Affinity of a vancomycin polymer with bacterial surface models

Article abstract of DOI:10.1016/S0040-4039(01)00429-4

The affinity between a vancomycin polymer (3) and cell wall intermediate mimics of vancomycin resistant bacteria (VRE) was determined by use of surface plasmon resonance (SPR). The increased affinity of 3 over monomeric vancomycin derivatives 1 and 2 suggests the importance of tighter binding to VRE surfaces in the enhanced antibacterial activities of 3.

Full text of DOI:10.1016/S0040-4039(01)00429-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3347–3350
Affinity of a vancomycin polymer with bacterial surface models
a,
a
a
b
Hirokazu Arimoto, * Takehisa Oishi, Manabu Nishijima and Tomoya Kinumi
a
Department of Chemistry, Faculty of Science, Shizuoka University, 836 Ohya, Shizuoka 422-8529, Japan
b
Department of Biochemistry and Cell Biology, National Institute of Infectious Diseases, Tokyo 162-8640, Japan
Received 16 February 2001; revised 7 March 2001; accepted 9 March 2001
Abstract—The affinity between a vancomycin polymer (3) and cell wall intermediate mimics of vancomycin resistant bacteria
VRE) was determined by use of surface plasmon resonance (SPR). The increased affinity of 3 over monomeric vancomycin
derivatives 1 and 2 suggests the importance of tighter binding to VRE surfaces in the enhanced antibacterial activities of 3.
2001 Elsevier Science Ltd. All rights reserved.
(
©
The evolution of multi-resistant bacteria is a serious
worldwide problem. The vancomycin-class glycopeptide
antibiotics are the last resort against MRSA (methicilin
resistant Staphylococcus aureus). However, VRE (van-
comycin resistant enterococci ) has emerged, and the
possibility of this resistance being transferred to S.
aureus is a cause of major concern. The binding of
vancomycin is believed to be important in interference
with bacterial peptideglycan biosynthesis.
The utility of the SPR methods has been demonstrated
in a variety of applications in the past decade, including
a recent determination of the binding constant of van-
1
OH
H
N
Me
R= H
vancomycin 1 O H
O
HO
HO
O
Me
R
O
HO
O
N
Cl
O
O
O
O
O
H
In our efforts toward the design of new semisynthetic
antibiotics, we recently reported that a vancomycin
polymer 3 displayed potency against VRE up to 60
OH
O
HO
Cl
O
H
H
N
H
N
H
N
O
N
H
N
H
N
H
Me
Me
H
H
monomer 2
O
Ph
H
O
HN
O
2
O
H
O
times greater than vancomycin itself (Fig. 1).
HOOC
Me
H
NH2
N
OH
OH
n
CH3
HO
O
H
The mechanism of this enhancement in antibacterial
activity remains elusive, but might derive from tighter
binding of the polymer to the resistant bacteria surface.
MIC of 3 (VRE, Van B)
mg / mL-1
polymer 3
2
It has also been demonstrated that the array of sugars
on the polymers could enhance its binding to the recep-
tors due to the ‘multi-valent or cluster effect’, which
Figure 1. Structure and antibacterial activity of a vancomycin
polymer 3.
3
suggests that a vancomycin polymer could have similar
types of interactions (Fig. 2).
To probe the basis for the enhanced activity of van-
comycin polymer 3, we examined the binding of 3 with
bacterial cell wall model peptides by surface plasmon
resonance (SPR).
SPR is a powerful method for examination of the
affinity between surface presenting receptor molecules
4
and ligand in solution.
Figure 2. Schematic presentation of the possible interaction
of a vancomycin polymer with a bacterial cell wall biosynthe-
sis intermediate.
*
0
Corresponding author.
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00429-4

3
348
H. Arimoto et al. / Tetrahedron Letters 42 (2001) 3347–3350
NH2
O
O
O
independent flow cells to form amide linkages with the
activated dextran layer. The remaining activated car-
boxylate groups were capped by injection of
ethanolamine. The control flow cell was prepared by
immobilizing only ethanolamine without the addition
of 4 and 5.
i-v
i-v
OH
H2N
X
N
H
OH
5
D-alanine
O
4
X= NH
vancomycin-susceptible type
D-Ala-D-Ala)
vi
(
OH
O
5
X= O
OH
To the resultant SPR sensors, vancomycin solution (in
vancomycin-resistant type
(D-Ala-D-Lac)
1
0 mM HEPES buffer, containing NaCl 75 mM, sur-
D-lactic acid
factant P20 0.5 ppm, 1% DMSO, pH 7.4) at concentra-
tions ranging from 22 to 0.2 mM, were injected. The
Scheme 1. Synthesis of the cell wall models. Reagents and
conditions: (i) benzyl alcohol, H ; (ii) N-Boc-alanine, EDCI,
equilibrium dissociation constants K were estimated by
+
d
an analysis (BIAevaluation 3.0 program) of the sensor-
grams (Fig. 3). Under these conditions, nonspecific
binding to the control surface was not observed (data
are not shown).
HOBt, DMAP, CH Cl ; (iii) TFA, rt; (iv) N-Cbz-8-amino-
2
2
caprylic acid, EDCI, HOBt, CH Cl ; (v) Pd–C, H , methanol;
2
2
2
(
vi) NaNO , aq. CH COOH.
2 3
comycin dimer to a bacterial cell surface membrane
model.
−5
The K values obtained were 3.25×10 M for
D
-Ala-
D
-
-
5
d
−4
Ala and >1×10 M (above measurable limit) for
Ala-
with the trends observed in previous solution phase
studies (K of vancomycin with N-Ac - -Lys- -Ala-
-Ala- -Lac 10 –
M), which proved the validity of the SPR surface.
D
D
-Lac, respectively. These values are in accordance
Our studies employed a commercially available auto-
mated SPR system (BIAcore 2000, PE biosystems). The
tripeptides 4 and 5 were designed as immobilized lig-
ands on optical biosensors, which mimic the C-terminal
L
2
D
D-
−3
d
−6
Ala 10 M and with N-Ac2-
L
-Lys-
D
D
−4
6
1
0
segments of bacterial peptideglycan precursors,
nine- -alanine (vancomycin susceptible) and -alanine-
-lactate (vancomycin resistant), respectively (Scheme
D-ala-
D
D
The affinities of the monomer 2 and polymer 3 to the
optical sensors presenting peptide 4 (vancomycin sus-
ceptible type) and 5 (vancomycin resistant type) were
determined in a similar fashion (Figs. 4 and 5). Since
D
1
). The synthesis of 4 and 5 was achieved with tradi-
tional techniques, as depicted in Scheme 1.
7
polymer 3 is a mixture of 2- to ca. 15-mers, rigorous
These peptides were immobilized to an optical biosen-
sor (‘sensor chip’ CM5) with four independent flow
cells, where thin gold surfaces are coated with carboxy-
lated dextran.
analysis of the data was not trivial. Thus, apparent
dissociation constants Kd were estimated on a per
residue basis. These results are summarized in Table 1.
The sensorgram clearly indicates an enhanced affinity
of polymer 3 for each SPR surface presenting 4 and 5
relative to monomer 2. Most strikingly, the apparent Kd
value of 3 with resistant bacteria mimic 5 was enhanced
First, the carboxylate terminus of carboxylated dextran
were activated by a mixture of N-ethyl-N%-(diethyl-
aminopropyl)-carbodiimide (EDC) and N-hydroxysuc-
cinimide (NHS). Then, ligands 4 and 5 were injected to
−
5
to the same range (ꢀ10 M) as that of vancomycin 1
Figure 3. The binding of vancomycin to a SPR surface presenting 4 (A) and 5 (B): concentration of vancomycin. Blue 22.9 mM,
red 11.5 mM, green 2.3 mM, yellow 0.23 mM, light blue 0 mM.

H. Arimoto et al. / Tetrahedron Letters 42 (2001) 3347–3350
3349
Figure 4. The binding of vancomycin monomer 2 (A) and polymer 3 (B) to a SPR surface presenting vancomycin susceptible
Gram positive bacteria model 4: concentration of vancomycin, blue 22.9 mM, red 11.5 mM, green 2.3 mM, yellow 0.23 mM, light
blue 0 mM.
Figure 5. The binding of vancomycin monomer 2 (A) and polymer 3 (B) to a SPR surface presenting VRE model peptide 5:
concentration of vancomycin, blue 22.9 mM, red 11.5 mM, green 2.3 mM, yellow 0.23 mM, light blue 0 mM.
to sensitive bacteria mimic 4. This might be important
(SPR). The SPR analyses by use of commercially avail-
able carboxylated dextran optical sensors would be a
convenient and powerful tool in the search for more
potent vancomycin polymers. A screening of polymeric
vancomycin derivatives is currently under way using
SPR and will be reported in due course.
to an increase of the antibacterial activity of polymeric
2
vancomycin 3 against VRE (Table 2).
The results also suggest that the observed increases in
affinity for 4 and 5 do not quantitatively account for
the biological activities. For example, polymer 3 also
binds with increased affinity to peptide 4, but it was not
reflected in the potency against vancomycin susceptible
bacteria. Other additional factors could also contribute
to the antibacterial nature of these compounds.
Acknowledgements
We thank Professor Hirokazu Kawagishi (Shizuoka
University) for the use of the Surface Plasmon Reso-
nance spectrometer, Professor William Moser (IUPUI)
for critiquing the manuscript, and the Grant-in Aid for
We have described in this letter, the enhanced affinities
of the polymer 3 for resistant bacterial cell wall models,
which was measured by surface plasmon resonance

3
350
H. Arimoto et al. / Tetrahedron Letters 42 (2001) 3347–3350
Table 1. Ligand-binding affinities of vancomycin 1, monomer 2, and polymer 3
2
Table 2. In vitro antibacterial activities of vancomycin 1, monomer 2, and polymer 3
Scientific Research from the Ministry of Education,
Science, Sports and Culture, Japan (No. 12558076,
3. (a) Kiessling, L. L.; Pohl, N. L. Chem. Biol. 1996, 3, 71;
(b) Mammen, M.; Choi, S. K.; Whitesides, G. M. Angew.
Chem., Int. Ed. 1998, 37, 2754.
12780436) and Shionogi award in synthetic organic
chemistry, Japan and the Research grants from the
Sumitomo foundation, Novartis science foundation,
and the Inamori foundation.
4. For a short review of SPR principles and applications,
Weimer, T. Angew. Chem., Int. Ed. 2000, 39, 1219.
5. Rao, J.; Yan, L.; Xu, B.; Whitesides, G. M. J. Am. Chem.
Soc. 1999, 121, 2629.
6
. For some recent examples: (a) Xu, R.; Greiveldinger,
G.; Marenus, L. E.; Cooper, A.; Ellman, J. A. J.
Am. Chem. Soc. 1999, 121, 4898. (b) Sundram, U. N.;
Griffin, J. H.; Nicas, T. I. J. Am. Chem. Soc. 1996, 118,
References
1
. For recent reviews, see: Niccolai, D.; Tarsi, L.; Thomas,
R. J. Chem. Commun. 1997, 2333. See also: Williams, D.
H.; Bardsley, B. Angew. Chem., Int. Ed. 1999, 38, 1173.
. Arimoto, H.; Nishimura, K.; Kinumi, T.; Hayakawa, I.;
Uemura, D. Chem. Commun. 1999, 1361.
13107.
2
7. Polymer 3 prepared in our preliminary studies was used in
this study. The degree of polymerization was judged from
SDS-PAGE analysis, see Figure 2 in Ref. 2.
2
.
.