Synthesis of rigidly-linked vancomycin dimers and their in vivo efficacy against resistant bacteria

Article abstract of DOI:10.1039/b613319c

A novel and efficient avenue for the preparation of dimeric vancomycins is described, and the dimers exhibited excellent antibacterial activities in the murine infection model. The Royal Society of Chemistry.

Full text of DOI:10.1039/b613319c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis of rigidly-linked vancomycin dimers and their in vivo efficacy
against resistant bacteria{
Jun Lu,a Osamu Yoshida,b Sayaka Hayashic and Hirokazu Arimoto*c
Received (in Cambridge, UK) 13th September 2006, Accepted 10th October 2006
First published as an Advance Article on the web 24th October 2006
DOI: 10.1039/b613319c
A novel and efficient avenue for the preparation of dimeric
vancomycins is described, and the dimers exhibited excellent
antibacterial activities in the murine infection model.
The emergence of multidrug-resistant gram-positive pathogens
represented by MRSA (methicillin resistant S. aureus) is of concern
in clinical institutions. The emergence of vancomycin-resistant
bacteria (VRE:1 vancomycin-resistant enterococci; VRSA:2 vanco-
mycin-resistant S. aureus) is of particular concern,3 because
vancomycin is the last resort for the treatment nosocomial
infections by multi-resistant gram-positive bacteria. Currently,
Zyvox (Linezolid; Pfizer), Synercid (Monarch and King), and
Tygacil (Wyeth) are available for VRE infections, but have
problems such as side effects and drug resistance.4 Here we
describe the synthesis of novel vancomycin dimers with excellent
antibacterial activity both in vitro and in vivo.
Fig. 1 Actinocin, a natural privileged scaffold for dimeric peptidic
ligands.
Two cyclic depsipeptide subunits of actinomycin were arranged
in parallel on a rigid tricyclic actinocin template. This led to the
design of novel vancomycin dimers (Fig. 2). It should be noted that
the actinocin-type linkers are quite rigid. Since flexible linkers (e.g.
–(CH2)n–) are thought to reduce the entropic advantages in
dimeric binding modes,10 the rigid nature of the linker was
expected to minimize the entropic loss arising from the linker unit
in dimeric binding events. There are only four rotatable bonds
between the linked vancomycin molecules in 1 (Fig. 2). Molecular
modeling studies (data not shown) revealed that two vancosamine
nitrogen atoms should exist approximately 7 angstroms apart
in 1.11,12
Vancomycin binds to a bacterial cell wall intermediate by
recognizing the D-Ala–D-Ala terminus of lipid intermediates, and
thus exhibits antibacterial activities against Gram-positive bac-
teria.5 The dipeptide terminal of lipid II is transformed to D-Ala–
D-lactate in VREs (VanA and VanB phenotypes) and VRSAs
(Van A phenotypes), and this lowers the binding affinity of
vancomycin toward the mutated lipid II.6
With the biosynthetic pathway to actinocin nuclei in mind, an
alkoxynitrobenzaldehyde 6 was introduced to vancomycin by
The hypothesis that the associated dimer of vancomycin plays
important roles in the recognition events of the lipid intermediates7
prompted synthetic chemists to design covalently-linked glycopep-
tide dimers. It has been shown that the choice of the linker
structure significantly affects the antibacterial profiles.8
In order to develop lead-drug compounds based on the
dimerization strategy, advances in novel synthetic protocols are
necessary. The polyfunctionalized nature of vancomycin limits the
reactions applicable for dimerization. It should be possible to link
bulky vancomycin molecules without tedious protection/deprotec-
tion procedures over a short distance. Most of the reported dimers
have long flexible linkers, and some include biologically unstable
disulfide bonds in the linker portion.
To overcome these problems, we focused our attention on the
use of natural privileged templates found in actinomycin9 (Fig. 1).
aDepartment of Chemistry, Graduate School of Science, Nagoya
University, Chikusa, Nagoya, 464-8602, Japan
bDiscovery Research Laboratories, Shionogi & Co., Ltd., Fukushima,
Osaka, 553-0002, Japan
cGraduate School of Life Sciences, Tohoku University, Sendai, 981-
8555, Japan. E-mail: arimoto@biochem.tohoku.ac.jp;
Fax: +81 22 717 8806; Tel: +81 22 717 8803
{ Electronic supplementary information (ESI) available: Experimental
procedures and selected physical data for compounds 1–5. See DOI:
10.1039/b613319c
Fig. 2 Structures of vancomycin dimers with actinocin-based linkers.
Chem. Commun., 2007, 251–253 | 251
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 1 The first set of in vitro evaluations of vancomycin dimers
with different linker lengthsa
Dimer 1
2 (N.T.)
16 (8)
2
3
4
Vancomycin
S. aureus FDA209P
JC-1
S. aureus SR3637
(MRSA)
2
2 (N.T.)
1
0.5
1
32 16 (2)
64 16 (4)
32
32
S. aureus ATCC700787 32 (32)
(VISA:Mu50)
E. faecalis ATCC29212 2 (N.T.)
1
2
2
4 (N.T.)
8 (4)
1
8
2
.64
E. faecalis SR7914
(VRE:VanA)
8 (4)
2 (1)
E. faecalis SR23630
(VRE:VanB)
E. faecium NCTC7171 0.5 (N.T.)
4
4 (0.5)
2
32
1
2
2 (N.T.) 0.5
8 (N.T.) 32 .64
0.5
E. faecium SR7917
(VRE:VanA)
E. faecium SR23598
(VRE:VanB)
a
32 (N.T.)
Scheme 1 Synthesis of vancomycin dimer 1 by mild oxidative coupling
of an amino-phenol. Reagents and conditions: (i) aldehyde 6, DMF–MeOH
(1 : 1), i-Pr2NEt, 70 uC, 2 h; NaBH3CN (4 eq.), 70 uC, 20 h (59%); (ii) Pd–
C/H2, MeOH; (iii) p-quinone, MeOH, RT, 20 h (68%).
1 (0.25)
1
2 (0.5) 1 .64
Minimum inhibitory concentration (MIC) in micrograms per
millilitre as determined by microdilution broth assay in vitro. TFA
salts of dimers 1–4 were employed for the evaluation. Antibacterial
activities of dimers 1 and 3 were also examined with their HCl salts,
and the MIC values are shown in parentheses.
reductive alkylation. The regioselectivity of the alkylation was
proven by fragmentation analysis of ESI mass spectra (Scheme 1).
Conversion of a nitro group to an amine and removal of benzyl
ether were achieved in a one-pot reaction by catalytic hydrogena-
tion. The obtained aminophenol intermediate was unstable, and
thus was employed directly in the dimerization step. Treatment of
the aminophenol in methanol by quinone furnished a pink-colored
dimer at room temperature. After HPLC purification (Develosil
ODS HG-5, w 20 mm 6 250 mm, CH3CN : H2O : TFA 33 : 66 :
0.1, flow rate 2 mL min21, tR = 21.82 min) and lyophilization,
dimer 1 was isolated in 68% yield as a TFA salt. Various reactions,
such as amide formation, disulfide coupling, and cross metathesis,
have been employed for the synthesis of vancomycin dimers. Our
oxidative coupling to actinocin-linked dimers has notable advan-
tages, including mild reaction conditions, high yield and selectivity
and giving stable products in the in vivo test. Dimers 2, 3, and 4
were also prepared.
Table 2 The second set of in vitro evaluations of antibacterial
activitya
1
3
Vancomycin
.64
S. aureus VRSA-2
(VCM-resistant)
E. faecium SR7940
(VRE:VanA)
8 (4)
8 (4)
32 (1)
8 (4)
16 (2)
.64
S. aureus Smith
S. pneumoniae Type1
S. pneumoniae SR23928 ¡0.063 (¡0.063) 1 (0.5)
(PRSP)
16 (4)
1
0.125 (¡0.063) 1 (0.25)
0.5
0.5
S. pneumoniae tupelo
(VTSP)
0.125 (¡0.063) 1 (0.125)
0.25
a
Minimum inhibitory concentration in micrograms per millilitre as
determined by microdilution broth assay in vitro. The TFA salts and
HCl salts of dimers 1 and 3 were employed for the evaluation.
Antibacterial activities of the HCl salts are shown in parentheses.
The antibacterial activities of the actinocin-based dimers were
evaluated with a set of vancomycin-susceptible and -resistant
bacterial strains (Table 1). The synthetic dimers were found to be
active against VREs. No clear correlation was found between the
linker structures and the antibacterial activities. We next selected 1
and 3 and conducted further investigations. The potencies of
dimers for vancomycin-susceptible S. aureus were somehow lower
than those for enterococci, but the problem could be solved by
changing the counter anion of the dimers 1 and 3 from
trifluoroacetate. The minimum inhibitory concentrations (MICs)
of dimer-HCl salt are shown in parentheses (Table 1 and 2). The
antibacterial activities of these dimer-HCl salts were enhanced 2 to
32-fold, not only for S. aureus but also for other strains including
vancomycin-resistant strains. The dimers thus possessed balanced
and practical in vitro antibacterial activities against both
vancomycin-sensitive and vancomycin-resistant strains.
These promising antibacterial results prompted us to examine
the efficacy of the selected dimers in vivo (Table 3).13 A test was
conducted with S. pneumoniae-infected mouse, which is resistant to
penicillin, using 1, 3, and linezolid because the treatment of upper
Table 3 In vivo efficacy in murine lung infection model with
S. pneumoniaea
Two-log kill
dose/mg kg21
Three-log kill
dose/mg kg21
Compounds
MIC/mg mL21
1
4.49
34.8
21.3
6.54
72.2
33.1
¡0.063
0.125
0.5
3
Linezolid
a
Animal: CBA mouse (6 weeks old, N = 5). Infection: S. pneumoniae
105 cfu mouse21
was administered
We next conducted investigations with the second set of
bacteria, including vancomycin-resistant S. aureus (VRSA:
Table 2). VRSA and VanA-type VRE share the same vanA-type
resistant gene. However, the pathogenic and more harmful nature
of the parent S. aureus strain than those of enterococci leads to
VRSA being a cause of serious concern in clinics. The dimers were
found to be quite active against VRSA.
SR11031 (PRSP:
6
6
)
intranasally (i.n.). Compounds (HCl salt) were administered by the
intravenous route (i.v.) 17 h and 24 h post infection. Viable count
per lung was carried out at 48 h postinfection. A 2-log kill dose was
defined as the dose required to produce a decrease in titer of 2 log
CFU g21 from pretreatment controls. Similarly, 3-log kill doses were
defined as the doses required to produce decreases in titer of 3 log
CFU g21
.
252 | Chem. Commun., 2007, 251–253
This journal is ß The Royal Society of Chemistry 2007

View Article Online
3 D. Kahne, C. Leimkuhler, W. Lu and C. Walsh, Chem. Rev., 2005, 105,
425–448.
4 As Zyvox, Synercid and Tygacil inhibit the protein synthesis of bacteria,
all differ from vancomycin in their mechanisms of action.
and lower respiratory tract infections by multi-resistant
S. pneumoniae has become a therapeutic problem. Among the
tested compounds, 1 exhibited the most promising bactericidal
effect. Dimer 1 at 10 mg kg21 dose yielded undetectable bacterial
counts (i.e. ,50 CFU mL21, data not shown). Pharmacokinetic
profiles greatly affect the in vivo potency, which includes the
protein binding, and the tissue penetration of the drug showing
that there are significant barriers to attaining good in vivo activity.
It is noteworthy that 1 showed a preliminary but promising
curative effect in vivo.
5 (a) D. H. Williams, M. P. Williamson, D. W. Butcher and
S. J. Hammond, J. Am. Chem. Soc., 1983, 105, 1332–1339; (b)
P. Reynolds, Eur. J. Clin. Microbiol. Infect. Dis., 1989, 943–950.
6 (a) C. T. Walsh, S. L. Fisher, I. S. Park, M. Prahalad and Z. Wu, Chem.
Biol., 1996, 3, 21–28; (b) T. D. H. Bugg, G. D. Wright, S. Dutka-Malen,
M. Arthur, P. Courvalin and C. T. Walsh, Biochemistry, 1991, 30,
10408–10415; (c) R. Xu, G. Greiveldinger, L. E. Marenus, A. Cooper
and J. A. Ellman, J. Am. Chem. Soc., 1999, 121, 4898–4899.
7 D. H. Williams and B. Bardsley, Angew. Chem., Int. Ed., 1999, 38,
1172–1193.
8 (a) U. N. Sundram, J. H. Griffin and T. I. Nicas, J. Am. Chem. Soc.,
1996, 118, 13107–13108; (b) J. H. Griffin, M. S. Linsell, M. B. Nodwell,
Q. Chen, J. L. Pace, K. L. Quast, K. M. Krause, L. Farrington,
T. X. Wu, D. L. Higgins, T. E. Jenkins, B. G. Christensen and J. K.
Judice, J. Am. Chem. Soc., 2003, 125, 6517–6531; (c) D. R. Stack. and
R. G. Thompson, Eur. Pat. EP 0801075 A1, 1997; (d) K. C. Nicolaou,
R. Hughes, S. Y. Cho, N. Winssinger, C. Smthurst, H. Labischinski and
R. Endermann, Angew. Chem., Int. Ed., 2000, 39, 3823–3828; (e)
K. C. Nicolaou, R. Hughes, S. Y. Cho, N. Winssinger, H. Labischinski
and R. Endermann, Chem.–Eur. J., 2001, 3824–3843.
At present, it is not clear how actinocin-based vancomycin
dimers exhibit their enhanced antibacterial activities. Biochemical
studies are underway to investigate the mechanism of these
actions. The answer appears to be more complex than simple
cooperative binding to transformed peptidoglycan intermediates.8b
The novel and efficient synthesis route presented here should be
applicable to the preparation of highly potent dimeric vancomy-
cins that are effective in the animal infection model.
We are grateful to Mr Kenichi Okazaki (Shionogi) for the
evaluation of compounds in vivo, and to Prof. Daisuke Uemura
(Nagoya Univ.), Dr Hideki Maki, and Mr Kenji Miura (Shionogi)
for their valuable suggestions. This work was supported in part by
PRESTO, JST, and Grants-in-Aid (Nos. 16310150 and 17035039)
for Scientific Research from the Ministry of Education, Sports,
Science and Technology (MEXT), Japan.
9 (a) S. A. Waksman and H. B. Woodruff, Proc. Soc. Exp. Biol. Med.,
1940, 45, 609–614; (b) H. Brockmann, Angew. Chem., 1954, 66, 1–10.
10 Whitesides also pointed out that a flexible linker should reduce the
entropic advantages in dimeric binding modes;M. Mammen, S.-K. Choi
and G. M. Whitesides, Angew. Chem., Int. Ed., 1998, 37, 2754–2794.
11 Recognition of the D-Ala–D-Ala motif by the associated vancomycin
dimer in the crystal structure seemed to be a good starting point for the
rational design of covalently linked vancomycin dimers. Amino groups
of vancosamine sugars were shown to be located at distances ranging
from 7 to 11 angstroms with the ‘‘back-to-back dimers’’ and located at
shorter distances than the ‘‘face-to-face dimers’’; P. J. Loll and
P. H. Axelsen, Annu. Rev. Biophys. Biomol. Struct., 2000, 29, 265–289.
12 In 2000, Nitanai and Aoki deposited the crystal structure of diacetyl-
Lys–D-Ala–D-Ala complexed with vancomycin with the Protein Data
Bank (code: 1FVM). The distance of vancosamine amino groups in
associated back-to-back dimers was, at the shortest, 7.8 angstroms.
13 In all animal studies, experiments were performed in compliance with
the relevant laws and institutional guidelines and were approved by the
Institutional Animal Care and Use Committee of Shionogi Discovery
Research Laboratories.
Notes and references
1 (a) R. E. Leclerq, R. E. Derlot, J. Duval and P. Courvalin, N. Engl.
J. Med., 1988, 319, 157–161; (b) A. H. C. Uttley, C. H. Collins,
J. Naidoo and R. C. George, Lancet, 1988, 1, 57–58.
2 D. M. Sievert, M. L. Boulton, G. Stoltman, D. Johnston,
M. G. Stobierski, F. P. Downes, P. A. Somsel, J. T. Rudrik,
W. Brown, W. Hafeez, T. Lundstrom, E. Flanagan, R. Johnson,
J. Mitchell and S. Chang, Morbid. Mortal. Weekly Rep., 2002, 51,
565–567.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 251–253 | 253