Comparative study of the inhibition of metallo-β-lactamases (IMP-1 and VIM-2) by thiol compounds that contain a hydrophobic group

Article abstract of DOI:10.1248/bpb.27.851

For the purpose of screening of inhibitors that are effective for wide range of metallo-β-lactamases, the inhibitory effect of two series of compounds, 2-ω-phenylalkyl-3-mercaptopropionic acid (PhenylCnSH (n=1-4)) and N-[(7-chloro-quinolin-4-ylamino)-alkyl]-3-mercapto-propionamide (QuinolineCnSH (n=2-6)), where n denotes the alkyl chain length, on metallo-β-lactamases IMP-1 and VIM-2 was examined. These inhibitors contain a thiol group and a hydrophobic group linked by variable-length methylene chain. PhenylCnSH (n=1-4) was found to be a potent inhibitor of both IMP-1 and VIM-2. PhenylC4SH was the potent inhibitor of both IMP-1 (IC₅₀=1.2 μM) and VIM-2 (IC₅₀=1.1 μM) among this study. When the number of methylene units was varied, QuinolineC4SH showed the maximum inhibitory activity against IMP-1 and VIM-2 (IC₅₀=2.5 μM and IC ₅₀=2.4 μM). The relationship between the inhibitory effect of the alkyl chain length was different for both series of inhibitors, suggesting that IMP-1 has a tighter binding site than VIM-2. QuinolineCnSH did not serve as a fluorescence reagent for metallo-β-lactamases.

Full text of DOI:10.1248/bpb.27.851

June 2004
Biol. Pharm. Bull. 27(6) 851—856 (2004)
851
Comparative Study of the Inhibition of Metallo-b-Lactamases (IMP-1 and
VIM-2) by Thiol Compounds That Contain a Hydrophobic Group
a
b
a
a
a
Wanchun JIN, Yoshichika ARAKAWA, Hisami YASUZAWA, Tomoko TAKI, Ryo HASHIGUCHI,
a
a
a
a
Kana MITSUTANI, Asumi SHOGA, Yoshihiro YAMAGUCHI, Hiromasa KUROSAKI,
Naohiro SHIBATA, Michio OHTA, and Masafumi GOTO*
b
c
,a
a
Graduate School of Pharmaceutical Sciences, Kumamoto University; 5–1 Oe Honmachi, Kumamoto 862–0973, Japan:
b
Department of Bacterial Pathogenesis and Infection Control, National Institute of Infectious Diseases; 4–7–1 Gakuen,
c
Musashi-Murayama, Tokyo 208–0011, Japan: and Department of Bacteriology, Nagoya University School of Medicine;
6
5 Tsurumai-cho, Showa-ku, Nagoya 466–8550, Japan. Received January 14, 2004; accepted March 9, 2004
For the purpose of screening of inhibitors that are effective for wide range of metallo-b-lactamases, the in-
hibitory effect of two series of compounds, 2-w-phenylalkyl-3-mercaptopropionic acid (PhenylCnSH (nꢀ1—4))
and N-[(7-chloro-quinolin-4-ylamino)-alkyl]-3-mercapto-propionamide (QuinolineCnSH (nꢀ2—6)), where n de-
notes the alkyl chain length, on metallo-b-lactamases IMP-1 and VIM-2 was examined. These inhibitors contain
a thiol group and a hydrophobic group linked by variable-length methylene chain. PhenylCnSH (nꢀ1—4) was
found to be a potent inhibitor of both IMP-1 and VIM-2. PhenylC4SH was the potent inhibitor of both IMP-1
(
IC ꢀ1.2 mM) and VIM-2 (IC ꢀ1.1 mM) among this study. When the number of methylene units was varied,
50 50
QuinolineC4SH showed the maximum inhibitory activity against IMP-1 and VIM-2 (IC ꢀ2.5 mM and
50
IC ꢀ2.4 mM). The relationship between the inhibitory effect of the alkyl chain length was different for both se-
5
0
ries of inhibitors, suggesting that IMP-1 has a tighter binding site than VIM-2. QuinolineCnSH did not serve as
a fluorescence reagent for metallo-b-lactamases.
Key words metallo-b-lactamase; inhibitor; thiol; zinc; 3-mercaptopropionic acid
A wide variety of b-lactam antibiotics are used in the addition, the water-1 molecule would be expected to bridge
treatment of infectious diseases. These antibiotics block the the two Zn(II) ions presumably in the form of a hydroxide
cell wall biosynthesis of bacteria by inhibiting a transpepti- ion, although this is not confirmed in the three-dimensional
dase. However, bacteria have developed several strategies for structure of IMP-1 but is based on the three-dimensional
13)
resisting b-lactam antibiotics, including the production of a structure of CcrA (PDB 1ZNB).
family of enzymes, b-lactamases. The b-lactamases catalyze
VIM-2 was isolated from Pseudomonas aeruginosa in
the hydrolysis of b-lactam antibiotics, opening the b-lactam 1996 in France, and has also been identified recently in
9)
1—3)
14)
ring and rendering the antibiotics inactive.
Japan. On the basis of amino acid sequences, 31% of the
b-Lactamases have been grouped into four molecular residues are identical in these enzymes, indicating that VIM-
9)
8)
classes, A, B, C and D, based on their amino-acid sequence
2
and IMP-1 retain the binuclear metal binding motif
4)
homologies by Ambler et al. Those that belong to classes where Zn(II) ions are coordinated by the side chains of con-
A, C and D are serine-b-lactamases which use serine as the served amino acids. The amino acid sequences of a flexible
active site to facilitate catalysis, and those that belong to loop, a b-sheet flap, which is thought to be important in the
class B are metallo-b-lactamases (MBLs) that contain one or binding of particular inhibitors and substrates, are different
two zinc ions at the active site to facilitate b-lactam cleavage. between IMP-1 and VIM-2 especially in the nature and posi-
Based on the difference in their primary structures, MBLs tion of the hydrophobic amino acid: Trp28 which is located
5)
have been classified into three subfamilies, B1 (BcII from in the edge of the flap in the case of IMP-1 but is replaced
6
)
7)
Bacillus cereus, CcrA from Bacteroides fragilis, IMP-1 with Ala in VIM-2.
8)
from Serratia marcescens, VIM-2 from Pseudomonas
A number of compounds have been examined as inhibitors
aeruginosa ), B2 (CphA from Aeromonas hydrophila of MBLs since 1996, including trifluoromethyl alcohols and
AE036 ), and B3 (L1 from Stenotrophomonas malto- ketones,
9)
1
0)
15)
16)
17—25)
hydroxamates,
cysteinyl peptides, biphenyl tetrazoles,
thiols,
thioester deriva-
1
1)
19,26—29)
30)
31,32)
philia ). MBLs are able to hydrolyze most of the b-lactam tives,
6—11)
12,21,29)
antibiotics
and pose a serious clinical threat because of mercaptocarboxylates,
1b-methylcarbapenem deriva-
3
3,34)
35)
their ability to degrade carbapenems including imipenem and tives,
meropenem, which are not degraded by most serine-b-lacta- cyclic natural products,
2,3-disubstituted succinic acid derivatives, tri-
3
6)
37)
sulfonyl hydrazones,
and a
1—3)
38)
mases.
synthetic cephamycin. Most of these compounds, however,
The genes that encode IMP-1 and VIM-2 have been found have been shown to exert their adverse effects only on a
to be located in plasmids that are horizontally transferable to limited number of MBLs.
8
)
other bacteria and no clinically available inhibitors for them
exist at present.
The X-ray crystal structures of MBLs complexed with in-
hibitors have been reported, including CcrA-4-morpholi-
39)
36)
IMP-1 consists of an ab/ba motif, and two Zn(II) ions are neethanesulfonic acid, CfiA-tricyclic natural products,
located at the bottom of a wide shallow groove between the IMP-1-mercaptocarboxylate,
b-sheets. One of the Zn(II) ions is coordinated by three acid derivatives,
histidine residues (3H site), and the other is coordinated by pril. From these reports, these enzymes have conserved the
His, Asp, Cys residues and water-2 molecule (DCH site). In amino acid residues and the binuclear metal center at the ac-
1
2)
2,3-disubstituted succinic
12)
35)
40)
BlaB-D-captopril,
and FEZ-D-capto-
41)
∗
To whom correspondence should be addressed. e-mail: gmphiwin@gpo.kumamoto-u.ac.jp
© 2004 Pharmaceutical Society of Japan

8
52
Vol. 27, No. 6
lined), and IMP-BamHI (5ꢁ-gTTggATCCTAgAAATT
TAgTTgCTTggTTTTgATggT-3ꢁ), designed to add the
BamHI linker (underlined). PCR was performed with 350 ng
of template DNA (plasmid pKF19k/IMP), 50 pmol of each
primer, 2.5 U of taq DNA polymerase (TaKaRa Bio INC.,
Shiga, Japan) in a final volume 100 ml in the buffer system
provided by the manufacturer. The resulting PCR product
was digested with NdeI and BamHI and cloned in pET9a
(Novagen, Madison, Wis.), digested with the same enzymes,
to give the recombinant plasmid. The constructed expression
vectors, denoted as pET9a/IMP, were introduced into E. coli
BL21 (DE3) (Novagen).
Expression and Purification of the IMP-1 Enzymes E.
coli BL21 (DE3), carrying plasmids pET9a/IMP, was cul-
tured in 10 l of LB broth supplemented with kanamycin
Fig. 1. Chemical Structures of Inhibitors Used in This Study
tive site. There are various combinations with respect to the (50 mg/ml). At an Abs of 3, isopropyl-b-D-thiogalactopyra-
6
00
enzyme-inhibitor binding. MBLs offer an opportunity for the noside (IPTG) was added, at final concentration of 1 mM, and
structure-aided design of inhibitors as well as a better under- the culture was further incubated for 8 h. Cells were har-
standing of structure–activity relationships.
vested by centrifugation and suspended in 30 ml of 50 mM
In 1997, we reported that mercaptocarboxylic acids are po- phosphate buffer (pH 7.0, containing 100 mM Zn(NO ) ) and
3
2
1
7)
tent inhibitors of IMP-1. The thiol group of inhibitors was disrupted by sonication (60 cycles, each cycle for 5 s at 1 min
assumed to coordinate with the metal ions in the active site, intervals), then centrifuged at 150000ꢂg for 60 min at 4 °C.
and this was supported by an X-ray crystallographic analysis The cleared supernatant was used in the subsequent chro-
44)
of 2-[5-(1-tetrazolyl-methyl)thien-3-yl]-N-[2-(mercaptometh- matographic purification, as described previously. The pu-
12)
yl)-4-(phenylbutyrylglycine)] and the IMP-1 complex.
rity of the enzyme was verified by sodium dodecyl sulfate
A fluorescent probe of IMP-1 having a dansyl and a thiol polyacrylamide gel electrophoresis (SDS-PAGE) and
group, DansylC2SH, has recently been reported by our Coomassie blue staining. Final preparation showed a single
25)
group. The fluorescent probe also functions as a potent in- band with more than 95% purity.
hibitor of IMP-1. The dansyl group appears to interact with
hydrophobic groups near the active site of IMP-1.
Production and Purification of VIM-2 VIM-2 was ex-
pressed by E. coli NCB326-1B2 harboring pVM4k/VIM-2
In this study, we used three functional groups to strengthen and purified using methods described by Poirel et al. in
9)
the interactions with MBLs: a thiol group, presumed to bind 2000.
with Zn(II) ions, an aromatic ring to bind with the hydropho-
E. coli NCB326-1B2 carrying plasmids pVM4k/VIM-2
bic pocket, and a carboxyl group to interact via hydrogen was cultured in 10 l of LB broth supplemented with ABPC
bonding or electrostatic interactions with conserved residues, (50 mg/ml). The culture was incubated for 12 h at 37 °C
such as Lys. Three types of compounds that contain the three (Abs was 1.0). Cells were harvested by centrifugation and
6
00
functional groups above, PhenylCnSH, QuinolineCnSH, and suspended in 30 ml of 50 mM Bis-Tris buffer (pH 6.5, con-
DansylC2SH shown in Fig. 1 were examined as inhibitors of taining 10 mM Zn(NO ) ) and disrupted by sonication (60 cy-
3
2
IMP-1 and VIM-2.
cles, each cycle for 5 s in a minite), then centrifuged at
1
50000ꢂg for 60 min at 4 °C. The cleared supernatant was
MATERIALS AND METHODS
used in the subsequent chromatographic purification. Anion-
exchange chromatography was performed on a Q-Sepharose
Fast Flow (Pharmacia Biotech, Uppsala, Sweden, f26 mmꢂ
Materials The blaIMP gene is derived from S.
42)
marcescens TN9160 and was cloned into pKF19k (denoted
100 mm) preequilibrated with 50 mM Bis-Tris buffer (pH 6.5,
as pKF19k/IMP). The construction of pKF19k/IMP will be
containing 10 mM Zn(NO ) ) and was eluted with a linear gra-
dient of 0 to 0.3 M NaCl in 50 mM Bis-Tris buffer (pH 6.5,
containing 10 mM Zn(NO ) ). Fractions containing b-lacta-
mase activity were collected, concentrated and the buffer
changed to 50 mM phosphate buffer (pH 7.0, containing
3
2
43)
reported elsewhere. The P. aeruginosa strain NCB326 was
isolated from the drainage fluid of a liver abscess patient in
3
2
1999 in Nara, Japan. BamHI digested fragments of whole-
cell DNA from P. aeruginosa NCB326 were ligated into the
BamHI-restricted plasmid pBCSK(ꢀ) (Stratagene, La Jolla,
Calif., U.S.A.). Recombinant plasmids were introduced by
electroporation into E. coli HB101 competent cells. E. coli
NCB326-1B2 carried a recombinant plasmid pVM4k (named
pVM4k/VIM-2) and this transformant successfully produced
a VIM-2 metallo-b-lactamase.
2
mM Zn(NO ) ) containing 0.3 M NaCl, were applied to a gel-
3
2
filtlation column, Sephadex G-75 (Pharmacia Biotech, Upp-
sala, Sweden, f50 mmꢂ800 mm), preequilibrated with
5
containing 0.3 M NaCl. The purified enzyme was used in sub-
sequent b-lactamase assays. The purity of the preparation
was checked by SDS-PAGE, and final preparation showed a
single band with more than 95% purity. The molecular
0 mM phosphate buffer (pH 7.0, containing 2 mM Zn(NO ) )
3
2
Construction of Expression Vectors, pET9a/IMP The
over-expression of IMP-1 metallo-b-lactamase in E. coli was
achieved as follows: the bla
gene without the nucleotide
IMP
weight of purified VIM-2, M ꢃ29.7 kDa as SDS-PAGE, was
r
9)
coding the signal peptide was amplified using two primers
IMP-NdeI (5ꢁ-TTCCTTTCATATggCAgAgTCTTTgCCA
gATTT-3ꢁ), which was designed to add NdeI linker (under- A spectrophotometric method was used to measure the initial
in agreement with the reported value.
Assays of Substrate Hydrolysis of IMP-1 and VIM-2

June 2004
853
rate of consumption of a chromophoric substrate, nitrocefin,
by IMP-1 (VIM-2). A solution (3.1 ml) of nitrocefin
(100 mM) in 50 mM Tris–HCl buffer (pH 7.4, 0.5 M NaCl) was
incubated at 30 °C for 5 min. The enzyme solution of 0.4 mM
IMP-1 (VIM-2) (10 ml) was added to the substrate solution
and the increase in absorbance at 491 nm (Deꢃ20000) was
recorded. Enzymic rate parameters, K and k , were deter-
m
cat
mined by a nonlinear least-squares method (Kaleida graph)
to fit the Michaelis–Menten equation.
Determination of IMP-1 and VIM-2 The concentra-
tions of IMP-1 and VIM-2 were determined from the ab-
Chart 1. Synthetic Route of QunolineCnSH (nꢃ2—6)
ꢄ1
ꢄ1
sorbance at 280 nm: e280ꢃ49000 M cm for IMP-1 and
ꢄ1
ꢄ1
e
ꢃ32000 M cm for VIM-2.
122.09, 125.37, 127.59, 135.36, 148.65, 150.85, 151.45,
PhenylCnSH (nꢀ1—4) PhenylCnSH (nꢃ1—4) were 171.72.
2
80
45)
prepared using a method described by Park J. D. et al.
QuinolineC5SH, N-[5-(7-Chloro-quinolin-4-ylamino)-
1
QuinolineC3SH, N-[3-(7-Chloro-quinolin-4-ylamino)- pentyl]-3-mercapto-propionamide Yield, 57.3%.
H-
propyl]-3-mercapto-propionamide 4,7-Dichloroquinoline NMR (CDCl ): d 1.20 (t, 1H) 1.51 (q, 2H), 1.61 (q, 2H),
3
(1.64 g, 0.0083 mol) was added to 1,3-diaminopropane 1.80 (q, 2H), 2.48 (t, 1H), 2.76 (t, 2H), 3.24 (t, 2H), 3.37 (t,
(6.15 g, 0.083 mol). After refluxing at 100 °C for 4 h, the 1H), 3.38 (t, 2H), 6.49 (d, 1H, Jꢃ5.49 Hz), 7.40 (d, 1H,
mixture was concentrated to give the crude product, which Jꢃ9.16 Hz), 7.80 (s, 1H), 8.09 (d, 1H, Jꢃ9.15 Hz), 8.33 (d,
1
3
was chromatographed with a Wakogel C-300 (Silica Gel) 1H, Jꢃ5.49 Hz). C-NMR (CDCl ): d 20.84, 24.95, 28.50,
3
column by elution with methanol : chloroform : 28% aqueous 29.75, 39.74, 40.64, 43.68, 99.19, 117.99, 123.73, 126.01,
ammoniaꢃ40 : 360 : 1 to give a yellow solid (1.65 g, 84.5%).
To the yellow solid (1.54 g, 0.0065 mol) and BPO reagent 351.1139 [Calcd for C H N OSCl (M ) 351.1172].
136.55, 147.89, 150.63, 152.73, 173.13. HR-FAB MS: m/z
ꢀ
1
7
22
3
(
benzotriazol-1-yloxytris-(dimethylamino)
phosphonium
QuinolineC6SH, N-[6-(7-Chloro-quinolin-4-ylamino)-
1
hexafluorophosphate) (2.88 g, 0.0065 mol) dissolved in hexyl]-3-mercapto-propionamide Yield, 50.4%. H-NMR
1
50 ml of chloroform, 3-mercaptopropionic acid (0.69 g, (CDCl ): d 1.24 (t, 1H), 1.48 (q, 2H), 1.52 (q, 2H), 1.66 (q,
3
0.0065 mol) and triethylamine (1.76 g, 0.017 mol) were 2H), 1.78 (q, 2H), 2.50 (t, 2H), 2.79 (t, 2H), 3.29 (t, 2H),
added. The mixture was stirred under argon at room tempera- 3.35 (t, 2H), 6.39 (d, 1H, Jꢃ6.10 Hz), 7.35 (d, 1H,
ture overnight and concentrated under reduced pressure to Jꢃ6.71 Hz), 7.83 (s, 1H), 7.97 (d, 1H, Jꢃ8.55 Hz), 8.34
1
3
give a pale yellow solid. This was purified by Wakogel C-300 (d, 1H, Jꢃ5.5 Hz). C-NMR (CDCl ): d 19.44, 25.38,
3
(
Silica Gel) column chromatography eluted with the solution 25.54, 27.41, 28.41, 35.71, 38.20, 39.24, 41.94, 97.78,
containing methanol, chloroform, and 28% aqueous ammo- 116.38, 121.02, 124.30, 126.52, 134.29, 147.58, 149.78,
1
nia (40 : 360 : 1) to give a white solid. (0.71 g, 33.9%). H- 150.38, 170.65. HR-FAB-MS: m/z 365.1274 [Calcd for
ꢀ
NMR (CDCl ): d 1.26 (t, 1H), 1.92 (q, 2H), 2.18 (s, 1H), C H N OSCl (M ) 365.1329].
3
18 24
3
2
1
8
.57 (t, 2H), 2.82 (t, 2H), 3.38 (t, 2H), 3.51 (t, 2H), 6.57 (d,
Inhibition of IMP-1 and VIM-2 A portion of 0.1 ml of
H, Jꢃ6.71 Hz), 7.47 (d, 1H, Jꢃ1.83 Hz), 7.74 (s, 1H), an inhibitor [PhenylCnSH (nꢃ1—4), DansylCnSH (nꢃ2),
1
3
.13 (d, 1H, Jꢃ9.15 Hz), 8.23 (d, 1H, Jꢃ6.71 Hz). C- QuinolineCnSH (nꢃ2—6)] at various concentrations in
NMR (CDCl ): d 20.48, 27.97, 36.55, 40.08, 40.39, 98.55, methanol and 0.01 ml of 310 nM enzyme (IMP-1 and VIM-2)
3
1
1
16.37, 122.35, 123.68, 127.10, 138.41, 142.37, 145.79, in 50 mM Tris–HCl (pH 7.4, 0.5 M NaCl) was added to 2.9 ml
54.04, 173.15. HR-FAB-MS: m/z 323.0847 [Calcd for of 50 mM Tris–HCl buffer (pH 7.4, 0.5 M NaCl). The enzyme–
ꢀ
C H N OSCl (M ) 323.0859].
inhibitor mixture was incubated for 10 min at 30 °C, and a
1
5
18
3
QuinolineCnSH (nꢀ2, 4—6) These compounds were 0.1 ml DMSO solution of 3.1 mM nitrocefin was then added
synthesized, as described for the synthesis of Quino- and the absorbance change at 491 nm monitored for 3 min.
lineC3SH, as described above. A flow chart of the synthesis Methanol was used instead of an inhibitor as a control.
ꢄ1
is shown in Chart 1.
The molar activity, k /min , was determined from the ini-
2
QuinolineC2SH, N-[2-(7-Chloro-quinolin-4-ylamino)- tial velocity, v, by v/[E], and plotted against the concentration
1
ethyl]-3-mercapto-propionamide Yield, 13.8%. H-NMR of the inhibitors. IC50 values were determined from this and
(
2
(
4
1
CDCl ): d 1.24 (t, 1H), 2.59 (t, 2H), 2.78 (t, 2H), 3.64 (t, K values were determined by fitting k to a competitive inhi-
3
i
2
H), 3.7 (q, 2H), 6.7 (d, 1H), 7.58 (d, 1H), 7.73 (s, 1H), 8.12 bition of Eq. 1 using Km values of 20 mM and 22.5 mM for
13
d, 1H), 8.19 (d, 1H). C-NMR (CDCl ): d 38.04, 39.52, IMP-1 and VIM-2 respectively,
3
5.62, 47.29, 98.46, 115.33, 119.58, 124.12, 127.98, 138.63,
40.00, 142.76, 155.78, 174. HR-FAB-MS: m/z 309.0684
V
max[S]
k2 ꢃ
(1)
ꢀ
[Calcd for C H N OSCl (M ) 309.0703].

[I] 
1
4
16
3
Km 1ꢀ
ꢀ[S]


QuinolineC4SH, N-[4-(7-Chloro-quinolin-4-ylamino)-
Ki


1
butyl]-3-mercapto-propionamide Yield, 27.1%. H-NMR
(
2
CDCl ): d 1.22 (t, 1H), 1.54 (q, 2H), 1.74 (q, 2H), 2.48 (t, where, k : molar activity, k2maxꢃVmax/[E], Vmaxꢃthe maxi-
3
2
H), 2.78 (t, 2H), 3.24 (t, 2H), 3.30 (t, 2H), 6.38 (d, 1H, mum velocity.
Fluorescence Titration of IMP-1 with QuinolineCnSH
H, Jꢃ4.88 Hz), 8.39 (d, 1H, Jꢃ5.5 Hz). C-NMR (CDCl3): (nꢀ2—6) The method used to examine fluorescence re-
Jꢃ5.49 Hz), 7.34 (d, 1H, Jꢃ7.33 Hz), 7.86 (s, 1H), 7.87 (d,
1
d 20.51, 26.61, 29.48, 39.27, 40.31, 43.01, 98.85, 117.45, ported by Kurosaki et al. was used in this experiment. Flu-
1
3
25)

8
54
Vol. 27, No. 6
Fig. 2. Typical Plots of Molar Activity of IMP-1 (Left) and VIM-2 (Right) against the Concentration of PhenylC3SH
orescence spectra were measured at 25 °C, with a 5.0 nm of Table 1. IC50 and K
Values for PhenylCnSH (nꢃ1—4) and Quinoline
i
CnSH (nꢃ2—6) to IMP-1 and VIM-2
slit width, using 1 mM of QuinolineCnSH (nꢃ2—6) as a fluo-
rescence agent with an excitation of 340 nm and 280 nm
using 50 mM Tris–HCl buffer (pH 7.4, 0.5 M NaCl) containing
IC /mM
5
0
Inhibitors
IMP-1
VIM-2
1
0% of methanol.
RESULTS
K (Inhibition Constant) and IC (50% Inhibitory
PhenylC1SH
PhenylC2SH
PhenylC3SH
PhenylC4SH
QuinolineC2SH
QuinolineC3SH
QuinolineC4SH
QuinolineC5SH
QuinolineC6SH
16.4
7.4
9.7
1.2
14.2
6.6
2.5
6.7
3.4
14.3
2.6
1.3
1.1
6.3
6.5
2.4
7.6
4.9
i
50
Concentration) of Inhibitors agaist IMP-1 and VIM-2
Using nitrocefin as a substrate, the molar activity k /min in
the presence of each inhibitor was obtained from the initial
rate measurement and was plotted against the concentration
of each inhibitor, (PhenylCnSH (nꢃ1—4), DansylCnSH
ꢄ1
2
(nꢃ2), and QuinolineCnSH (nꢃ2—6)). Typical plots are
shown in Fig. 2 (IMP-1 and VIM-2). Values for the IC50 of effect on the number of methylene units of the linker, n, is
IMP-1 and VIM-2 were calculated from these concentration different for IMP-1, compared to VIM-2. PhenylC4SH
response curves. Inhibition constants, K , were determined by inhibits both IMP-1 and VIM-2 more strongly and
i
fitting the kinetic data to that for the competitive inhibition PhenylC1SH the most weakly. VIM-2 is as susceptible when
using Eq. 1.
Inhibition by PhenylCnSH (nꢀ1—4) PhenylCnSH nificant only for nꢃ4.
nꢃ1—4) were synthesized and tested as inhibitors of car-
Inhibition by a Fluorescent Probe, DansylCnSH (nꢀ2)
A fluorescent probe, DansylC2SH for IMP-1, containing a
nꢃ2 and nꢃ3 as nꢃ4 but the susceptibility of IMP-1 is sig-
(
45)
boxypeptidase A (CPA), a Zn(II) enzyme, by Park et al.
The IC50 and K values for IMP-1 and VIM-2 are listed in thiol group and a dansyl group was synthesized, and reported
i
2
5)
Table 1.
as potent inhibitor for IMP-1. DansylC2SH (nꢃ2) inhib-
The IC s for IMP-1 were found to be in a range between ited IMP-1 and VIM-2: the IC50 values were 2.3 mM and
5
0
1
.2 mM (PhenylC4SH) and 16.4 mM (PhenylC1SH), and the 5.2 mM, the inhibition constant K was 0.42 mM and 1.1 mM for
i
25)
inhibition constants, K , had values between 0.21 mM VIM-2 and IMP-1.
i
(
PhenylC4SH) and 2.78 mM (PhenylC1SH). When the num-
QuinolineCnSH (nꢀ2—6) An increase in fluorescence
ber of methylene units in the linker was varied, PhenylC4SH intensity was found on the addition of IMP-1 to
was found to inhibit IMP-1 more strongly among Phenyl- DansylC2SH, but the addition of IMP-1 to QuinolineCnSH
CnSH (nꢃ1—4). 3-Mercaptopropionic acid has been re- led to a decrease in fluorescence intensity. A comparison of
1
7)
ported to have a K value of 1.2 mM to IMP-1. In compari- structures of QuinolineCnSH and with that of DansylC2SH
i
son, PhenylC4SH was found to be a more potent inhibitor.
shows that the sulfonyl group between the quinoline group, a
IC values for VIM-2 were found to be in a range between fluorescence group, and the HN group is missing in Quino-
5
0
1
.1 mM (PhenylC4SH) and 14.3 mM (PhenylC1SH). The lineCnSH.
The IC50 and K values for IMP-1 and VIM-2 are listed in
inhibition constants, K , have values between 0.19 mM
i
i
(
PhenylC4SH) and 2.36 mM (PhenylC1SH). PhenylCnSH Table 1.
(nꢃ1—4) were found to be potent inhibitors of VIM-2 as
QuinolineCnSH (nꢃ2—6) act as effective inhibitors
well as of IMP-1. When the number of methylene units of the against IMP-1 and VIM-2. The IC50 for IMP-1 was found to
linker was varied, PhenylC3SH and PhenylC4SH was also be in a range between 2.5 mM (QuinolineC4SH) and 14.2 mM
the potent inhibitor of VIM-2.
(QuinolineC2SH). The inhibition constants, K , have values
i
PhenylCnSH (nꢃ1—4) were found to be potent inhibitor between 0.44 mM (QuinolineC4SH) and 2.97 mM (Quino-
for both IMP-1 and VIM-2, and a comparison of the inhibi- lineC2SH). The IC for VIM-2 was found to be in a
5
0
tion constant K for PhenylCnSH (nꢃ1—4) for IMP-1 and range between 2.4 mM (QuinolineC4SH) and 7.6 mM (Quino-
i
VIM-2 is shown in Fig. 3. The dependency of the inhibitory lineC5SH). The inhibition constants, K , have values between
i

June 2004
855
Fig. 3. Variation of Inhibition Constants of IMP-1 and VIM-2 with n, for PhenylCnSH (Left) and QuinolineCnSH (Right)
drophobic pocket of VIM-2 is larger than IMP-1, and is lo-
cated near the Zn(II) site. PhenylC1SH showed only a weak
inhibition for VIM-2. When the number of methylene units is
3
and 4, it can be considered to be a suitable distance for in-
teracting with the hydrophobic pocket of VIM-2 (Fig. 3).
For the case of the inhibition of CPA by PhenylCnSH
(
nꢃ1—4), the K of PhenylCnSH (nꢃ1—4) was determin-
i
45)
ed to be 0.010, 2.1, 1.35, and 1.3 mM respectively.
PhenylC1SH inhibits CPA most strongly by electrostatic in-
teractions between the oxygen of the carboxyl group and the
ionic side chain of CPA as well as binding of the thiol with
Zn(II) and hydrophobic interactions. Although PhenylC4SH
Fig. 4. Relation of Inhibitory Effect versus Structures, for Several In-
hibitors of IMP-1
inhibited most strongly, the K values for IMP-1 and VIM-2
i
0
.42 mM (QuinolineC4SH) and 1.39 mM (QuinolineC5SH). are not greatly different from the others with respect to CPA.
When the number of methylene units was varied, Quino- The observation that PhenylCnSH (nꢃ1—4) inhibited both
lineC4SH showed the maximum inhibitory activity against of IMP-1 and VIM-2 indicates that these compounds are in-
both IMP-1 and VIM-2, as shown in Fig. 3. The magnitude hibitors of other MBLs.
of K decreased with the number of methylene units from 2 to
A comparison of the inhibition constants and structures of
i
4
for IMP-1, but remained unchanged for VIM-2.
other inhibitors are shown in Fig. 4.
Although only the thiol group of PhenylC1SH is replaced
with a carboxyl group for inhibitor B, a 2,3-disubstituted
DISCUSSION
35)
succinic acid, reported by Toney a large difference in the
PhenylCnSH (nꢃ1—4) and QuinolineCnSH (nꢃ2—6) are IC : 16.4 mM was found for PhenylC1SH and 490 mM by B
5
0
found to be potent inhibitors for MBLs in this study.
for IMP-1 when the substrate was nitorocefin. This suggests
The following interactions of PhenylCnSH (nꢃ1—4) with that a thiol group is a very important group in terms of in-
IMP-1 and VIM-2 would be predicted: 1) replacement of the hibiting MBLs. PhenylC4SH was the potent inhibitor of
water-1 between the two Zn(II) ions with the thiol group in IMP-1 among the PhenylCnSH series. However, Mollard re-
23)
the active site, 2) hydrogen bonding of the oxygen of the car- ported that inhibitor C has a K value of 29 nM lower by 10
i
boxyl group with amino acids of the enzyme or replacement fold than PhenylC4SH. We have reported that mercaptoacetic
1
7)
of water-2 coordinated to the Zn(II) ion in the DCH site, 3) acid is a reversible inhibitor with a K of 0.23 mM. Inhibitor
i
hydrophobic interactions between the phenyl group and the C having the phenyl group on C2 position of mercaptoacetic
hydrophobic pockets of MBLs.
The thiol group of PhenylCnSH (nꢃ1—4) inhibitors mercaptoacetic acid. This higher K may reflect much higher
acid led to approximately 10 fold high activity compared to
1)
i
appears to bind with Zn(II) ions of the active site of MBLs. hydrophobic Toney et al. have reported that the addition of
This was supported by the X-ray crystallographic analysis of the benzyl group at C3 position of inhibitor B increases K by
i
12)
5
35)
a complex of IMP-1 with a thiol compound.
The phenyl group of PhenylCnSH (nꢃ1—4) may in-
1.8ꢂ10 fold (0.0027 mM) compared to inhibitor B.
2)
Although QuinolineCnSH could not be used for the detec-
teract with hydrophobic amino acid(s) near the active site. tion of MBLs by fluorescence, QuinolineCnSH are potent in-
The phenyl ring of PhenylC4SH is located at a sufficient dis- hibitors for both IMP-1 and VIM-2.
tance to effect hydrophobic interactions with a hydrophobic
For both PhenylCnSH and QuinolineCnSH, the inhibitory
pocket near the active site of IMP-1. PhenylC3SH and effect is maximum for nꢃ4 for IMP-1 and VIM-2 and least
PhenylC4SH inhibit both IMP-1 and VIM-2 more strongly, for nꢃ1 for IMP-1. But the effect of variation in the number
and PhenylC2SH was found to strongly inhibit VIM-2 to a of methylene units in the linker, n, is different between IMP-
greater extent than IMP-1. These results suggest that the hy- 1 and VIM-2. The thiol group acts as a coordinating ligand

8
56
Vol. 27, No. 6
of one of the two Zn(II) ions or both as a bridging ligand and 20) Scrofani S. D. B., Chung J., Huntley J. J. A., Benkovic S. J., Wright P.
E., Dyson H. J., Biochemistry, 38, 14507—14514 (1999).
provides strong driving force for binding with the active cen-
ter. However, for being a good or relevant inhibitor, candidate
compounds should be involved in specific interactions
2
1) Greenlee M. L., Laub J. B., Balkovec J. M., Hammond M. L., Ham-
mond G. G., Pompliano D. L., Epstein-Toney J. H., Bioorg. Med.
Chem. Lett., 9, 2549—2554 (1999).
around the environment of the active site. The phenyl group 22) Arakawa Y., Shibata N., Shibayama K., Kurokawa H., Yagi T., Fuji-
wara H., Goto M., J. Clin. Microbiol., 38, 40—43 (2000).
in PhenylCnSH and the quinoline group in QuinolineCnSH
2
3) Mollard C., Moali C., Papamicael C., Damblon C., Vessilier S., Ami-
cosante G., Schofield C. J., Galleni M., Frère J.-M., Roberts G. C. K.,
J. Biol. Chem., 276, 45015—45023 (2001).
are intended to interact with the hydrophobic region of the
enzymes. The dependency of the inhibitory effect on the
number of the methylene (n) reflects the difference in the
geometry of MBL: IMP-1 has a narrower or tighter hydro-
phobic pocket than VIM-2.
2
2
2
4) Siemann S., Clarke A. J., Viswanatha T., Dmitrienko G. I., Biochem-
istry, 42, 1673—1683 (2003).
5) Kurosaki H., Yasuzawa H., Yamaguchi Y., Jin W., Arakawa Y., Goto
M., Org. Biomol. Chem., 1, 17—20 (2003).
6) Payne D. J., Bateson J. H., Gasson B. C., Khushi T., Proctor D., Pear-
son S. C., Reid R., FEMS Microbiol. Lett., 157, 171—175 (1997).
Acknowledgments This study was partially supported
by the Research Foundation for Kumamoto Techno Associa- 27) Payne D. J., Bateson J. H., Gasson B. C., Proctor D., Khushi T., Farmer
tion, Sagawa Science and Technology promotion Foundation
to Yamaguchi, a Regional Science Promotion Program of
Japan Science and Technology Agency to Kurosaki, Lions
Clubs International district 337-D to Jin, and H15-Shinkou-9
from the Ministry of Health, Labor and Welfare of Japan to
Goto.
T. H., Tolson D. A., Bell D., Skett P. W., Marshall A. C., Reid R.,
Ghosez L., Combret Y., Marchand-Brynaert J., Antimicrob. Agents
Chemother., 41, 135—140 (1997).
2
8) Hammond G. G., Huber J. L., Greenlee M. L., Laub J. B., Young K.,
Silver L. L., Balkovec J. M., Pryor K. D., Wu J. K., Leiting B., Pompli-
ano D. L., Toney J. H., FEMS Microbiol. Lett., 459, 289—296 (1999).
29) Payne D. J., Du W., Bateson J. H., Exp. Opin. Invest. Drugs, 9, 247—
61 (2000).
0) Bounaga S., Galleni M., Laws A. P., Page M. I., Bioorg. Med. Chem.,
, 503—510 (2001).
2
3
3
REFERENCES
9
1) Toney J. H., Fitzgerald P. M. D., Grover-Sharma N., Olson, S. H., May
W. J., Sundelof J. G., Vanderwall D. E., Cleary K. A., Grant S. K., Wu
J. K., Kozarich J. W., Pompliano D. L., Hammond G. G., Chem. Biol.,
1)
2)
3)
4)
5)
Page M. I., Laws A. P., Chem. Commun., 1998, 1609—1617 (1998).
Page M. I., Curr. Pharm. Des., 5, 895—913 (1999).
Cricco J. A., Vila A. J., Curr. Pharm. Des., 5, 915—927 (1999).
Ambler R. P., Phil. Trans. R. Soc. Lon., B289, 321—331 (1980).
Galleni M., Lamotte-Brasseur J., Rossolini G. M., Spencer J., Dide-
berg O., Frère J.-M., Antimicrob. Agents Chemother., 45, 660—663
5
, 185—196 (1998).
3
2) Toney J. H., Cleary K. A., Hammond G. G., Yuan X., May W. J.,
Hutchins S. M., Ashton W. T., Vanderwall D. E., Bioorg. Med. Chem.
Lett., 9, 2741—2746 (1999).
3) Nagano R., Adachi Y., Imamura H., Yamada K., Hashizume T., Mo-
rishima H., Antimicrob. Agents Chemother., 43, 2497—2503 (1999).
4) Nagano R., Adachi Y., Hashizume T., Morishima H., J. Antimicrob.
Chemother., 45, 271—276 (2000).
5) Toney J. H., Hammond G. G., Fitzgerald P. M. D., Sharma N.,
Balkovec J. M., Rouen G. P., Olson S. H., Hammond M. L., Greenlee
M. L., Gao Y.-D., J. Biol. Chem., 276, 31913—31918 (2001).
6) Payne D. J., Hueso-Rodríguez J. A., Boyd H., Concha N. O., Janson C.
A., Gilpin M., Bateson J. H., Cheever C., Niconovich N. L., Pearson
S., Rittenhouse S., Tew D., Díez E., Pérez P., de la Fuente J., Rees M.,
Rivera-Sagredo A., Antimicrob. Agents Chemother., 46, 1880—1886
(
2001).
Hussain M., Carlino A., Madonnan M. J., Lampen J. O., J. Bacteriol.,
64, 223—229 (1985).
Rasmussen B. A., Gluzman Y., Tally F. P., Antimicrob. Agents
Chemother., 34, 1590—1592 (1990).
3
3
3
6)
7)
8)
1
Arakawa Y., Murakami M., Suzuki K., Ito H., Wacharotayankun R.,
Ohsuka S., Kato .N., Ohta M., Antimicrob. Agents Chemother., 39,
1
612—1615 (1995).
3
9)
Poirel L., Naas T., Nicolas D., Collet L., Bellais S., Cavallo J.-D.,
Nordmann P., Antimicrob. Agents Chemother., 44, 891—897 (2000).
1
0) Zervosen A., Valladares M. H., Devreese B., Prosperi-Meys C.,
Adolph H.-W., Mercuri P. S., Vanhove M., Amicosante G., van Beeu-
men J., Frère J.-M., Galleni M., Eur. J. Biochem., 268, 3840—3850
(2002).
3
7) Siemann S., Evanoff D. P., Marrone L., Clarke A. J., Viswanatha T.,
(2001).
Dmitrienko G. I., Antimicrob. Agents Chemother., 46, 2450—2457
11) Walsh T. R., Hall L., Assinder S. J., Nichols W. W., Cartwright S. J.,
(2002).
MacGowan A. P., Bennett P. M., Biochim. Biophys. Acta, 1218, 199—
3
8) Quiroga M. I., Franceschini N., Rossolini G. M., Gutkind G., Bon-
2
01 (1994).
figlio G., Franchino L., Amicosante G., Chemotherapy, 46, 177—183
1
2) Concha N. O., Janson C. A., Rowling P., Pearson S., Cheever C. A.,
Clarke B. P., Lewis C., Galleni M., Frère J.-M., Payne D. J., Bateson J.
H., Abdel-Meguid S. S., Biochemistry, 39, 4288—4298 (2000).
(2000).
3
9) Carfi A., Duée E., Paul-Soto R., Galleni M., Frère J.-M., Dideberg O.,
Acta Cryst., D54, 47—57 (1998).
0) García-Sáez I., Hopkins J., Papamicael C., Franceschini N., Ami-
cosante G., Rossolini G. M., Galleni M., Frère J.-M., Dideberg O., J.
Biol. Chem., 278, 23863—23873 (2003).
13) Concha N. O., Rasmussen B. A., Bush K., Herzberg O., Structure, 4,
4
8
23—836 (1996).
14) Shibata N., Doi Y., Yamane K., Yagi T., Kurokawa H., Shibayama K.,
Kato H., Kai K., Arakawa Y., J. Clin. Microbiol., 41, 5407—5413
4
4
4
4
4
1) García-Saez I., Mercuri P. S., Papamicael C., Kahn R., Frère J.-M.,
(2003).
Galleni M., Rossolini G. M., Dideberg O., J. Mol. Biol., 325, 651—
1
5) Walter M. W., Felici A., Galleni M., Soto R. P., Adlington R. M., Bald-
win J. E., Frère J.-M., Gololobov M., Schofield C. J., Bioorg. Med.
Chem. Lett., 6, 2455—2458 (1996).
6
60 (2003).
2) Osano E., Arakawa Y., Wacharotayankun R., Ohta M., Horii T., Ito H.,
Yoshimura F., Kato N., Antimicrob. Agents Chemother., 38, 71—78
1
6) Walter M. W.,Valladares M. H., Adlington R. M., Amicosante G.,
Baldwin J. E., Frère J.-M., Galleni M., Rossolini G. M., Schofield C.
J., Bioorg. Chem., 27, 35—40 (1999).
(1994).
3) Yamaguchi Y., Kuroki T., Yasuzawa H., Jin W., Higashi T., Kawanami
A., Taki T., Yamagata Y., Arakawa Y., Goto M., Manuscript in prepa-
ration.
4) Goto M., Yasuzawa H., Higashi T., Yamaguchi Y., Kawanami A., Mi-
fune S., Mori H., Nakayama H., Harada K., Arakawa Y., Biol. Pharm.
Bull., 26, 589—594 (2003).
1
1
1
7) Goto M., Takahashi T., Yamashita F., Koreeda A., Mori H., Ohta M.,
Arakawa Y., Biol. Pharm. Bull., 20, 1136—1140 (1997).
8) Bounaga S., Laws A. P., Galleni M., Page M. I., Biochem. J., 331,
7
03—711 (1998).
9) Fitzgerald P. M. D., Wu J. K., Toney J. H., Biochemistry, 37, 6791—
800 (1998).
5) Park J. D., Kim D. H., J. Med. Chem., 45, 911—918 (2002).
6