Structure-activity relationship study of the bacterial actin-like protein MreB inhibitors: Effects of substitution of benzyl group in S-benzylisothiourea

Article abstract of DOI:10.1271/bbb.60443

We comprehensively investigated the effects of substitution of the benzyl group in S-benzylisothiourea derivatives on antibacterial activity, because we found previously that some substitutions enhanced it. A 2,4-Cl ₂-derivative was found to be the most effective compound, it was stronger than the original one in Gram-negative rod shaped-bacteria such as Escherichia coli and Salmonella typhimurium.

Full text of DOI:10.1271/bbb.60443

Biosci. Biotechnol. Biochem., 71 (1), 246–248, 2007
Note
Structure-Activity Relationship Study of the Bacterial Actin-Like
Protein MreB Inhibitors: Effects of Substitution
of Benzyl Group in S-Benzylisothiourea
y
1
Noritaka IWAI,1;2; Takuya FUJII,1 Hirokatsu NAGURA,1 Masaaki WACHI,1 and Tomoya KITAZUME
1Department of Bioengineering, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226-8501, Japan
2Frontier Collaborative Research Center, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226-8501, Japan
Received August 10, 2006; Accepted October 23, 2006; Online Publication, January 7, 2007
[doi:10.1271/bbb.60443]
We comprehensively investigated the effects of sub-
stitution of the benzyl group in S-benzylisothiourea
derivatives on antibacterial activity, because we found
previously that some substitutions enhanced it. A 2,4-
Cl2-derivative was found to be the most effective
compound, it was stronger than the original one in
Gram-negative rod shaped-bacteria such as Escherichia
coli and Salmonella typhimurium.
against multi-drug resistant pathogens.
Our previous structure-activity relationship study
showed that S-benzylisothiourea structure is necessary
and sufficient to induce spherical cells, and that 3- and/
or 4-chloro-substitution of the S-benzyl group enhance
antibacterial activity.5) This led us to expect that mo-
dification of the benzyl group in S-benzylisothiourea
would improve the antibacterial activity as well as
the antibacterial spectrum of S-benzylisothiourea com-
pounds. In this study, we investigated effects of sub-
stitutions of benzyl group in S-benzylisothiourea struc-
ture on antibacterial activity comprehensively. The re-
sults obtained in this study should be a hint for devel-
oping new bacterial actin-like protein inhibitors.
Chemical compounds, except for entry 1, were syn-
thesized by the method reported previously. Briefly,
thiourea was suspended in dehydrated ethanol, and cor-
responding benzyl halide was added to the suspension.
The mixture was heated at 130 ꢀC to reflux for several
h, and then cooled to room temperature. The reaction
mixture was concentrated under vacuum and the residue
was diluted with methanol. The product was purified by
recrystallization from diethyl ether. The structure of the
compound was confirmed by 1H-NMR spectrometry and
elemental analysis.
Minimal inhibitory concentrations (MIC) of the
compounds for E. coli and several other bacteria were
determined by the agar-dilution method using Mueller
Hinton Broth (Becton Dickinson and Company, MD).
The tested bacterial strains were E. coli MG1655,
Pseudomonas putida NBRC14164 (the same as
ATCC12633), S. typhimurium NBRC13245 (the same
as LT2), Bacillus subtilis 168, and Staphylococcus
aureus subsp. aureus NBRC15035 (the same as
ATCC29213).
Key words: A22; actin-like protein; benzylisothiourea;
MreB; rod-shape determination
To overcome bacterial infectious diseases caused by
multi-drug resistant pathogens, it is important to find a
new drug that has a new target. For this purpose, we
developed a new screening system for inhibitors of
bacterial chromosome partitioning.1) The assay system is
designed to detect production of chromosome-less cells
(anucleate cells) caused by inhibition of chromosome
partitioning in E. coli. Random screening for inhibitors
of chromosome partitioning in E. coli was done using
this assay system, and a novel S-benzylisothiourea
derivative, S-(3,4-dichlorobenzyl) isothiourea, tentative-
ly named A22, was found.2) A22 induced spherical cells
and spherical anucleate cells in E. coli. Genetic analysis
revealed that the target molecule of A22 was the actin-
like cytoskeletal protein MreB in Caulobacter cres-
centus and E. coli.3,4) Using A22 as a probe, it was
found that the actin-like MreB cytoskeleton was in-
volved in chromosome segregation, at least in C. cres-
centus and E. coli. A22 is active only against Gram-
negative rods, as tested so far. Since no antibacterial
agents targeting bacterial cytoskeletal proteins have
been introduced into the chemotherapeutic treatment of
bacterial infectious diseases, A22 is expected to be a
lead compound in developing new drugs effective
To examine the position effect of mono-substitution
y
To whom correspondence should be addressed. Tel: +81-45-924-5775; Fax: +81-45-924-5780; E-mail: niwai@bio.titech.ac.jp
Abbreviations: MIC, minimal inhibitory concentration; PBP, penicillin-binding protein

Structure-Activity Relationship of MreB Inhibitors
247
Table 1. Summary of the Structure-Activity Relationship Study of S-Benzylisothiurea Derivatives
Compoundsa
NH
HX
R
S
NH2
MIC (mg/ml)b
S. typhimurium
R
X
E. coli
P. putida
B. subtilis
S. aureus
Shapec
1
2
Ph
Cl
Cl
Cl
Cl
Br
Cl
Br
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Br
Cl
Cl
100
12.5
50
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
100
>100
25
>100
100
>100
50
>100
>100
>100
>100
>100
>100
>100
50
Sphereꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂꢂ
Sphere
Sphereꢂꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂꢂ
Sphereꢂꢂ
Sphere
Sphereꢂꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂ
Sphereꢂ
2-ClC6H4
3-ClC6H4
4-ClC6H4
3
4
3.13
50
3.13
>100
>100
>100
>100
50
5
2-BrC6H4
3-BrC6H4
50
>100
6
>100
3.13
12.5
7
4-BrC6H4
4-FC6H4
4-IC6H4
6.25
12.5
8
9
12.5
25
25
25
>100
25
>100
>100
>100
>100
>100
>100
25
10
11
12
13
14
15
16
17
18
19
4-NO2C6H4
4-CH3C6H4
4-CF3C6H4
4-CH3OC6H4
4-C2H5C6H4
2,3-Cl2C6H3
2,4-Cl2C6H3
2,5-Cl2C6H3
2,6-Cl2C6H3
3,4-Cl2C6H3
(A22)
12.5
50
25
>100
>100
>100
>100
>100
>100
>100
>100
100
>100
>100
>100
>100
>100
>100
25
1.56
12.5
12.5
3.13
1.56
>100
>100
>100
>100
25
25
3.13
20
21
3,5-Cl2C6H3
2,4,6-Cl3C6H2
Cl
Cl
>100
100
>100
>100
>100
>100
>100
100
>100
>100
Sphereꢂꢂ
Sphereꢂ
aCompounds were hydrochloride or hydrobromide salt.
bMICs were determined for E. coli MG1655, P. putida NBRC14164 (the same as ATCC12633), S. typhimurium NBRC13245 (the same as LT2), B. subtilis 168, and
S. aureus subsp. aureus NBRC15035 (the same as ATCC29213).
cCell shapes were determined from E. coli cultures treated with the compounds at 3ꢃ MIC (ꢂ) or at 300 mg/ml (ꢂꢂ) at 30 ꢀC for 16 h.
of the benzyl group, mono-chloro- and mono-bromo-
substitutions at positions 2-, 3- and 4- of the benzyl
group were done. All compounds showed activity
inducing spherical cells in E. coli. Antibacterial activity
against E. coli decreased in the order 4- > 2- > 3- in
both chloro- and bromo-substitutions (Table 1, entries
2–7). A similar tendency was also seen in S. typhimu-
rium. These compounds did not show antibacterial
activity against P. putida, but induced spherical cells at
100 mg/ml (data not shown). This is probably because of
multi-drug efflux systems those are able to accommo-
date a variety of structurally unrelated antimicrobial
agents in P. putida.6) It is concluded that the most
effective position of the benzyl group in S-benzyliso-
thiourea is the 4-site.
To identify effective substituents, several 4-substitut-
ed compounds were synthesized and examined (Table 1,
entries 4, 7–14). The most effective substituents were
chloro-groups and bromo-groups for E. coli, but the
bromo group was slightly less effective for S. typhimu-
rium. The 4-Fluoro group, which is smaller than the
chloro group, was less effective, while the 4-iodo group,
which is lager than the bromo group, was also less
effective. The order of antibacterial activity of 4-mono-
halogenized compounds was Cl ꢁ Br > F ꢁ I. It is
suggested that a better molecular size-fitting in the
supposed drug-biding pocket of the MreB protein is Cl-
or Br-. Substitutions by alkyl groups (CH3-, C2H5-) and
several other substituents (CH3O-, NO2-, CF3-) were all
less effective.
Next, di- and tri-chloro-substituted compounds were
synthesized and examined (Table 1, entries 15–21). In
the case of dichloro-substituted compounds, as expected
from the results for mono-substituted compounds, 2,4-
Cl2-substituted compounds showed the most effective
antibacterial activity against Gram-negative bacteria,
and also induced spherical formation (Fig. 1C). Con-
trary to our expectations, a compound substituted at both
meta-sites, 3,5-Cl2-, was less effective than the non-
substituted compound. It appears that the mono-meta-
substituted compound is acceptable to the supposed
drug-binding pocket of MreB, but simultaneous sub-
stitution in another meta-site prevents docking with
MreB. The 2,4,6-Cl3-substituted compound was less
effective.
Most of the compounds tested in this study showed no
antibacterial activity against B. subtilis, but NO2- and
CF3-substituted compounds showed anti-B. subtilis ac-
tivity, although weak (the MICs of NO2- and CF3- were
25 and 50 mg/ml respectively). However, they showed
almost no effect on cell shape (Fig. 1E, and not shown).
These compounds also showed antibacterial activity

248
N. IWAI et al.
Acknowledgments
A
D
B
E
C
This work was supported in part by Grants-in-Aid for
Scientific Research (B) (15380059, to W.M.) from the
Japan Society for the Promotion of Science and the
Grant of the 21st Century COE Program from the
Ministry of Education, Culture, Sports, Science and
Technology of Japan.
References
1) Wachi, M., Iwai, N., Kunihisa, A., and Nagai, K.,
Irregular nuclear localization and anucleate cell produc-
tion in Escherichia coli induced by a Ca2þ chelator,
EGTA. Biochimie, 81, 909–913 (1999).
10 µm
2) Iwai, N., Nagai, K., and Wachi, M., Novel S-benzyliso-
thiourea compound that induces spherical cells in Esche-
richia coli probably by acting on a rod-shape-determining
protein(s) other than penicillin-binding protein 2. Biosci.
Biotechnol. Biochem., 66, 2658–2662 (2002).
3) Gitai, Z., Dye, N. A., Reisenauer, A., Wachi, M., and
Shapiro, L., MreB actin-mediated segregation of a
specific region of a bacterial chromosome. Cell, 120,
329–342 (2005).
Fig. 1. Effects of S-Benzylisothiourea Derivatives on E. coli and
B. subtilis Cell Shape.
E. coli MG1655 (A–C) and B. subtilis 168 (D, E) cells growing
at 30 ꢀC were treated for 4 h with: A, no treatment: B, 75 mg/ml of
10: C, 4.58 mg/ml of 16: D, no treatment: or E, 75 mg/ml of 10.
Differential interference contrast microphotographs are shown. Bar,
10 mm.
4) Kruse, T., Blagoev, B., Lobner-Olesen, A., Wachi, M.,
Sasaki, K., Iwai, N., Mann, M., and Gerdes, K., Actin
homolog MreB and RNA polymerase interact and are
both required for chromosome segregation in Escherichia
coli. Genes Dev., 20, 113–124 (2006).
5) Iwai, N., Ebata, T., Nagura, H., Kitazume, T., Nagai, K.,
and Wachi, M., Structure-activity relationship of S-
benzylisothiourea derivatives to induce spherical cells in
Escherichia coli. Biosci. Biotechnol. Biochem., 68, 2265–
2269 (2004).
6) Kieboom, J., and de Bont, J., Identification and molecular
characterization of an efflux system involved in Pseudo-
monas putida S12 multidrug resistance. Microbiology,
147, 43–51 (2001).
7) Bork, P., Sander, C., and Valencia, A., An ATPase
domain common to prokaryotic cell cycle proteins, sugar
kinases, actin, and hsp70 heat shock proteins. Proc. Natl.
Acad. Sci. USA, 89, 7290–7294 (1992).
against Staphylococcus aureus, which does not possess
the mreB homolog. Therefore, these phenomena were
probably due to non-specific toxic effects of these
compounds.
We found S-(2,4-dichlorobenzyl)isothiourea to be a
more effective compound than the original A22 (3,4-
Cl2-). With respect to the antibacterial spectrum, we did
not find a compound effective against Gram-positive rod
B. subtilis. This problem remains to be solved.
It was recently found that the MreB protein is a direct
target molecule of A22 by analysis of spontaneous A22-
resistant mutants of C. cresentus and E. coli. MreB
protein is known as a bacterial actin-like cytoskeltal
protein, and MreB homologs are found in various non-
coccal bacteria, such as rod-shaped bacteria, helical
bacteria, and hyphal bacteria.7–9) It is expected that S-
benzylisothiourea derivatives will prove useful as lead-
ing compounds for developing novel antibacterial drugs,
and as a molecular probes for analyzing the cellular
functions of MreB.
8) van den Ent, F., Amos, A. L. A., and Lowe, J.,
¨
Prokaryotic origin of the actin cytoskeleton. Nature,
413, 39–44 (2001).
9) Jones, C. A., Carballido-Lopez, R., and Errington, J.,
Control of cell shape in bacteria: helical, actin-like
filaments in Bacillus subtilis. Cell, 104, 913–922 (2001).