig. 1a). Outside of each end of the tun-
nel, the L side locating at the 7-side chain binding site (B
site) and the catalytic center (A site), and the R side lo-
cated at the outside of the 3-side chain binding site (C
site), was exposed to solvent as shown in Figure 1a
and b. It was speculated that FR259647 and FK518 ap-
proached the binding pocket from the direction of the L
side, because the size of their 7-side chain was larger
than the space of the tunnel and their 7-side chain can-
not pass through the tunnel from the R side. A gate on
the L side of the tunnel located in the white rectangle
was surrounded by the following four residues,
Ala319, Tyr177, Thr343, and Asn373, as shown in Fig-
ure 1c. The gate was almost the same size as the tunnel,
so we speculated that the 3-side chain that has passed
through the gate could be accommodated in the 3-side
Scheme 1. Hydrolysis pathway of cephalosporins by AmpC b-lactamase (E); (A) a precovalent first encounter complex of the enzyme and
cephalosporin (R1, 7-side chain; R, 3-side chain); (B) an acyl-enzyme complex; (C) a complex of a hydrolyzed product and the intact enzyme.

2264
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
Figure 1. (a) Cephalosporin binding pocket of AmpC b-lactamase. Ser90 (ball and stick display) is a catalytic center. The binding pocket forms a cleft
structure and the Connolly surface is displayed with a yellow color. The loop Ala312-Pro322 (green stick display) forms a bridged structure over the
cleft, which shows a tunnel structure. (b) Diagrammatic illustration of cephalosporin binding pocket: A site (Ser90) shows active center; B site shows
7-side chain binding site; C site shows 3-side chain binding site; solid line shows the gate of the 3-side chain binding pocket; rectangle shows tunnel
structure part; outline arrow indicated direction of approach of FR259647 derivatives to binding pocket. (c) The 3-side chain binding site viewed
from L side (see b). The entrance of the 3-side chain binding site displays white rectangle surrounded by the following four residues, Ala319, Tyr177,
Thr343, and Asn373 (ball and stick display). Green stick display shows the loop Ala312-Pro322 except for Ala319. (d) The gate of the 3-side chain
binding pocket of AmpC b-lactamase and 3-side chain of FR259647 approaching the gate, rotating aminopropyl group on pyrazolium ring; CPK
display shows pyrazolium ring colored by atom type. Stick display shows rotating aminopropyl moiety colored by atom type. Yellow solid display
shows active pocket Connolly surface. The blue part shows the gate of the 3-side chain binding pocket. Pictures (a) and (c) were drawn by Viewer Lite
5.0. Picture (d) was drawn by InsightII 2000.
chain binding pocket. Therefore, among various possi-
ble causes for the stability, we speculated that steric hin-
drance between a cephalosporin and the gate in its entry
into the 3-side chain binding pocket prevented and
inhibited formation of complex A, and this inhibition re-
sulted in the stability against AmpC b-lactamase. In or-
der to validate this speculation, the Connolly surface of
the binding pocket was displayed and the 3D-structure
of the gate was investigated. The shape of the gate was
approximated to a rectangle, and the gate size was mea-
sured by creating manually the rectangle fitted to the
surface in the proximity of the white rectangle shown
in Figure 1c using the Discover/Insight II program.19
The result showed that the gate was a rectangle of
4.5 A · 8.0 A displayed as blue as shown in Figure 1d.
Considering the structure of both FR259647 and the
gate, we reasoned that FR259647 would approach the
gate from the L side, rotating the 4-aminopropyl moiety
on the pyrazolium ring, putting the 3- or 4-substituent
on the ring at the head (Fig. 1d). If FR259647 is bound
to AmpC b-lactamase, the 3-side chain of FR259647 has
to pass through the gate of the 3-side chain binding
pocket. Therefore, we predicted the possibility of entry
into the binding pocket as the probability of entry
through the gate by conformational analysis and at-
tempted to evaluate the relationship between the proba-
bility of entry and the stability to AmpC b-lactamase.
Conformational analysis of the structure of the 3-side
chain in the case of FR259647 was executed for the
rotatable bond s1 ꢀ s3 of the aminopropyl moiety as a
variable using the systematic search module in SYBYL
6.9 (Fig. 2a).20 According to the conditions described
above, we defined that a conformer can pass through
the gate of the 3-side chain binding pocket in the case
that the maximum value (Dmax) of the distance D be-
tween the p-plane of the pyrazolium ring and each atom
of 3- or 4-substituent on the pyrazolium ring including
˚
van der Waals radius is less than the height H (4.5 A)
˚
˚
of the gate as shown in Figure 2b. The probability P(i)
of conformer i amongst all possible conformers of
FR259647 was calculated from total energy of each con-
former using the Maxwell–Boltzmann distribution rule
(Eq. 1), and the sum of the probability P(i) of the con-
formers which can pass through the gate is defined as
the probability of entry of FR259647. As shown in Fig-
ure 2b, in the case that Dmax is less than H, the conform-
ers could pass through the gate to be bound to the active
center. In the case that Dmax is more or equal to H, the

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2265
a
b
R
NH2
N
Dmax
Me
N
NH2
H=4.5
+N
τ3
pyrazolium ring
C
D3
C
8.0
i) in case of binding at active site
max < H
ii) in case of non-binding at activesite
τ2
τ1
D2
D1
π-plane
D
C
v
>
Dmax
H
=
Figure 2. (a) The 3-side chain of FR259647; p-plane shows the p-plane of pyrazolium ring; s1 ꢀ s3 shows free rotatable bond. v shows van der Waals
radius. Circles surrounded around atomic symbol show van der Waals surface. Red dashed lines show distance D1 ꢀ D3 between p-plane of
pyrazolium ring and each atom of the aminopropyl moiety. (b) Condition whereby FR259647 derivatives can pass through the gate from the L side
(see Fig. 1b and d) for binding to active center (Ser90) of AmpC b-lactamase: solid rectangle shows the gate of the 3-side chain binding pocket; H
shows height of the gate; Dmax shows maximum value of distance of each atom of 3- or 4-substituent R on pyrazolium ring from p-plane of
pyrazolium ring including van der Waals radius.
conformers may not pass through the gate. We reasoned
that compounds with a higher probability of the latter
type of conformer amongst all conformers should be
more stable to the enzyme
b-lactamase has been inserted in the location of the
binding for the second strand in PBP.21 Accordingly,
we considered that modification of the substituents at
the 3- or 4-position of the pyrazolium ring of
FR259647, aiming at steric hindrance of the gate, may
generate cephalosporins which are stable to AmpC b-
lactamase, keeping their affinity for PBP. Compounds
4–9 with various lengths of methylene chain at the 3-
or 4-position on the pyrazolium ring were designed, cal-
culated, synthesized, and evaluated. Conformational
analysis of the structure of the 3-side chain of each com-
pound was executed for the rotatable bonds of the ami-
no, aminoalkyl or alkyl moiety at the 3- or 4-position on
the pyrazolium ring as a variable, and the probability of
entry of each compound was calculated using a proce-
dure similar to that employed for the calculation of
FR259647. As shown in Table 2, the probabilities of en-
try of compound 4, bearing an 4-amino group on the
pyrazolium ring, and compound 5 bearing a 4-amino-
methyl group were 100% and 85%, respectively. These
results predicted that compounds 4 and 5 would be la-
bile to AmpC b-lactamase. The DampD/PAO1:PAO1
strain MIC ratios of compounds 4 and 5 were 8, and
indicated that these compounds were labile to AmpC
b-lactamase. On the other hand, the probabilities of en-
try of compound 6 bearing a 4-(4-aminobutyl) group on
the pyrazolium ring, compound 7 bearing a 4-(5-amin-
opentyl) group, and compound 8 bearing a 3-(2-amino-
ethyl) group were 0.71%, 0.43%, and 5.9%, respectively.
These results predicted that compounds 6–8 would be
stable to AmpC b-lactamase. The DampD/PAO1:PAO1
MIC ratios of 6–8 were all 2, indicating that these com-
pounds were stable to AmpC b-lactamase. However,
these designed compounds all have an amino group at
the edge of the 3- or 4-substituent on the pyrazole ring,
it was also considered that FR259647 might be stable
because of electrostatic repulsion between the amino
moiety of the aminoalkyl group on the pyrazolium ring
of FR259647 and the basic group of AmpC b-lactamase.
In order to evaluate the possibility of repulsion, com-
pound 9, without the amino group of 4-(3-aminopropyl)
N
X
PðiÞ ¼ expðꢀEi=RT Þ=
expðꢀEi=RTÞ
ð1Þ
i¼1
P(i): probability of conformer i amongst all possible
conformers of a compound,
N: the total number of conformers,
i: conformer number,
Ei: total energy of conformer i,
R = 0.00199 kcal/mol/K,
T = 298.15 K.
As shown in Table 1, the probabilities of entry of
FR259647, bearing a 4-(3-aminopropyl) group, and
FK518 bearing an 3-amino group were 0.10% and
100%, respectively. From these calculation results, we
proposed FR259647 derivatives with lower probability
of entry would be more stable to AmpC b-lactamase.
2.3. Model validation of structural requirements for the
stability to AmpC b-lactamase using designed compounds
We next attempted to verify the validity of the structural
requirements using designed compounds. Considering
the conformational space of compounds, we speculated
that compounds with a short methylene chain at the 3-
or 4-position on the pyrazolium ring would have a high-
er probability of entry and could be labile to AmpC
b-lactamase, and that compounds with a longer methy-
lene chain on the same position would have a lower
probability of entry and could be stable because of steric
hindrance against the gate. A part of the gate of the 3-
side chain binding pocket of AmpC b-lactamase com-
prises of loop (Ala312-Pro322). PBP, which is a target
enzyme for cephalosporins, and AmpC b-lactamase
are closely related and have a similar 3D-structure, but
such a loop does not exist in PBP.21 The loop of AmpC

2266
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
Table 2. MICs (lg/mL) of compounds 4–9 against P. aeruginosa PAO1 and DampD/PAO1 strain, and probability of entry through the gate of the 3-
side chain binding pocket by conformational analysis
CO2H
O
N
R1
H
N
S
N
N
O
+N
R2
N
H2N
N
S
O
-
CO2 Me
No.
Compounds
R2=
DampD/PAO1a
PAO1a
MIC ratiob
Calculation probability of entry (%)
R1=
4
5
6
7
8
9
–NH2
–NH2
–NH2
–NH2
–NH2
4/0.5
4/0.5
1/0.5
2/1
8
8
2
2
2
2
100
85
–CH2NH2
–(CH2)4NH2
–(CH2)5NH2
–H
0.71
0.43
5.9
–(CH2)2NH2 (H2SO4 salt)
–NH2
1/0.5
2/1
–(CH2)2CH3
2.7
a MIC (lg/mL).
b MIC ratio of DampD/PAO1 to PAO1 strain.
group on the pyrazolium ring from FR259647, was syn-
thesized and evaluated. The probability of entry of com-
pound 9 was 2.7% and this result predicted that
compound 9 would be stable to AmpC b-lactamase. If
compound 9 is labile to AmpC b-lactamase, our model
is not supported. The DampD/PAO1:PAO1 MIC ratio
of 9 was 2 and indicated that this compound was stable
to AmpC b-lactamase. Because both FR259647 and 9
were stable to AmpC b-lactamase, this revealed that
electrostatic repulsion of the amino moiety of the ami-
noalkyl group on the pyrazolium ring of FR259647
did not induce the stability. These results supported
the validity of our proposal and revealed the structural
requirements for stability and that FR259647 deriva-
tives with lower probability of entry were more stable
to AmpC b-lactamase.
3. Chemistry
The preparation of the 3-side chain pyrazoles was per-
formed according to the procedures shown in Scheme
2. 4-tert-Butoxycarbonylamino-1-methyl-5-tritylamino-
pyrazole (11) was prepared by nitrosation of 5-amino-
1-methylpyrazole (10), followed by hydrogenation,
N-tert-butoxycarbonylation, and tritylation. 5-Formyla-
mino-4-formylaminomethyl-1-methylpyrazole (13) was
prepared by reduction of 5-amino-4-cyano-1-methylpy-
razole (12) with LiAlH4, followed by formylation. tert-
Butyl 2-(1-methylpyrazol-5-yl)-ethylcarbamate (15) was
prepared by Horner–Emmons reaction of 1-methylpy-
razole-5-carbaldehyde (14), followed by hydrogenation,
hydrolysis, and Curtius rearrangement. l-Methyl-5-tri-
tylaminopyrazole-4-carbaldehyde (17) was prepared by
tritylation of 5-amino-1-methylpyrazole-4-carbaldehyde
(16). 4-(1-Propyl)-1-methyl-5-tritylaminopyrazole (18)
was prepared by Wittig reaction of 17, followed by
hydrogenation. 4-(3-Hydroxypropyl)-5-tritylamino-1-
methylpyrazole (19) was prepared by Horner–Emmons
reaction of 17, followed by reduction with LiAlH4. 4-
(4-tert-Butoxycarbonylaminobutyl)-5-tritylamino-1-meth-
ylpyrazole (20) was prepared by iodination of 19 and
treatment with NaCN, LiAlH4, and Boc2O. 4-(5-tert-
Butoxycarbonylaminopentyl)-5-tritylamino-1-methylpy-
razole (21) was prepared by Swern oxidation of 19,
followed by Horner–Emmons reaction, reduction with
LiAlH4, and treatment with Boc2O.
Taken together, these studies indicated that the proba-
bility of entry of FR259647 derivatives through the gate
of the 3-side chain binding pocket of AmpC b-lactamase
correlated with the MIC ratio of the mutant strain
DampD/PAO1 to the parent strain PAO1, and that
FR259647 derivatives with lower probability of entry
are more stable to AmpC b-lactamase (Fig. 3).
100
80
Cephalosporins 4–9 were obtained by coupling of pro-
tected cephalosporin derivatives and the pyrazoles
according to the procedures shown in Scheme 3. Ceph-
alosporin 4 was prepared by Vilsmeier acylation of 7-
ACA derivative 2522 and acetic acid derivative 23, which
was obtained by N-tert-butoxycarbonylation of the
known carboxylic acid 22,23 followed by coupling with
pyrazole 11, and subsequent removal of the protecting
groups. Cephalosporin 5 was prepared by coupling of
pyrazole 13 and 7-ACA derivative 26,24 followed by re-
moval of the protecting groups, and acylation with acyl
chloride 24, which was prepared by treatment of 22 with
PCl5, and subsequent removal of the protecting groups.
60
more stable to
AmpC β-lactamase
40
20
0
0
2
4
6
8
10
MIC ratio of ΔampD/ PAO1 to PAO1 strain
Figure 3. Correlation between MIC ratio of DampD/PAO1 to PAO1
strain and probability of entry. FR259647 derivatives with lower
probability of entry are more stable to AmpC b-lactamase.

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2267
Scheme 2. Reagents and conditions: (a) NaNO2, concd HCl, H2O, 5 ꢁC; (b) H2, 10% Pd–C, H2SO4, H2O, rt, 73% in 2 steps; (c) Boc2O, i-Pr2EtN,
CH2Cl2, rt; (d) TrCl, Et3N, CH2Cl2, rt, 23% in 2 steps; (e) LiAlH4, THF, rt; (f) HCO2H, Ac2O, MeOH, rt, 43% in 2 steps; (g) (EtO)2POCH2CO2Et,
NaH, THF, 0 ꢁC to rt, 62%; (h) H2, 10% Pd–C, EtOH, rt, 98%; (i) 1 N NaOH, EtOH, 70 ꢁC, 71%; (j) (PhO)2P(O)N3, Et3N, t-BuOH, reflux, 86%; (k)
TrCl, Et3N, CH2Cl2, rt, 92%; (l) Ph3PEtBr, NaH, DMSO, 70 ꢁC–rt, 36%; (m) H2, 10% Pd–C, EtOAc–THF, rt, 91%; (n) (EtO)2POCH2CO2Et, NaH,
THF, 0 ꢁC–rt, 76%; (o) LiAlH4, THF, rt, 74%; (p) I2, imidazole, Ph3P, THF, rt, 34%; (q) NaCN, NH4Cl, DMF, 70 ꢁC, 83%; (r) LiAlH4, THF, reflux,
90%; (s) Boc2O, CH2Cl2, rt, 51%; (t) (COCl)2, DMSO, i-Pr2EtN, CH2Cl2, ꢀ78 to 0 ꢁC, 25%; (u) (EtO)2POCH2CN, NaH, THF, rt, 74%; (v) LiAlH4,
THF, rt, 99%; (w) Boc2O, CH2Cl2, rt, 28%.
Scheme 3. Reagents and conditions: (a) Boc2O, sodium bis(trimethylsilyl)amide, THF–DMF, 0 ꢁC, 47%; (b) PCl5, CH2Cl2, ꢀ20 to ꢀ10 ꢁC, 76%; (c)
N-(trimethylsilyl)acetamide (MSA), THF, then 23, POCl3, DMF, AcOEt–THF, ꢀ10 to 0 ꢁC, 87%; (d) 11, NaI, DMF–CH2Cl2, rt; (e) TFA, anisole,
CH2Cl2, rt, 3.4% in 2 steps; (f) 13, NaI, DMF–CH2Cl2, rt; (g) TFA, anisole, CH2Cl2, rt; (h) 24, MSA, THF–DMF, 0 ꢁC; (i) concd HCl, MeOH, rt,
2.2% in 4 steps; (j) 24, MSA,THF, 0 ꢁC, 87%; (k) NaI, Aliquat 336, toluene-phosphate buffer, rt, 88%; (l) 20, MSA, DMF, 37 ꢁC; (m) TFA, anisole,
CH2Cl2, rt, 16% in 2 steps; (n) 21, MSA, DMF, 37 ꢁC; (o) TFA, anisole, CH2Cl2, rt, 21% in 2 steps; (p) 15, MSA, DMF, 37 ꢁC; (q) TFA, anisole,
CH2Cl2, rt, 3.8% in 2 steps; (r) 18, MSA, DMF, 37 ꢁC; (s) TFA, anisole, CH2Cl2, rt, 18% in 2 steps.

2268
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
Key intermediate iodomethyl cephalosporin derivative
27 was prepared by acylation of 25 with acyl chloride
24, followed by iodination. Cephalosporins 6–9 were
prepared by coupling of the pyrazoles 20, 21, 15, and
18 with 27, respectively, and subsequent removal of
the protecting groups. The cephalosporin compounds
were then purified by column chromatography by
HPLC and HP-20 to evaluate biological activities.
AmpC b-lactamase (1KVL) by overlapping correspond-
ing Ca atoms using the Discover/Insight II program19
on a Silicon Graphics Workstation Octane2 with
R12000 CPU, respectively. Superimposition of the mod-
eling structure on the template X-ray structure indicated
that root mean square distance between Ca atoms was
˚
0.50 A. The amino acid residues of the active pocket
˚
of AmpC b-lactamase less than 6 A from the bound
cephalothin have 60% identical residues, no insertions,
and no deletions. These results show the validity of the
modeled structure.
4. Conclusions
We have investigated the structural requirements for the
stability of FR259647 derivatives to AmpC b-lactamase
based on 3D-structure. Comparison of the structural
difference between FR259647 and FK518, considering
the 3D-structure of the active pocket of AmpC b-lacta-
mase, led to structural requirements for stability to the
enzyme that steric hindrance of the 3- or 4-substituent
on the pyrazole ring at the 3-position of cephalosporin
nucleus against the gate of the 3-side chain binding
pocket interfered with the entry of FR259647 into the
3-side chain binding pocket. We predicted the possibility
of the entry into the binding pocket as the probability of
entry by conformational analysis. In order to validate
our speculation, novel FR259647 derivatives 4–9 were
designed, calculated, synthesized, and evaluated. As a
result, we demonstrated that their probability of entry
correlated with their MIC ratio of DampD/PAO1 strain
to PAO1 strain as an index of stability and validity of
our model. In conclusion, we propose novel structural
requirements for the stability of FR259647 derivatives
to AmpC b-lactamase that the derivatives with the lower
probability of entry through the gate of the 3-side chain
binding pocket of AmpC b-lactamase are more stable.
Our continuing efforts to generate more potent cephalo-
sporins, which are more stable to AmpC b-lactamase
than FR259647, against P. aeruginosa overproducing
AmpC b-lactamase based on these structural require-
ments will be reported in the future.
5.2. Conformational analysis and calculation method of
probability of entry
Conformational studies of the 3-side chain structures of
FR259647, FK518, and compounds 4–9 were performed
with the systematic search module in SYBYL 6.920 on a
Silicon Graphics Workstation Octane2 with R12000
CPU. Each torsion angle s was rotated in 30ꢁ incre-
ments. Other parameters used default values. Total en-
ergy of each conformer was calculated by systematic
search module, and distance D between the p-plane of
the pyrazolium ring and each atom of the 3- or 4-substi-
tuent on the pyrazolium ring of each conformer was cal-
culated using PLANE in the SYBYL function on a
spread sheet and was added to the van der waals radius
of each atom.20 The probability P(i) of conformer i
amongst all possible conformers of each compound
was calculated from the total energy of each conformer
using the Maxwell–Boltzmann distribution rule (Eq. 1).
The probability of entry of each compound was calcu-
lated as a sum of the probability P(i) of the conformers
which satisfied the following conditions: maximum va-
lue (Dmax) of the distance D is less than the height H
˚
(4.5 A) of the gate. The size of the gate was measured
by creating manually a rectangle fitting to the surface
of the 3D-structure of AmpC b-lactamase using the Dis-
cover/Insight II program.19
5.3. Antibacterial assay against DampD/PAO1 and
PAO1 strains
5. Experimental
It is known that P. aeruginosa PAO1 AmpD-defective
mutant (DampD/PAO1) leads to the overproduction of
AmpC b-lactamase.7,12 Construction of DampD/PAO1
mutant was performed by disrupting ampD of P. aeru-
ginosa PAO1 using a streptomycin-resistant gene cas-
sette according to the literature method.7,13 The
antibacterial activity (MIC) ratio of DampD/PAO1 to
PAO1 was used as an index of stability to AmpC b-lac-
tamase. MICs were determined by an agar dilution
method using Mueller-Hinton agar (Becton–Dickinson,
Tokyo, Japan) and an inoculum of 104 CFU per
spot. Results of susceptibility testing were recorded
according to Clinical and Laboratory Standards Insti-
tute reference method M7-A5, after inoculation at
35 ꢁC for 18 h.25
5.1. Homology modeling
Homology modeling of AmpC b-lactamase from
P. aeruginosa was performed with the FAMS (Full
Automatic Modeling System) program18 using the crys-
tal structure of AmpC b-lactamase from E. coli as a tem-
plate. The amino acid sequences of P. aeruginosa and
E. coli AmpC b-lactamase have 43% identical residues,
two insertion residues, and 10 deletion residues. A mod-
eled structure from P. aeruginosa was obtained using the
crystal structure of complex of mutant AmpC b-lacta-
mase S64G from E. coli with cephalothin (PDB entry
1KVL) as a template.15,16 We used the sequence from
Lys35 to Leu384 as sequence information for P. aerugin-
osa AmpC b-lactamase in accordance with amino acid
sequence of template crystal structures that are visible
in the electron density. In order to compare their struc-
tures, the modeling structure of P. aeruginosa was super-
imposed on the complex crystal structure of E. coli
5.4. Chemistry
5.4.1. General methods. Proton NMR spectra were re-
corded on a Brucker BIOSPIN AVANCE400 or

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2269
DPX200. d Values in ppm relative to tetramethylsilane
are given. IR spectra were recorded with the compound
(neat) on a sodium chloride disk or as KBr pellets or
Nujol suspension using HORIBA FT-710. Mass spectra
were recorded with Hewlett Packard 1100LC/MSD.
High resolution mass spectra were recorded by micro-
mass LCT. Elemental analysis was obtained with Per-
kin-Elmer 2400II.
5.4.3.2. (Z)-2-{5-[(tert-Butoxycarbonyl)amino]-1,2,4-
thiadiazol-3-yl}-2-[(1-tert-butoxycarbonyl-1-methylethoxy)
imino]acetic acid (23). To a solution of 2223 (5.0 g,
15.1 mmol) in a mixed solvent of THF (80 mL) and
DMF (20 mL) was added a solution of sodium bis(tri-
methylsilyl)amide (8.33 g, 45.4 mmol) in THF (12 mL),
and the mixture was stirred for 15 min. To the reaction
mixture was added a solution of di-tert-butyl dicarbon-
ate (3.3 g, 15.1 mmol) in THF (20 mL) under ice-cool-
ing, and the mixture was stirred under ice-cooling for
3 h. To the reaction mixture was added EtOAc, and
the mixture was washed with 10% aqueous KHSO4 solu-
tion and then washed with a phosphate buffer (pH 6.86).
The organic layer was separated, dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The residue
was triturated with IPE, filtered, and dried in vacuo to
give 23 (3.10 g, 47%): 1H NMR (DMSO-d6) d 1.37
(9H, s), 1.45 (6H, s), 1.50 (9H, s), 12.7 (1H, s); IR(KBr)
5.4.2. Starting materials. (Z)-2-(5-Amino-1,2,4-thia-
diazol-3-yl)-[2-(1-tert-butoxycarbonyl-1-methylethoxy)-
imino]acetic acid (22), benzhydryl 7b-amino-3-
chloromethyl-3-cephem-4-carboxylate (25), and benzhy-
dryl 7b-tert-butoxycarbonylamino-3-chloromethyl-3-ce-
phem-4-carboxylate (26) were prepared according to the
procedure described in the literature.22–24
m
max 3192, 2981, 1714, 1551, 1153, 1001 cmꢀ1; MS (ESI):
5.4.3. Synthesis of cephalosporin 4.
m/z 429 (MꢀH)+.
5.4.3.1. 4-tert-Butoxycarbonylamino-1-methyl-5-trityl-
aminopyrazole (11). To a solution of 5-amino-1-meth-
ylpyrazole (10) (100 g, 1.03 mol) in water (700 mL)
were added concd HCl (86 mL) and sodium nitrite
(63.9 g, 927 mmol) in water (200 mL) below 10 ꢁC. The
reaction mixture was stirred for 30 min at 5 ꢁC. The pre-
cipitated solid was filtered off and dried to give 5-amino-
1-methyl-4-nitrosopyrazole (117 g). To a suspension of
this product (117 g, 928 mmol) in water (819 mL) were
added concd H2SO4 (91 g, 928 mmol) and 10% Pd–C
(58 g) at room temperature. The resulting mixture was
hydrogenated under balloon pressure for 10 h at room
temperature. The reaction mixture was filtered and the
resulting filtrate was concentrated in vacuo. To the
resulting suspension was added isopropanol (IPA;
2.3 L) and the mixture was stirred for 1 h. The precipi-
tated solid was filtered and dried to give 4,5-diamino-
5.4.3.3. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-(1-
carboxy-1-methylethoxyimino)acetamido]-3-[3,4-diamino2-
methyl-1-pyrazolio]methyl-3-cephem-4-carboxylate (4). A
mixture of DMF (0.648 mL, 8.37 mmol) and POCl3
(0.781 mL, 8.37 mmol) was stirred at room temperature
for 30 min. To the mixture was added 23 (3 g,
6.98 mmol) in THF (4 mL) at 4 ꢁC, and the reaction
mixture was stirred at room temperature for 1 h. Mean-
while, a mixture of 2522 (3 g, 6.65 mmol) and N-(tri-
methylsilyl)acetamide (8.72 g, 66.4 mmol) in THF
(15 mL) was warmed to make a clear solution. The solu-
tion was then cooled to ꢀ20 ꢁC and added to the acti-
vated acid solution obtained above. The reaction
mixture was stirred at ꢀ10 ꢁC to 0 ꢁC for 1 h and poured
into mixture of EtOAc and water. The aqueous layer
was separated, and the organic layer was washed with
brine, dried over anhydrous MgSO4, and filtered. The
filtrate was concentrated in vacuo and purified by col-
umn chromatography on silica gel eluting with hexane/
EtOAc (3:2) to give benzhydryl 7b-[(Z)-2-(5-tert-butoxy-
carbonylamino-1,2,4-thiadiazol-3-yl)-2-(1-tert-butoxycarbo-
nyl-1-methylethoxyimino)acetamido]-3-chloromethyl-3-
cephem-4-carboxylate (4.79 g, 87%): 1H NMR (DMSO-
d6) d 1.39 (9H, s), 1.48 (6H, s), 1.50 (9H, s), 3.58 (1H, d,
J = 18.3 Hz), 3.76 (1H, d, J = 18.3 Hz), 4.44 (2H, s), 5.29
(1H, d, J = 5.0 Hz), 6.01 (1H, dd, J = 8.6, 5.0 Hz), 6.97
(1H, s), 7.2–7.6 (10H, m), 9.65 (1H, d, J = 5.0 Hz),
12.7 (1H, s); IR(KBr) mmax 2981, 1794, 1720, 1525,
1371, 1247, 1151 cmꢀ1; MS (ESI): m/z 849 (M+Na)+.
1
1-methylpyrazole sulfate (158 g, 73%): H NMR (D2O)
d 3.74 (3H, s), 7.80 (1H, s).
To an ice-cooled solution of 4,5-diamino-1-methylpy-
razole sulfate (30 g, 143 mmol) in CH2Cl2 (300 mL) were
added di-tert-butyl dicarbonate (34.3 g, 157 mmol) and
diisopropylethylamine (49.7 mL, 285 mmol). The solu-
tion was stirred for 4 h at room temperature. To the
resulting mixture was added water and extracted with
EtOAc. The organic layer was washed with water twice
and brine. The extract was dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The residue
was triturated with diisopropyl ether (IPE), filtered, and
dried in vacuo to give 4-tert-butoxycarbonylamino-1-
methyl-5-aminopyrazole (10.8 g). To a solution of this
product (10.5 g, 49.5 mmol) and triethylamine
(34.5 mL, 247 mmol) in CH2Cl2 (150 mL) was added tri-
tyl chloride (16.5 g, 59.4 mmol) at room temperature
and was stirred at room temperature overnight. The
resulting mixture was concentrated in vacuo and the
organic layer was washed with water three times and
brine. The extract was dried over anhydrous MgSO4, fil-
tered, and concentrated in vacuo. The residue was tritu-
rated with IPE, filtered, and dried in vacuo to give 11
(14.2 g, 23%): 1H NMR (DMSO-d6) d 1.35 (9H, s),
2.70 (3H, s), 5.66 (1H, s), 7.00–7.40 (17H, m); MS
(APCI) m/z 455 (M+H)+.
To a solution of benzhydryl 7b-[(Z)-2-(5-tert-butoxycar-
bonylamino-1,2,4-thiadiazol-3-yl)-2-(1-tert-butoxycarbonyl-
1-methylethoxyimino)acetamido]-3-chloromethyl-3-ce-
phem-4-carboxylate (0.65 g, 0.786 mmol) in a mixed sol-
vent of DMF (2 mL) and CH2Cl2 (2 mL) was added
sodium iodide (130 mg, 0.864 mmol), and the mixture
was stirred for 30 min at room temperature. To the reac-
tion mixture was added 11 (714 mg, 1.57 mmol). The
whole mixture was stirred for 36 h at room temperature
and poured into a mixture of EtOAc and water, and ex-
tracted with EtOAc. The organic layer was washed with
water, a mixture of aqueous 10% sodium thiosulfate and

2270
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
brine, brine, and aqueous 10% trifluoroacetic acid so-
dium salt solution, and dried over MgSO4, filtered,
and evaporated in vacuo. The residue was dissolved in
a small amount of EtOAc and poured into IPE. The
resulting precipitates were filtered. The filter cake was
washed with IPE and dried in vacuo. To a solution of
the resulting solid in CH2Cl2 (1.5 mL) were added ani-
sole (0.5 mL) and trifluoroacetic acid (1.0 mL). The
resulting solution was stirred for 2 h at room tempera-
ture and poured into IPE. The resulting precipitate
was collected by filtration and dried in vacuo. The crude
product was purified by preparative HPLC utilizing
ODS. The first eluate containing the desired product
was concentrated to about 30 mL in vacuo. The concen-
trate was adjusted to about pH 3 by addition of concd
HCl and purified by column chromatography on HP-
20 eluting with 20% aqueous IPA. The eluate was con-
centrated to about 30 mL in vacuo and lyophilized to
precipitate was collected by filtration to give 24
(14.8 g, 76%): IR (Nujol) mmax 3430, 3270, 3130, 1815,
1750, 1725, 1640 cmꢀ1
.
5.4.4.3. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-
(1-carboxy-1-methylethoxyimino)acetamido]-3-[3-amino-
4-aminomethyl-2-methyl-1-pyrazolio]methyl-3-cephem-4-
carboxylate (5). To
a
solution of 2624 (3.50 g,
6.80 mmol) in DMF (4.5 mL) was added sodium iodide
(1.02 g, 6.80 mmol), and the mixture was stirred for
30 min at room temperature. To the reaction mixture
was added 13 (3.71 g, 20.4 mmol). The whole mixture
was stirred for 29 h at room temperature and poured
into a mixture of EtOAc and water. The aqueous layer
was separated, and the organic layer was washed with
brine, dried over anhydrous MgSO4, and filtered. The
filtrate was concentrated to about 20 mL in vacuo.
The concentrate was poured into IPE (300 mL), and
the resulting precipitate was collected by filtration and
dried in vacuo. To a solution of the resulting solid in
CH2Cl2 (10.5 mL) were added anisole (3.5 mL) and tri-
fluoroacetic acid (7 mL). The resulting solution was
stirred for 3 h at room temperature and poured into
IPE. The resulting precipitate was collected by filtration
and dried in vacuo to give crude 7b-amino-3-[3-
formamido-4-formamidomethyl-2-methyl-1-pyrazolio]
1
give 4 (15.5 mg, 3.4%): H NMR (D2O) d 1.53 (6H, s),
3.13 and 3.30 (2H, ABq, J = 17.7 Hz), 3.65 (3H, s),
4.50–4.95 (2H, m), 5.20 (1H, d, J = 4.7 Hz), 5.84 (1H,
d, J = 4.7 Hz), 7.57 (1H, s); IR(KBr) mmax 1772 cmꢀ1
;
HRMS (ESI) m/z calcd for C20H25N10O7S2 [M+H]+
581.1349. Found: 581.1350. Anal. Calcd for
C20H24N10O7S2 3.89H2O: C, 36.92; H, 4.92; N, 21.53.
Found: C, 37.32; H, 5.42; N, 21.03.
methyl-3-cephem-4-carboxylate
bistrifluoroacetate
5.4.4. Synthesis of cephalosporin 5
(3.07 g). To a solution of the crude product (3.07 g,
4.93 mmol) and N-(trimethylsilyl)acetamide (6.47 g,
49.3 mmol) in a mixed solvent of DMF (15 mL) and
THF (15 mL) was added 24 (1.9 g, 4.93 mmol) under
ice-cooling. The solution was stirred for 2 h at the same
temperature. The reaction mixture was poured into
EtOAc (300 mL), and the mixture was stirred for
30 min. The resulting precipitate was collected by filtra-
tion, successively washed with EtOAc and IPE, and
dried in vacuo to give a solid (3.28 g). To a suspension
of the resulting solid in CH2Cl2 (9 mL) were added ani-
sole (3 mL) and trifluoroacetic acid (6 mL) under ice-
cooling. The resulting solution was stirred for 3 h at
room temperature and poured into IPE. The resulting
precipitate was collected by filtration and dried in
vacuo to give crude 7b-[(Z)-2-(5-amino-1,2,4-thia-
diazol-3-yl)-2-(1-carboxy-1-methylethoxyimino)acetam-
ido]-3-[3-formamido-4-formamidomethyl-2-methyl-1-
pyrazolio]methyl-3-cephem-4-carboxylate (3.44 g). To a
solution of the crude product (2.5 g) in MeOH (25 mL)
was added concd HCl solution (2.5 mL) at room tem-
perature. The mixture was stirred for 17 h at room tem-
perature. The reaction mixture was adjusted to about
pH 7 by addition of saturated aqueous Na2CO3 solution
and concentrated in vacuo to remove MeOH. The
resulting residue was purified by preparative HPLC uti-
lizing ODS. The eluate containing the desired product
was concentrated to about 30 mL in vacuo. The concen-
trate was adjusted to about pH 1 by addition of concd
HCl and purified by column chromatography on HP-
20 eluting with 20% aqueous IPA. The eluate was con-
centrated in vacuo and lyophilized to give 5 (50.0 mg,
5.4.4.1. 5-Formylamino-4-formylaminomethyl-1-meth-
ylpyrazole (13). To ice-cooled THF (300 mL) was added
LiAlH4 (6.21 g, 164 mmol) under nitrogen gas. To the
ice-cooled mixture was added 5-amino-4-cyano-1-meth-
ylpyrazole (12) (10.0 g, 81.9 mmol) and stirred for 2 h at
room temperature under nitrogen gas. To the ice-cooled
mixture were successively added sodium fluoride (30.0 g)
and water (12 mL). The resulting mixture was filtered
and concentrated in vacuo. The residue was dissolved
in toluene, and the solution was concentrated in vacuo
again to give 5-amino-4-aminomethyl-1-methylpyrazole
(6.34 g). To formic acid (11.5 mL) was added acetic
anhydride (14.2 mL), and the mixture was stirred at
room temperature for 30 min. The resulting mixture
was added to 5-amino-4-aminomethyl-1-methylpyrazole
(6.34 g, 50.2 mmol) in MeOH (19.4 mL) and the mixture
was stirred for 2 h at room temperature. After concen-
tration in vacuo, to the residue was added water. The
mixture was adjusted to pH 9.5 with DIAIONꢂ
SA10A (OHꢀ). The mixture was filtered and concen-
trated in vacuo. The residue was purified by column
chromatography on silica gel eluting with CH2Cl2/
MeOH (3:1) to give 13 (3.97 g, 43%): 1H NMR
(DMSO-d6) d 3.59 (3H, s), 3.98 (2H, d, J = 5.8 Hz),
7.30 (1H, s), 8.01 (1H, t, J = 5.8 Hz), 8.15 (1H, s), 8.39
(1H, s), 9.95 (1H, s) MS (ESI): m/z 205 (M+H)+.
5.4.4.2. (Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-(1-tert-
butoxycarbonyl-1-methylethoxyimino)acetyl chloride hydro-
chloride (24). To a solution of PCl5 (11.1 g, 53.4 mmol)
in CH2Cl2 (168 mL) was added 22 (16.78 g, 50.8 mmol)
at ꢀ20 ꢁC. The resulting mixture was stirred at ꢀ20 to
ꢀ10 ꢁC for 1.5 h, and to the mixture was added drop-
wise IPE (671.2 mL) at ꢀ20 to ꢀ10 ꢁC. The mixture
was stirred under ice-cooling for 1 h and the resulting
1
2.2%) as an amorphous solid: H NMR (D2O) d 1.53
(3H, s), 1.54 (3H, s), 3.20 and 3.48 (2H, ABq,
J = 17.8 Hz), 3.73 (3H, s), 4.08 (2H, s), 4.99 and 5.19
(2H, ABq, J = 15.4 Hz), 5.26 (1H, d, J = 4.9 Hz), 5.86

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2271
(1H, d, J = 4.9 Hz), 8.04 (1H, s); IR(KBr) mmax 3346,
3184, 1770, 1595, 1398 cmꢀ1; HRMS (ESI) m/z calcd
for C21H27N10O7S2 [M+H]+ 595.1506. Found: 595.1505.
Anal. Calcd for C21H26N10O7S2 6H2O: C, 35.89; H,
5.45; N, 19.93. Found: C, 36.03; H, 5.85; N, 19.89.
5.4.5.3. 4-(4-tert-Butoxycarbonylaminobutyl)-5-tri-
tylamino-1-methylpyrazole (20). To a solution of 19
(3.0 g, 7.55 mmol), imidazole (1.03 g, 15.1 mmol), and
triphenylphosphine (3.96 g, 15.1 mmol) in THF
(35 mL) was added iodine (3.83 g, 15.1 mmol) at room
temperature. To the solution was further added iodine
until the color of the solution changed to pale yellow.
To the solution was added saturated aqueous NaHCO3
solution and the mixture was extracted with CHCl3. The
organic layer was dried over anhydrous MgSO4, filtered,
and concentrated in vacuo. The residue was purified by
column chromatography on silica gel eluting with a
mixed solvent of EtOAc and hexane to give 4-(3-iodo-
propyl)-5-tritylamino-1-methylpyrazole (1.31 g, 34%):
1H NMR (DMSO-d6) d 1.35–1.55 (2H, m), 1.85–2.00
(2H, m), 2.76 (3H, s), 2.93 (2H, t, J = 7.1 Hz), 5.74
(1H, s), 7.02 (1H, s), 7.10–7.35 (15H, m); MS (APCI)
m/z 508 (M+H)+, 530 (M+Na)+.
5.4.5. Synthesis of cephalosporin 6
5.4.5.1. 1-Methyl-5-(tritylamino)-pyrazole-4-carbalde-
hyde (17). To a suspension of 5-amino-1-methylpyraz-
ole-4-carbaldehyde (16) (25 g, 200 mmol) and Et3N
(22.2 g, 220 mmol) in CH2Cl2 (500 mL) was added trityl
chloride (61.3 g, 220 mmol) at room temperature. The
mixture was stirred at room temperature for 70 h. The
reaction mixture was washed with 10% aqueous citric
acid solution, 10% aqueous NaHCO3 solution, and
brine. The extract was dried over anhydrous MgSO4, fil-
tered, and concentrated in vacuo. The residue was tritu-
rated with EtOAc, filtered, washed with EtOAc, and
dried in vacuo to give 17 (67.6 g, 92%) as a colorless so-
A solution of 4-(3-iodopropyl)-5-tritylamino-1-methylpy-
razole (1.00 g, 1.97 mmol), sodium cyanide (97.6 mg,
1.99 mmol), and ammonium chloride (21.1 mg, 0.394
mmol) in DMF (5 mL) was stirred at 70 ꢁC for 5 h. After
concentration in vacuo, the residue was extracted with
EtOAc. The organic layer was dried over anhydrous
MgSO4, filtered, and concentrated in vacuo to give 4-(3-
cyanopropyl)-5-tritylamino-1-methylpyrazole (664 mg,
1
lid: H NMR (CDCl3) d 2.84 (3H, s), 7.26–7.34 (15H,
m), 7.60 (1H, s), 8.95 (1H, br s), 9.58 (1H, s); MS (APCI)
m/z 366 (MꢀH)+.
5.4.5.2. 3-(1-Methyl-5-triphenylmethylaminopyrazol-4-
yl) propanol (19). To a suspension of sodium hydride (60%
dispersion in mineral oil, 4.80 g, 120 mmol) in THF
(200 mL) was added dropwise triethyl phosphonoacetate
(4.80 g, 120 mmol) under ice-cooling. The mixture was
stirred under ice-cooling for 1 h. To the reaction mixture
was added 17 (36.7 g, 100 mmol) at room temperature,
and the mixture was stirred at room temperature for
17 h. After concentration in vacuo, the residue was dis-
solved in CHCl3. The solution was washed with 10%
aqueous citric acid solution, 10% aqueous NaCO3 solu-
tion, and brine. The extract was dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The residue
was triturated with EtOAc, filtered, and dried in vacuo to
give ethyl (E)-3-(1-methyl-5-triphenylmethylaminopyrazol-
4-yl)acrylate (33.0 g, 76%) as a colorless solid: 1H NMR
(DMSO-d6) d 1.14 (3H, t, J = 7.0 Hz), 3.04 (3H, s), 3.99
(2H, q, J = 7.0 Hz), 5.79 (1H, d, J = 16.0 Hz), 6.26 (1H,
s), 7.07 (1H, d, J = 16.0 Hz), 7.15–7.35 (15H, m), 7.62
(1H, s); MS (APCI) m/z 460 (M+Na)+.
1
83%): H NMR (CDCl3) d 1.30–1.50 (2H, m), 1.85–2.00
(2H, m), 2.06 (2H, t, J = 7.2 Hz), 2.86 (3H, s), 4.13 (1H,
s), 7.10–7.35 (16H, m); MS (APCI) m/z 429 (M+Na)+.
To ice-cooled THF (13 mL) was added LiAlH4 (185 mg,
4.87 mmol) under nitrogen gas. To the ice-cooled sus-
pension was added 4-(3-cyanopropyl)-5-tritylamino-1-
methylpyrazole (660 mg, 1.62 mmol) and then refluxed
for 5 h. After cooling to room temperature, to the
ice-cooled mixture were successively added sodium fluo-
ride (1.02 g) and water (350 lL). After filtration, the
solution was concentrated in vacuo and to the residue
was added toluene and the solution was concentrated
in vacuo again to give 4-(4-aminobutyl)-5-tritylamino-
1
1-methylpyrazole (602 mg, 90%): H NMR (DMSO-d6)
d 0.90–1.20 (4H, m), 1.70–1.85 (2H, m), 2.36 (2H, t,
J = 6.5 Hz), 2.75 (3H, s), 5.66 (1H, s), 6.98 (1H, s),
7.10–7.35 (15H, m); MS (APCI) m/z 409 (MꢀH)+.
To a suspension of LiAlH4 (5.7 g, 150 mmol) in THF
(200 mL) was added ethyl (E)-3-(1-methyl-5-trip-
henylmethylaminopyrazol-4-yl)acrylate (21.9 g, 50 mmol)
at room temperature under nitrogen gas. The mixture
was stirred at room temperature for 1 h. After cooling
on an ice bath, potassium fluoride (34 g) and water
(10 mL) were added to the reaction mixture. The insoluble
materials were removed by filtration. The filtrate was con-
centrated in vacuo, and the residue was dissolved in
CHCl3. The solution was washed with 10% aqueous citric
acid solution. The organic layer was dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The oily res-
idue was purified by column chromatography on silica gel
eluting with CHCl3 to give 19 (14.8 g, 74%) as a solid:
1H NMR (DMSO-d6) d 1.10–1.30 (2H, m), 1.83 (2H, d,
J = 7.7 Hz), 2.76 (3H, s), 3.05–3.20 (2H, m), 4.20–4.30
(1H, m), 5.67 (1H, s), 6.98 (1H, s), 7.10–7.30 (15H, m);
MS (APCI) m/z 396 (MꢀH)+.
To an ice-cooled solution of 4-(4-aminobutyl)-5-tri-
tylamino-1-methylpyrazole (600 mg, 1.46 mmol) in
CH2Cl2 (5 mL) was added dropwise a solution of di-
tert-butyl dicarbonate (335 mg, 1.53 mmol) in CH2Cl2
(1 mL). After stirring for 3 h at room temperature
and concentrating in vacuo, the residue was extracted
with EtOAc. The organic layer was successively
washed with water and brine and dried over MgSO4.
After filtration and concentration, the residue was
purified by column chromatography on silica gel elut-
ing successively with CH2Cl2 and a mixed solvent
(CH2Cl2/EtOAc = 3:1) to give 20 (383 mg, 51%): 1H
NMR (DMSO-d6) d 0.90–1.20 (4H, m), 1.37 (9H, s),
1.70–1.85 (2H, m), 2.70–2.80 (2H, m), 2.74 (3H, s),
5.66 (1H, s), 6.60–6.75 (1H, m), 6.98 (1H, s), 7.10–
7.35 (15H, m); MS (APCI) m/z 511 (M+H)+, 533
(M+Na)+.

2272
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
5.4.5.4. Benzhydryl 7b-[(Z)-2-(5-amino-1,2,4-thia-
4-(4-aminobutyl)-2-methyl-1-pyrazolio]methyl-3-cephem-
4-carboxylate (6). To solution of 27 (550 g,
diazol-3-yl)-2-(1-tert-butoxycarboxy-1-methylethoxyimi-
no)acetamido]-3-iodomethyl-3-cephem-4-carboxylate (27).
To a solution of 2524 (140 g, 310 mmol) in THF (1.4 L)
was added N-(trimethylsilyl)acetamide (142 g, 1.09 mol)
and the mixture was stirred at about 35 ꢁC for 40 min.
To the mixture was added 24 (143 g, 372 mmol) under
ice-cooling and the reaction mixture was stirred at the
same temperature for 1 h. The resulting reaction mixture
was added to mixture of EtOAc and ice-water. The mix-
ture was adjusted to pH 1.5 with aqueous NaHCO3 solu-
tion, and then the organic layer was separated. To the
organic layer was added aqueous NaHCO3 solution to
adjust to pH 7.5, andthen the organic layer was separated.
The aqueous layer was extracted again with EtOAc and
the combined organic layers were washed successively
with water and brine, dried over MgSO4, and filtered.
The filtrate was concentrated to about 0.8 kg in vacuo.
The concentrate was slowly poured into IPE (7 L), and
the resulting precipitate was collected by filtration,
washed with IPE, and dried in vacuo to give benzhydryl
7b-[(Z)-2-(5-amino-1,2,4-thiadiazol-3-yl)-2-(1-tert-but-
oxycarboxy-1-methylethoxyimino)acetamido]-3-chlo-
romethyl-3-cephem-4-carboxylate (220 g, 98%). This
product was identified by deprotection to the final com-
pound. To a mixture of the product (2.95 g, 4.06 mmol)
in a mixed solvent of CH2Cl2 (6 mL) and anisole
(1.5 mL) was added trifluoroacetic acid (2.95 mL) at 0–
5 ꢁC. After stirring at 0 ꢁC for 2 h, the resulting mixture
was poured into a mixed solvent of IPE (60 mL) and hex-
ane (120 mL) under ice-cooling. The resulting precipitate
was filtered, washed with hexane, and dried over P2O5 to
give 7b-[(Z)-2-(5-amino-1,2,4-thiadiazol-3-yl)-2-(1-tert-
butoxycarboxy-1-methylethoxyimino)acetamido]-3-chloro-
methyl-3-cephem-4-carboxylic acid trifluoroacetic acid
a
0.672 mmol) in DMF (1.7 mL) was added N-(trimethyl-
silyl)acetamide (441 mg, 3.36 mmol), and the mixture
was stirred at room temperature for 1.5 h. To the reac-
tion mixture was added 20 (377 mg, 0.739 mmol) and
the mixture was stirred at 37 ꢁC overnight. After cooling
to room temperature, to the resulting reaction mixture
was added water and the solution extracted with EtOAc.
The organic layer was washed with water, a mixture of
aqueous 10% sodium thiosulfate solution and brine,
brine, and aqueous 10% sodium trifluoroacetate solu-
tion, dried over MgSO4, and filtered. The filtrate was
concentrated to about 3 mL in vacuo. The concentrate
was poured into IPE, and the resulting precipitate was
collected by filtration and dried in vacuo. To a solution
of the resulting solid in CH2Cl2 (1.8 mL) were added
anisole (0.6 mL) and trifluoroacetic acid (1.8 mL), and
the mixture was stirred at room temperature for 2 h.
The reaction mixture was poured into IPE, and the
resulting precipitate was collected by filtration and dried
in vacuo to give a crude product, which was purified by
preparative HPLC utilizing ODS. The eluate containing
desired product was concentrated to about 30 mL in va-
cuo. The concentrate was adjusted to about pH 3 with
concd HCl and purified by column chromatography
on HP-20 eluting with 30% aqueous IPA. The eluate
was concentrated to about 30 mL in vacuo and lyophi-
1
lized to give 6 (69 mg, 16%): H NMR (D2O) d 1.53
(6H, s), 1.40–1.80 (4H, m), 2.30–2.50 (2H, m), 2.90–
3.10 (2H, m), 3.16 and 3.37 (2H, ABq, J = 17.7 Hz),
3.68 (3H, s), 4.91 and 5.18 (2H, ABq, J = 15.8 Hz),
5.23 (1H, d, J = 4.7 Hz), 5.84 (1H, d, J = 4.7 Hz), 7.74
(1H, s); IR(KBr) mmax 1772 cmꢀ1; HRMS (ESI) m/z calcd
for C24H33N10O7S2 [M+H]+ 637.1975. Found: 637.1976.
Anal. Calcd for C24H32N10O7S2 5.8H2O: C, 38.89; H,
5.93; N, 18.90. Found: C, 38.93; H, 6.04; N, 18.76.
1
salt (2.41 g, 88%): H NMR (DMSO-d6) d 1.39 (9H, s),
1.40 (6H, s), 3.47, 3.75 (2H, ABq, J = 18 Hz), 4.55 (2H,
s), 5.20 (1H, d, J = 5 Hz), 5.85 (1H, dd, J = 5 Hz, 8 Hz),
9.48 (1H, d, J = 8 Hz).
5.4.6. Synthesis of cephalosporin 7
5.4.6.1. 4-(5-tert-Butoxycarbonylaminopentyl)-5-tri-
tylamino-1-methylpyrazole (21). To a solution of oxalyl
chloride (857 lL, 9.82 mmol) in CH2Cl2 (15 mL) was
added DMSO (2.14 mL, 30.2 mmol) at ꢀ78ꢁC under a
nitrogen gas flow and the solution was stirred at the
same temperature for 40 min. To the solution was added
19 (3.00 g, 7.55 mmol) in CH2Cl2 (5 mL) and stirred for
1 h. To the solution was added diisopropylethylamine
(6.58 mL, 37.8 mmol) and gradually increased the tem-
perature to 0 ꢁC with stirring for 1 h. To the mixture
was added water and the solution extracted with EtOAc.
The organic layer was successively washed with water
and brine. The extract was dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The residue
was purified by silica gel column chromatography
eluting successively with hexane and a mixed solvent
(hexane/EtOAc = 1:1) to give 4-(2-formylethyl)-5-tri-
tylamino-1-methylpyrazole (754 mg, 25%): 1H NMR
(DMSO-d6) d 1.95–2.20 (4H, m), 2.73 (3H, s), 5.77
(1H, s), 7.00 (1H, s), 7.10–7.30 (15H, m), 9.39 (1H, s);
MS (APCI) m/z 394 (MꢀH)+.
To a solution of benzhydryl 7b-[(Z)-2-(5-amino-1,2,
4-thiadiazol-3-yl)-2-(1-tert-butoxycarboxy-1-methyleth-
oxyimino)acetamido]-3-chloromethyl-3-cephem-4-carbox-
ylate (60 g, 82.5 mmol) in toluene (600 mL) were added
a solution of sodium iodide (61.8 g, 412 mmol) in
0.05 mol phosphate buffer (pH 7, 500 mL) and trica-
prylylmethylammonium chloride (Aliquat 336) (6.67 g,
16.5 mmol). The mixture was stirred at room tempera-
ture for 15 h. The reaction mixture was added to a mix-
ture of EtOAc and water. The organic layer was washed
with water and brine, and then dried over MgSO4. The
MgSO4 was filtered off, and the filtrate was evaporated
to 255 g under reduced pressure. The concentrate was
poured into IPE (2 L). The resulting precipitate was col-
lected by filtration and dried to give 27 (59.4 g, 88%): 1H
NMR (DMSO-d6) d 1.39 (9H, s), 1.46 (6H, s), 3.57 and
3.87 (2H, ABq, J = 18.0 Hz), 4.30 (2H, br s), 5.25 (1H,
d, J = 4.9 Hz), 5.94 (1H, dd, J = 4.8, 8.7 Hz), 6.95 (1H,
br s), 7.15–7.60 (10H, m), 8.17 (2H, br s), 9.53 (1H, d,
J = 8.7 Hz).
5.4.5.5. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-
(1-carboxy-1-methylethoxyimino)acetamido]-3-[3-amino-
To an ice-cooled solution of diethyl cyanomethylphosph-
onate (806 mg, 4.55 mmol) in THF (15 mL) was added so-

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2273
dium hydride (60% dispersion in mineral oil, 182 mg,
4.55 mmol) under flowing nitrogen gas and then stirred
for 1 h at room temperature. To the ice-cooled reaction
mixture was added 4-(2-formylethyl)-5-tritylamino-1-
methylpyrazole (1.50 g, 3.79 mmol) and then stirred at
room temperature overnight. To the mixture was added
water and the solution extracted with EtOAc. The organic
layer was successively washed with water and brine. The
extract was dried over anhydrous MgSO4, filtered, and
concentrated in vacuoto give 4-(4-cyanobut-3-enyl)-5-tri-
tylamino-1-methylpyrazole (1.18 g, 74%): 1H NMR
(DMSO-d6) d 1.90–2.10 (4H, m), 2.74 and 2.76 (3H, s),
5.43 and 5.56 (1H, d, J = 16.6 Hz and 11.0 Hz) 5.76 and
5.77 (1H, s), 6.28 and 6.46 (1H, d, J = 11.0 Hz and
16.6 Hz) 7.02 and 7.04 (1H, s), 7.10–7.35 (15H, m); MS
(APCI) m/z 417 (MꢀH)+.
(26.9 g, 120 mmol) in THF (200 mL) was added sodium
hydride (60% dispersion in mineral oil, 4.80 g,
120 mmol) portionwise under ice-cooling. The mixture
was stirred for 1 h at the same temperature. To the reac-
tion mixture was added 1-methyl-1H-pyrazole-5-carbal-
dehyde (14) (33.0 g, 300 mmol) in THF (165 mL) at
room temperature and the mixture was stirred for
1.5 h at the same temperature. To the resulting solution
was added 10% aqueous KHSO4 solution. It was then
extracted with EtOAc twice. The combined organic lay-
ers were successively washed with saturated aqueous
NaHCO3 solution and brine. The extract was dried over
anhydrous MgSO4, filtered, and concentrated in vacuo.
The residue was purified by silica gel column chroma-
tography eluting with a mixed solvent (hexane/
EtOAc = 95:5 to 2:1) to give ethyl (2E)-3-(1-methyl-pyr-
1
azol-5-yl)-2-propenoate (33.4 g, 62%) as oil: H NMR
To ice-cooled THF (23 mL) was added LiAlH4 (313 mg,
8.24 mmol) under flowing nitrogen gas. To the ice-cooled
suspension was added 4-(4-cyanobut-3-enyl)-5-tritylami-
no-1-methylpyrazole (1.15 g, 2.75 mmol) and the tempera-
ture gradually increased toroom temperature withstirring.
The mixture was stirred for 2 h under the same conditions.
Totheice-cooledresultingmixtureweresuccessivelyadded
sodium fluoride (1.73 g) and water (594 lL). After filtra-
tion, the solution was evaporated in vacuo to give 4-(5-
(DMSO-d6) d 1.26 (3H, t, J = 7.1 Hz), 3.88 (3H, s),
4.20 (2H, q, J = 7.1 Hz), 6.54 (1H, d, J = 15.8 Hz),
6.88 (1H, d, J = 1.8 Hz), 7.45 (1H, d, J = 1.8 Hz), 7.60
(1H, d, J = 15.8 Hz); IR(neat) mmax 2981, 2943, 1713,
1703, 1180, 781 cmꢀ1; MS (APCI) m/z 181 (M+H)+.
A solution of ethyl (2E)-3-(1-methyl-pyrazol-5-yl)-2-pro-
penoate (33.4 g, 185 mmol) in EtOH (500 mL) was
treated with 10% Pd–C (6.60 g) under a hydrogen atmo-
sphere for 2 h at room temperature. After the catalyst
was filtered off, the filtrate was concentrated in vacuo to
give ethyl 3-(1-methyl-pyrazol-5-yl)propanoate (33.0 g,
aminopentyl)-5-tritylamino-1-methylpyrazole
(1.15 g,
1
99%): H NMR (DMSO-d6) d 0.80–1.30 (6H, m), 1.70–
1.90 (2H, m), 2.43 (2H, t, J = 6.9 Hz), 2.75 (3H, s), 5.67
(1H, s), 6.98 (1H, s), 7.05–7.40 (15H, m).
1
98%) as a colorless oil: H NMR (DMSO-d6) d 1.17
(3H, t, J = 7.1 Hz), 2.64 (2H, t, J = 7.3 Hz), 2.86 (2H, t,
J = 7.3 Hz), 3.33 (3H, s), 4.07 (2H, q, J = 7.1 Hz), 6.01
(1H, d, J = 1.7 Hz), 7.26 (1H, d, J = 1.7 Hz); IR(neat)
To asolution of 4-(5-aminopentyl)-5-tritylamino-1-meth-
ylpyrazole (1.10 g, 2.59 mmol) in CH2Cl2 (11 mL) was
added di-tert-butyl dicarbonate (622 mg, 2.85 mmol) at
room temperature and stirred at the same temperature
overnight. To the reaction mixture was added saturated
aqueous NaHCO3 solution and extracted with EtOAc.
The organic layer was successively washed with water
and brine. The extract was dried over anhydrous MgSO4,
filtered, and concentrated in vacuo to give 21 (383 mg,
m
max 2981, 2941, 1733, 1541, 1398, 1182 cmꢀ1; MS (APCI)
m/z 183 (M+H)+.
To a solution of ethyl 3-(1-methyl-pyrazol-5-yl)propan-
oate (33.0 g, 181 mmol) in EtOH (330 mL) was added
1 N aqueous NaOH solution (362mL), and the mixture
was stirred at 70 ꢁC for 1 h. After cooling to room tem-
perature, EtOH was evaporated off. The solution was
extracted with Et2O. The aqueous layer was acidified
to pH 3 using aqueous 1 N HCl solution and the mix-
ture was extracted with EtOAc four times. The com-
bined organic layers were dried over anhydrous
MgSO4, filtered, and concentrated in vacuo. The residue
was triturated with IPE, filtered, and dried in vacuo to
give 3-(1-methyl-pyrazol-5-yl)propanoic acid (20.1 g,
1
28%): H NMR (DMSO-d6) d 0.85–1.05 (4H, m), 1.10–
1.30 (2H, m), 1.38 (9H, s), 1.70–1.85 (2H, m), 2.74 (3H, s),
2.75–2.90 (2H, m), 5.66 (1H, s), 6.65–6.80 (1H, m), 6.98
(1H, s), 7.10–7.35 (15H, m); MS (APCI) m/z 525 (M+H)+.
5.4.6.2. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-
(1-carboxy-1-methylethoxyimino)acetamido]-3-[3-amino-
4-(5-aminopentyl)-2-methyl-1-pyrazolio]methyl-3-cephem-
4-carboxylate (7). This compound was prepared from 21
and 27 using a procedure similar to that employed for
1
71%) as a colorless solid: H NMR (DMSO-d6) d 2.56
(2H, t, J = 7.7 Hz), 2.82 (2H, t, J = 7.7 Hz), 3.73 (3H,
s), 6.01 (1H, d, J = 1.7 Hz), 7.26 (1H, d, J = 1.7 Hz),
11.90–12.70 (1H, br s); IR(KBr) mmax 1718, 1302, 1190,
1009, 945, 802, 633 cmꢀ1; MS (APCI) m/z 155 (M+H)+.
1
the preparation of 6 (90 mg, 21%): H NMR (D2O) d
1.25–1.80 (6H, m), 1.53 (3H, s), 1.54 (3H, s), 2.40 (2H, t,
J = 7.2 Hz), 2.98 (2H, t, J = 7.5 Hz), 3.14 and 3.34 (2H,
ABq, J = 17.6 Hz), 3.67 (3H, s), 4.80–5.00 (2H, m), 5.23
(1H, d, J = 4.9 Hz), 5.84 (1H, d, J = 4.9 Hz), 7.73 (1H,
s); IR(KBr) mmax 1772 cmꢀ1; HRMS (ESI) m/z calcd for
C25H35N10O7S2 [M+H]+ 651.2132. Found: 651.2126.
Anal. Calcd for C25H34N10O7S2 5.5H2O: C, 40.05; H,
6.05; N, 18.68. Found: C, 40.41; H, 6.00; N, 18.20.
To a solution of 3-(1-methyl-pyrazol-5-yl)propanoic
acid (10.0 g, 64.9 mmol) in t-BuOH (200 mL) were
added diphenylphosphoryl azide (16.8 mL, 77.8 mmol)
and triethylamine (10.8 mL, 77.8 mmol). The mixture
was refluxed for 7 h. After cooling to room temperature,
the reaction mixture was diluted with EtOAc, and the
solution was successively washed with saturated aque-
ous NaHCO3 solution, water, and brine. The extract
was dried over anhydrous MgSO4, filtered, and concen-
5.4.7. Synthesis of cephalosporin 8
5.4.7.1. tert-Butyl 2-(1-methyl-pyrazol-5-yl)ethylcar-
bamate (15). To a solution of triethyl phosphonoacetate

2274
K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
trated in vacuo. The residue was purified by silica gel
column chromatography eluting with a mixed solvent
(CH2Cl2/EtOAc = 9:1 to 1:1) to give 15 (12.6 g, 86%)
MgSO4, filtered, and concentrated in vacuo to give 4-
(1-propenyl)-1-methyl-5-tritylaminopyrazole as mixture
of Z: E = ca. 6:1 (1.12 g, 36%): the ratio of E/Z was
determined by the ratio of the integral of the methyl
group in the NMR spectrum. Major product; 1H
NMR (DMSO-d6) d 1.60 (3H, dd, J = 7.0, 1.5 Hz),
2.91 (3H, s), 5.0 (1H, dd, J = 11.4, 7.0 Hz), 5.67 (1H,
dd, J = 11.4, 1.5 Hz), 5.83 (1H, s), 6.44 (1H, s), 7.15–
1
as an oil: H NMR (DMSO-d6) d 1.37 (9H, s), 2.72
(2H, t, J = 7.0 Hz), 3.15 (2H, dt, J = 5.5, 7.0 Hz), 3.71
(3H, s), 6.04 (1H, d, J = 1.7 Hz), 6.96 (1H, t,
J = 5.5 Hz), 7.27 (1H, d, J = 1.7 Hz); IR(KBr) mmax
1701, 1541, 1275, 1250, 777 cmꢀ1; MS (ESI) m/z 226.4
(M+H)+.
1
7.35 (15H, m); Minor product; H NMR (DMSO-d6)
d1.44 (3H, d, J = 5.2 Hz), 2.81 (3H, s), 4.9–5.1 (1H,
m), 5.5–5.7 (1H, m), 5.83 (1H, s), 6.44 (1H, s), 7.15–
7.35 (15H, m); mixture; MS (APCI) m/z 378 (MꢀH)+.
5.4.7.2. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-(1-
carboxy-1-methylethoxyimino)acetamido]-3-[3-(2-amino-
ethyl)-2-methyl-1-pyrazolio]methyl-3-cephem-4-carboxyl-
ate hydrogen sulfate (8). To a solution of 27 (2.5 g,
3.44 mmol) in DMF (5.0 L) was added N-(trimethyl-
silyl)acetamide (2.26 g, 17.2 mmol). After stirring at room
temperature for 0.5 h, 15 (968 mg, 4.3 mmol) was added.
Stirring was continued for 24 h at 37 ꢁC. The resulting
mixture was poured into water and extracted with EtOAc.
The organic layer was successively washed with water,
aqueous 10% sodium thiosulfate, brine, and aqueous
10% trifluoroacetic acid sodium salt, and dried over
MgSO4, filtered, and evaporated in vacuo. The residue
was dissolved in a small amount of EtOAc and added to
IPE dropwise. The precipitates were filtered, and the filter
cake washed with IPE and dried under vacuum. To a solu-
tion of the resulting solid in CH2Cl2 (3.6 mL) were added
anisole (1.8 mL) and trifluoroacetic acid (5.4 mL). The
resulting solution was stirred for 4 h at room temperature
and poured into IPE. The resulting precipitate was col-
lected by filtration and dried in vacuo. The crude product
was purified by preparative HPLC utilizing ODS. The
first eluate containing the desired product was concen-
trated to about 30 mL in vacuo. The concentrate was ad-
justed to about pH 3 by addition of concd HCl and
purified by column chromatography on HP-20 eluting
with 30% aqueous IPA. The eluate was concentrated to
about 30 mL in vacuo, 2 M aqueous H2SO4 (80 lL,
0.16 mmol) was added, and the solution lyophilized to give
8 (90 mg, 3.8%): 1H NMR (D2O) d 1.61 (6H, s), 3.26 and
3.56 (2H, ABq, J = 17.9 Hz), 3.27 (2H, t, J = 7.4 Hz),
3.43 (2H, t, J = 7.4 Hz), 4.01 (3H, s), 5.29 (1H, d,
J = 4.8 Hz), 5.33 and 5.46 (2H, ABq, J = 15.5 Hz), 5.90
(1H, d, J = 4.8 Hz), 6.79 (1H, d, J = 3.1 Hz), 8.21 (1H, d,
J = 3.1 Hz); IR(KBr) mmax 1781, 1716, 1676, 1633,
1153 cmꢀ1; HRMS (ESI) m/z calcd for C22H28N9O7S2
[M+H]+ 594.1553. Found: 594.1554. Anal. Calcd for
C22H27N9O7S2ÆH2O4S 5.5H2O: C, 33.42; H, 5.10; N,
15.94. Found: C, 33.32; H, 4.71; N, 15.61.
A mixture of 4-(1-propenyl)-1-methyl-5-tritylaminopy-
razole (1.00 g, 2.64 mmol) and 10% Pd–C (500 mg) in
a mixed solvent of EtOAc (5 mL) and THF (1 mL)
was stirred under a H2 atmosphere for 10 min at room
temperature. After the catalyst was filtered off, filtrate
was concentrated in vacuo to give 18 (911 mg, 91%):
1H NMR (DMSO-d6) d 0.62 (3H, t, J = 7.2 Hz), 0.9–
1.1 (2H, m), 1.79 (2H, t, J = 7.8 Hz), 2.75 (3H, s), 5.67
(1H, s), 6.44 (1H, s), 7.10–7.35 (15H, m); MS (APCI)
m/z 380 (MꢀH)+.
5.4.8.2. 7b-[(Z)-2-(5-Amino-1,2,4-thiadiazol-3-yl)-2-(1-
carboxy-1-methylethoxyimino)acetamido]-3-[3-amino-2-
methyl-4-(1-propyl)-1-pyrazolio]methyl-3-cephem-4-car-
boxylate (9). This compound was prepared from 18 and
27 using a procedure similar to that employed for the
preparation of 6 (203 mg, 18%): 1H NMR (D2O) d 0.89
(3H, t, J = 7.2 Hz), 1.56 (6H, s), 1.45–1.65 (2H, m),
2.34 (2H, t, J = 7.3 Hz), 3.12 and 3.27 (2H, ABq,
J = 17.6 Hz), 3.64 (3H, s), 4.6–4.8 and 4.92 (2H, ABq,
J = 16.0 Hz), 5.20 (1H, d, J = 4.8 Hz), 5.84 (1H, d,
J = 4.8 Hz), 7.74 (1H, s); IR (KBr) mmax 1780,
1774 cmꢀ1; HRMS (ESI) m/z calcd for C23H30N9O7S2
[M+H]+ 608.1710. Found: 608.1703. Anal. Calcd for
C23H29N9O7S2 3.7H2O: C, 40.97; H, 5.44; N, 18.69.
Found: C, 40.95; H, 5.26; N, 18.37.
Acknowledgments
We are greatly indebted to Prof. Akito Tanaka, Depart-
ment of Pharmacy, School of Pharmacy, Hyogo Univer-
sity of Health Sciences, and Dr. David Barrett,
Chemistry Research Labs, Astellas Pharma Inc., for their
help in the preparation of this manuscript.
5.4.8. Synthesis of cephalosporin 9
References and notes
5.4.8.1. 4-(2-Propyl)-1-methyl-5-tritylaminopyrazole
(18). A suspension of sodium hydride (60% dispersion
in mineral oil, 392 mg, 16.3 mmol) in DMSO (15 mL)
was stirred at 70 ꢁC for 1 h. After cooling to room tem-
perature, to the suspension was added ethyltriphenyl-
phosphonium bromide (6.37 g, 17.1 mmol) and stirred
at room temperature for 1 h. To the mixture was added
17 (3.00 g, 8.16 mmol) in THF (10 mL) and stirred at
room temperature for 2 h. To the resulting mixture
was added water and the mixture was extracted with
EtOAc. The organic layer was successively washed with
water and brine. The extract was dried over anhydrous
1. Stover, C. K.; Pham, X. Q.; Erwin, A. L.; Mizoguchi, S.
D.; Warrener, P.; Hickey, M. J.; Brinkman, F. S. L.;
Hufnagle, W. O.; Kowalik, D. J.; Lagrou, M.; Garber, R.
L.; Goltry, L.; Tolentino, E.; Westbrock-Wadman, S.;
Yuan, Y.; Brody, L. L.; Coulter, S. N.; Folger, K. R.; Kas,
A.; Larbig, K.; Lim, R.; Smith, K.; Spencer, D.; Wong, G.
K.-S.; Wu, Z.; Paulsen, I. T.; Reizer, J.; Saier, M. H.;
Hancock, R. E. W.; Lory, S.; Olson, M. V. Nature 2000,
406, 959–964.
´
´
´
2. Juan, C.; Macia, M. D.; Gutierrez, O.; Vidal, C.; Perez, J.
L.; Oliver, A. Antimicrob. Agents Chemother. 2005, 49,
4733–4738.

K. Murano et al. / Bioorg. Med. Chem. 16 (2008) 2261–2275
2275
3. Livermore, D. M. J. Antimicrob. Chemother. 2001, 47,
247–250.
4. Walsh, C. Nature 2000, 406, 775–781.
5. Kato, K.; Iwai, S.; Kumasaka, K.; Horikoshi, A.; Inada,
S.; Inamatsu, T.; Ono, Y.; Nishiya, H.; Hanatani, Y.;
Narita, T.; Sekino, H.; Hayashi, I. J. Infect. Chemother.
2001, 7, 258–262.
6. Nordmann, P.; Guibert, M. J. Antimicrob. Chemother.
1998, 42, 128–132.
7. Langaee, T. Y.; Gagnon, L.; Huletsky, A. Antimicrob.
Agents Chemother. 2000, 44, 583–589.
8. Powers, R. A.; Caselli, E.; Focia, P. J.; Prati, F.; Shoichet,
B. K. Biochemistry 2001, 40, 9207–9214.
9. De Champs, C.; Poirel, L.; Bonnet, R.; Sirot, D.; Chanal,
C.; Sirot, J.; Nordmann, P. Antimicrob. Agents Chemother.
2002, 46, 3031–3034.
10. (a) Sakane, K.; Kawabata, K.; Inamoto, Y.; Okuda, S.;
Kamimura, T.; Takaya, T. Abstracts of Papers, 30th
Interscience Conference on Antimicrobial Agents and
Chemotherapy, Atlanta, GA, October 21–24, 1990, Abs.
454.; (b) Mine, Y.; Sakamoto, H.; Kamimura, T.;
Watanabe, Y.; Matsumoto, Y.; Kuwahara, S. Abstracts
of Papers, 30th Interscience Conference on Antimicrobial
Agents and Chemotherapy, Atlanta, GA, October 21–24,
1990, Abs. 455.; (c) Mine, Y.; Sakamoto, H.; Kamimura,
T.; Watanabe, Y.; Tawara, S.; Kuwahara, S. Abstracts of
Papers, 30th Interscience Conference on Antimicrobial
Agents and Chemotherapy, Atlanta, GA, October 21–24,
1990, Abs. 456.; (d) Sakamoto, H.; Mine, Y.; Kamimura,
T.; Hirose, T.; Hatano, K.; Kuwahara, S. Abstracts of
Papers, 30th Interscience Conference on Antimicrobial
Agents and Chemotherapy, Atlanta, GA, October 21–24,
1990, Abs. 457.
the signal molecule for induction of AmpC b-lactamase
expression.
13. Takeda, S.; Ishii, Y.; Hatano, K.; Tateda, K.; Yamaguchi,
K. Int. J. Antimicrob. Agents 2007, 30, 443–445.
14. Fisher, J. F.; Meroueh, S. O.; Mobashery, S. Chem. Rev.
2005, 105, 395–424.
15. Beadle, B. M.; Trehan, I.; Focia, P. J.; Shoichet, B. K.
Structure 2002, 10, 413–424.
16. Because cephalothin is easily hydrolyzed by wild-type
AmpC b-lactamase to eliminate the 3-side chain, a mutant
AmpC b-lactamase S64G which substitutes glycine for
serine in active center was used in the complex.
´
17. (a) Weston, G. S.; Blazquez, J.; Baquero, F.; Shoichet, B.
K. J. Med. Chem. 1998, 41, 4577–4586; (b) Tondi, D.;
´
Powers, R. A.; Caselli, E.; Negri, M.-C.; Blazquez, J.;
Costi, M. P.; Shoichet, B. K. Chem. Biol. 2001, 8, 593–610;
(c) Powers, R. A.; Shoichet, B. K. J. Med. Chem. 2002, 45,
3222–3234; (d) Powers, R. A.; Morandi, F.; Shoichet, B.
K. Structure 2002, 10, 1013–1023; (e) Morandi, F.; Caselli,
´
E.; Morandi, S.; Focia, P. J.; Blazquez, J.; Shoichet, B. K.;
Prati, F. J. Am. Chem. Soc. 2003, 125, 685–695; (f) Tondi,
D.; Morandi, F.; Bonnet, R.; Costi, M. P.; Shoichet, B. K.
J. Am. Chem. Soc. 2005, 127, 4632–4639.
18. Ogata, K.; Umeyama, H. J. Mol. Graph. Model 2000, 18,
258–272.
19. The Discover program (InsightII 2000, Accelrys Inc.).
20. SYBYL 6.9 Tripos Inc., 1699 South Hanley Road., Street.
Louis, Missouri, 63144, USA.
21. Massova, I.; Mobashery, S. Antimicrob. Agents Chemo-
ther. 1998, 42, 1–17.
22. Takaya, T.; Takasugi, H.; Masugi, T.; Yamanaka, H.;
Kawabata, K. Eur. Pat. Appl. EP30630-A2, 1981; . Chem.
Abstr. 1982, 96, 181061.
11. Toda, A.; Ohki, H.; Yamanaka, T.; Murano, K.; Okuda,
S.; Kawabata, K.; Hatano, K.; Matsuda, K.; Misumi, K.;
Itoh, K.; Satoh, K.; Inoue, S. Abstracts of Papers, 46th
Interscience Conference on Antimicrobial Agents and
Chemotherapy, San Francisco, CA, September 17–20,
2006, Abs. F1-0240.
23. Teraji, T.; Sakane, K.; Goto, J. Eur. Pat. Appl. EP13762-
B1, 1980; . Chem. Abstr. 1981, 94, 30773.
24. Albrecht, H. A.; Beskid, G.; Christenson, J. G.; Deitcher,
K. H.; Georgopapadakou, N. H.; Keith, D. D.; Konzel-
mann, F. M.; Pruess, D. L.; Chen Wei, C. J. Med. Chem.
1994, 37, 400–407.
12. DampD/PAO1 strain overproduces AmpC b-lactamase by
cytoplasmic accumulation of 1,6-anhydromuropeptide
because ampD encodes an N-acetyl-anhydromuramyl-L-
alanine amidase which cleaves 1,6-anhydromuropeptide,
25. Methods for dilution antimicrobial susceptibility tests for
bacteria that grow aerobically. Approved standard M7-
A5; National Committee for Clinical Laboratory Stan-
dards: Wayne, PA, 2000.