Synthesis and PDF inhibitory activities of novel benzothiazolylidenehydroxamic acid derivatives

Article abstract of DOI:10.1016/S0960-894X(03)00675-9

A novel series of benzothiazolylidenehydroxamic acid derivatives has been designed and synthesized as PDF inhibitors. Some of this novel class of PDF inhibitors exhibited micromolar order enzyme inhibitory activity and antibacterial activity.

Full text of DOI:10.1016/S0960-894X(03)00675-9

Bioorganic& Medicinal Chemistry Letters 13 (2003) 3273–3276
Synthesis and PDF Inhibitory Activities of Novel
Benzothiazolylidenehydroxamic Acid Derivatives
Wataru Takayama,* Yoshihisa Shirasaki, Yusuke Sakai, Emi Nakajima,
Shuhei Fujita, Kanako Sakamoto-Mizutani and Jun Inoue
Research Laboratories, Senju Pharmaceutical Co., Ltd., 1-5-4 Murotani, Nishi-Ku, Kobe, Hyogo 651-2241, Japan
Received 9 May 2003; revised 20 June 2003; accepted 23 June 2003
Abstract—A novel series of benzothiazolylidenehydroxamicacid derivatives has been designed and synthesized as PDF inhibitors.
Some of this novel class of PDF inhibitors exhibited micromolar order enzyme inhibitory activity and antibacterial activity.
# 2003 Elsevier Ltd. All rights reserved.
In recent years, many new antibiotic-resistant bacteria
have been appeared throughout the world, and the
increasing incidence emphasizes the need to discover
and develop novel antibacterial drugs with new modes
of action. So far, many successful antibacterial drugs
are targeted to inhibit the protein synthesis of bacteria.
Prokaryoticprotein synthesis begins with for-
mylmethionine-tRNA resulting in the synthesis of
N-terminally formylated polypeptides, and the formyl
group must be subsequently removed for the proteins to
function. Peptide deformylase (PDF), a unique subclass
of metalloenzymes, catalyzes the removal of the formyl
group at the N-terminus of bacterial proteins.1 Defor-
mylation of nascent peptides is a unique feature of bac-
terial cells and is essential for their survival,2 whereas
the peptide deformylase is absent in mammalian cells.3
Therefore, inhibition of PDF represents an attractive
target for the discovery of novel antibiotics. Recent
studies from several research groups have shown that
PDF inhibitors act as broad-spectrum antibacterial
agents.4
The general route for the synthesis of the benzothiazo-
lylidenehydroxamicaidc derivatives is outlined in
Scheme 1. Benzothiazolylacetic acid ethyl esters 2(a–d)
were prepared following a published method.6 Oxida-
tion of 2(a–d) with equimolar amount of mCPBA in
CH2Cl2 afforded olefin esters 3(a–d). The resulting
esters 3(a–d) were hydrolyzed with 4 N NaOH in etha-
nol to give the corresponding carboxylic acids 4(a–d).
To convert the carboxyl groups to hydroxamic function
moieties, 4(a–d) were treated with hydroxylamine
hydrochloride in the presence of 1-hydroxybenzo-
triazole (HOBt), 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide hydrochloride (EDC) and Et3N in DMF
to afford 5(a–d). Oxidation of 2c with three equivalents
of mCPBA gave a sulfonyl compound 6c, which was
then converted to 8c by using a similar method to the
preparation of 5c. The structures of novel compounds
were assigned based on NMR spectra and elemental
analyses.7 The stereochemistry of the double bonds of
conjugated carbonyl compounds (3–8) was determined
based on the result of nuclear Overhauser effect (NOE)
experiment. For example, irradiation at a frequency
corresponding to the methylene protons at d=3.46 (2H,
q, J=7.2 Hz) of 5a revealed a clear signal enhancement
of proton signal (d=6.96) of the double bond of con-
jugated carbonyl group. Thus, the double bond was
assigned to Z.
Based on the mechanism of deformylation of peptides1
and the pharmacophore requirements for PDF inhib-
itor,5 a novel series of benzothiazolylidenehydroxamic
acid derivatives was designed and synthesized (Fig. 1).
They were subsequently evaluated for their in vitro
ability to inhibit PDF and antibacterial activity, and the
observed structure–activity relationships were analyzed
using computational docking simulation.
The newly synthesized compounds were evaluated for
their in vitro inhibitory effects on Escherichia coli Ni-
PDF using a published method,8 and their in vitro
antibacterial activities were tested following a guideline
of The Chemotherapy Society of Japan.9 The activities
*Corresponding author. Tel.: +81-78-997-1010; fax: +81-78-997-
1016; e-mail: yingshe-cui@senju.co.jp
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00675-9

3274
W. Takayama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3273–3276
Oxidation of sulfur of 5c to sulfone (8c) resulted in
complete loss of PDF inhibitory activity. The PDF
inhibitors, 5b, 5c and 5d, showed antibacterial activity
against S. aureus and P. aeruginosa, but they were inactive
against E. coli (Table 1). There was not a good correlation
between the IC50 values and the MICs. The precise reason
remains to be investigated, but we can speculate that there
might be several possible factors, such as poor penetration
of the bacterial cell membrane, or recognition by an effi-
cient efflux system of the bacteria.8b
Computational docking simulation provided a good
explanation for the observed structure–activity rela-
tionship in enzyme inhibitory activity. A docking model
was prepared by using a crystal structure of E. coli
Ni-PDF with a ligand MAS (Met-Ala-Ser) reported by
Becker et al.10 and deposited in the Protein Data Bank
(code 1BS6). In order to conduct the docking simula-
Figure 1. Chemical structures of general PDF substrate, actinonin and
the novel benzothiazolylidenehydroxamicacid.
tions by using the Tripos docking program FlexXTM 11
,
the Ni atom was displaced with a Zn atom. Minimiza-
tion of the molecules was conducted by using the Tripos
force field (Trips 60) with the Powell method,12 and the
enzyme–inhibitor interactions in FlexX include both
polar (hydrogen bond and charge–charge) and non-
polar (hydrophobic) terms. Many of these interactions
are geometrically quite restrictive, which allows FlexX
to accurately place inhibitors and their fragments.
Directly docking a potential inhibitor into the active site
of the enzyme covers all possible docking combinations
of inhibitor conformations and binding modes. Finally,
the docking results were minimized again with the Tri-
pos force field, defining monomers outside the 5-A area
around the inhibitor as an aggregate.
Scheme 1. Reagents and conditions: (a) 1 equiv mCPBA, CH2Cl2, rt,
overnight; (b) 4 N KOH, EtOH, reflux, 5 h; (c) HOBt, EDC,
.
NH2OH HCl, Et3N, rt overnight; (d) 3 equiv mCPBA, CH2Cl2, rt,
overnight.
of these compounds compared to actinonin8b are sum-
marized in Table 1.
The hydroxamic acid function group, a bidentate metal
chelating moiety, is critical for the enzyme inhibitory
activity (5b, 5c, and 5d). The compound with a car-
boxylic acid as a metal chelating group (4c) is devoid of
PDF inhibitory activity. It is not sufficient, however, to
only have hydroxamic acid functionality, because the
PDF inhibitory activity of these inhibitors is strongly
influenced by the nature of the N-alkyl substituent R.
The activity enhanced as the length of the N-alkyl chain
increased from ethyl (5a, IC50>100 mM), to n-propyl
(5b, IC50=8.39 mM), and n-butyl (5c, IC50=1.04 mM).
When the N-butyl group was replaced with n-pentyl,
however, the activity decreased (5d, IC50=9.91 mM).
The docking result of actinonin is shown in Figure 2.
The actinonin bounds in a similar fashion to that of the
crystal structure of actinonin-PDF (PDB code 1G2A),13
which suggests the reliability of this docking model. The
Table 1. In vitro PDF inhibitory and antibacterial activities of ben-
zothiazolylidenehydroxamicacids
Compd
IC50 (mM)a
MIC (mg/mL)b
S. aureusc
E. colid
P. aeruginosae
1f
1.50ꢀ10ꢁ3
>100
>100
8.39
1.04
9.91
>100
3
>100
>100
10
30
100
50
>100
>100
>100
>100
>100
>100
100
>100
>100
100
100
100
4c
5a
5b
5c
5d
8c
>100
>100
Figure 2. Binding mode of the most stable docking model for actino-
nin within the substrate pocket of PDF. Hydrogen bonds (distances
less than 2.8 A between the hydrogen bonded to an H-bond donor and
an acceptor) are shown in dotted yellow lines. Carbon atoms are
shown in white, hydrogens are cyan, nitrogens are blue, and oxygens
are red, except those of the amino acids (orange). Zn atom is shown as
a red ball.
aThe inhibitor concentration that can inhibit 50% of enzyme activity.
bMinimal inhibitory concentration.
cStaphylococcus aureus (ATCC 10390).
dEscherichia coli (ATCC 25922).
ePseudomonas aeruginosa (ATCC 9027).
fKi=0.3 nM (lit.8b).

W. Takayama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3273–3276
3275
binding mode of 5c (Fig. 3), the most potent PDF inhib-
itor of this study, suggests that the inhibitor lie in a cleft
on the enzyme surface and within the active site. The
hydroxamicaicd moiety was positioned favorably for
chelating the Zn atom in the active site of the PDF. The
distances from the metal to nitrogen-bound oxygen and
carbonyl oxygen of 5c are 1.70 and 2.65 A, respectively,
while in the case of actinonin the distances are 2.34 and
3.22 A, respectively. Some hydrogen bonds were also
made between carbonyl oxygen of 5c and Leu 91,
between nitrogen-bound oxygen of 5c and Gln 50, and
between NH of 5c and His 136 of the enzyme. The N-n-
study helps rationalize the structure–activity relation-
ship and will assist further lead optimization studies.
Acknowledgements
The authors thank Dr. Yutaka Kawamatsu (Senju) for
his helpful advice.
References and Notes
0
butyl chain of 5c sat in the hydrophobicS1 pocket,
1. (a) Becker, A.; Schlichting, I.; Kabsch, W.; Groche, D.;
Schultz, S.; Wagner, A. F. Nat. Struct. Biol. 1998, 5, 1053. (b)
Groche, D.; Becker, A.; Schlichting, I.; Kabsch, W.; Schultz,
S.; Wagner, A. F. Biochem. Biophys. Res. Commun. 1998, 246,
342. (c) Becker, A.; Schlichting, I.; Kabsch, W.; Schultz, S.;
Wagner, A. F. J. Biol. Chem. 1998, 273, 11413.
2. (a) Mazel, D.; Pochet, S.; Marliere, P. EMBO J. 1994, 13,
914. (b) Meinnel, T.; Lazennec, C.; Villoing, S.; Blanquet, S.
J. Mol. Biol. 1997, 267, 749.
which was generated by the residues Gly 43, Gly 45,
Glu-88, Ile 128, Cys-129, and Glu-133 of the enzyme.
The n-butyl group made a better fit than the other N-
alkyl groups, such as ethyl (5a), n-propyl (5b), and n-
pentyl (5d) groups. Since the ethyl group is not long
enough to fit into the S10 pocket of the enzyme, and the
n-pentyl group is slightly long for the pocket, resulting
in an eclipsing interaction at its terminus. While the n-
propyl group is slightly shorter than that of n-butyl
group for the S10 pocket of the enzyme, which accounts
for the less potent PDF inhibitory activity of 5b com-
paring to 5c. The benzothiazole ring of the inhibitor is
positioned in the S20 binding pocket of the enzyme to
keep hydrophobic interaction with side chains of Ile 44
and Gly 89 of the enzyme. In addition, the FlexX energy
of 5c (ꢁ5.08 kcal/mol) was lower than that of 5b (ꢁ5.04
kcal/mol) and 5d (ꢁ4.95 kcal/mol). In the case of acti-
nonin in this study, the FlexX energy was ꢁ5.36 kcal/
mol. There is a trend of increased energy of interaction
in FlexX resulting in decreased activity of enzyme inhi-
bition. The sulfonyl compound 8c could not fit into the
active site of present docking model, which resulted in
its lack of PDF inhibitory activity.
3. Rajagopalan, P. T. R.; Yu, X. C.; Pei, D. J. Am. Chem.
Soc. 1997, 119, 12418.
4. (a) Waller, A. S.; Clements, J. M. Curr. Opin. Drug Discov.
Devel. 2002, 5, 785. (b) Smith, H. K.; Beckett, R. P.; Clements,
J. M.; Doel, S.; East, S. P.; Launchbury, S. B.; Pratt, L. M.;
Spavold, Z. M.; Thomas, W.; Todd, R. S.; Whittaker, M.
Bioorg. Med. Chem. Lett. 2002, 12, 3595. (c) Hackbarth, C. J.;
Chen, D. Z.; Lewis, J. G.; Clark, K.; Mangold, J. B.; Cramer,
J. A.; Margolis, P. S.; Wang, W.; Koehn, J.; Wu, C.; Lopez,
S.; Withers, G., III; Gu, H.; Dunn, E.; Kulathila, R.; Pan, S.-
H.; Porter, W. L.; Jacobs, J.; Trias, J.; Patel, D. V.; Weid-
mann, B.; White, R. J.; Yuan, Z. Antimicrob. Agents Che-
mother. 2002, 46, 2752. (d) Yuan, Z.; Trias, J.; White, R. J.
DDT 2001, 6, 954. (e) Pei, D. Emerg. Ther. Targets 2001, 5, 23.
(f) Meinnel, T. Parasitol. Today 2000, 16, 165. (g) Giglione, C.;
Pierre, M.; Meinnel, T. Mol. Microbiol. 2000, 36, 1197.
5. Hao, B.; Gong, W.; Rajagopalan, P. T. R.; Zhou, Y.; Pei,
D.; Chan, M. K. Biochemistry 1999, 38, 4712.
In summary, we have discovered benzothiazolylidene-
hydroxamic acid as novel class of PDF inhibitors. Some
of these inhibitors showed moderate antibacterial activ-
ity. In addition, the computational docking simulation
6. Chikashita, H.; Takegami, N.; Yanase, Y.; Itoh, K. Bull.
Chem. Soc. Jpn 1989, 62, 3389.
7. 5a: Orange crystals, mp 63.7–65.9 ꢂC. 1H NMR (DMSO-
d6): d 1.16 (3H, t, J=7.2 Hz), 3.46 (2H, q, J=7.2 Hz), 6.69
(1H, m), 6.78–6.79 (2H, m), 6.96 (1H, s), 6.98 (1H, m), 8.70
(1H, br s), 10.34 (1H, s). 5b: Orange crystals, mp
118.6–119.4 ꢂC. 1H NMR (DMSO-d6): d 0.92 (3H, t, J=7.2
Hz), 1.52–1.65 (2H, m), 3.36 (2H, t, J=7.2 Hz), 6.68 (1H, m),
6.77–6.79 (2H, m), 6.96 (1H, s), 6.98 (1H, m), 8.69 (1H, s),
10.33 (1H, s). 5c: Orange crystals, mp 126.5–126.9 ꢂC. 1H
NMR (DMSO-d6): d 0.91 (3H, t, J=7.2 Hz), 1.30–1.42 (2H,
m), 1.49–1.59 (2H, m), 3.40 (2H, t, J=7.2 Hz), 6.68 (1H, m),
6.76–6.79 (2H, m), 6.95 (1H, s), 6.98 (1H, m), 8.70 (1H, s),
10.34 (1H, s). 5d: Orange crystals, mp 108.1–108.3 ꢂC. 1H
NMR (DMSO-d6): d 0.87 (3H, d, J=7.2 Hz), 1.32–1.33 (4H,
m), 1.54–1.56 (2H, m), 3.39 (2H, t, J=7.2 Hz), 6.67 (1H, m),
6.78–6.79 (2H, m), 6.95 (1H, s), 6.98 (1H, m), 8.67 (1H, s),
10.31 (1H, s). 8c: Colorless crystals, mp 179.2–189.9 ꢂC. 1H
NMR (DMSO-d6): d 0.89 (3H, t, J=7.2 Hz), 1.24–1.37 (2H,
m), 1.60–1.70 (2H, m), 4.14 (2H, t, J=7.2 Hz), 7.45 (1H, m),
7.60 (1H, m), 7.74 (1H, m), 7.97 (1H, s), 7.99 (1H, m), 9.19
(1H, s), 10.20 (1H, s).
Figure 3. Binding mode of the most stable docking model for 5c
within the substrate pocket of PDF. Hydrogen bonds (distances less
than 2.8 A between the hydrogen bonded to an H-Bond donor and an
acceptor) are shown in dotted yellow lines. Carbon atoms are shown
in white, hydrogens are cyan, nitrogens are blue, oxygens are red, and
sulfur is yellow, except those of the amino acids (orange). Zn atom is
shown as a red ball.
8. (a) Lazennec, C.; Meinnel, T. Anal. Biochem. 1997, 244, 180.
(b) Chen, D. Z.; Patel, D. V.; Hackbarth, C. J.; Wang, W.;
Dreyer, G.; Young, D. C.; Margolis, P. S.; Wu, C.; Ni, Z. J.;
Trias, J.; White, R. J.; Yuan, Z. Biochemistry 2000, 39, 1256.
9. (a) Chemotherapy 1981, 29, 76. (b) Chemotherapy 1990, 38,
103.

3276
W. Takayama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3273–3276
10. Becker, A.; Schlichting, I.; Kabsch, W.; Groche, D.;
Schultz, S.; Wagner, A. F. Nat. Struct. Biol. 1998, 5, 1053.
11. Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. J. Mol.
Biol. 1996, 261, 470.
13. Clements, J. M.; Beckett, R. P.; Brown, A.; Catlin, G.;
Lobell, M.; Palan, S.; Thomas, W.; Whittaker, M.; Wood, S.;
Salama, S.; Baker, P. J.; Rodgers, H. F.; Barynin, V.; Rice,
D. W.; Hunter, M. G. Antimicrob. Agents. Chemother. 2001,
45, 563.
12. Powell, M. J. D. Math. Program 1977, 12, 241.