Synthesis and biological evaluation of vancomycin dimers with potent activity against vancomycin-resistant bacteria: Target-accelerated combinatorial synthesis

Article abstract of DOI:10.1002/1521-3765(20010903)7:17<3824::AID-CHEM3824>3.0.CO;2-1

Based on the notion that dimerization and/or variation of amino acid 1 of vancomycin could potentially enhance biological activity, a series of synthetic and chemical biology studies were undertaken in order to discover potent antibacterial agents. Herein we describe two ligation methods (disulfide formation and olefin metathesis) for dimerizing vancomycin derivatives and applications of target-accelerated combinatorial synthesis (e.g. combinatorial synthesis in the presence of vancomycin's target Ac₂-L-Lys-D-Ala-D-Ala) to generate libraries of vancomycin dimers. Screening of these compound libraries led to the identification of a number of highly potent antibiotics effective against vancomycin-suspectible, vancomycin-intermediate resistant and, most significantly, vancomycin-resistant bacteria.

Full text of DOI:10.1002/1521-3765(20010903)7:17<3824::AID-CHEM3824>3.0.CO;2-1

FULL PAPER
Synthesis and Biological Evaluation of Vancomycin Dimers with Potent
Activity against Vancomycin-Resistant Bacteria: Target-Accelerated
Combinatorial Synthesis
K. C. Nicolaou,*[a] Robert Hughes,[a] Suk Young Cho,[a] Nicolas Winssinger,[a]
Harald Labischinski,[b] and Rainer Endermann[b]
Abstract: Based on the notion that
dimerization and/or variation of amino
acid 1 of vancomycin could potentially
enhance biological activity, a series of
synthetic and chemical biology studies
were undertaken in order to discover
potent antibacterial agents. Herein we
describe two ligation methods (disulfide
formation and olefin metathesis) for
dimerizing vancomycin derivatives and
applications of target-accelerated com-
binatorial synthesis (e.g. combinatorial
synthesis in the presence of vancomy-
cinꢀs target Ac2-l-Lys-d-Ala-d-Ala) to
generate libraries of vancomycin dimers.
Screening of these compound libraries
led to the identification of a number of
highly potent antibiotics effective
against vancomycin-suspectible, vanco-
mycin-intermediate resistant and, most
significantly, vancomycin-resistant bac-
teria.
Keywords: antibiotics
´ biological
evaluation ´ combinatorial synthesis
´ synthesis design ´ vancomycin
cin (1) by parallel and target-accelerated combinatorial
synthesis in a study that led to the identification of a number
of highly potent antibiotics effective against vancomycin-
resistant bacteria.
Introduction
The emergence of microorganisms resistant to antibiotics
poses a serious threat to public health.[1] In particular, the rise
of bacterial resistance to vancomycin (1, Figure 1), the drug of
last resort for the treatment of many Gram-positive infections,
has spurred vigorous research activities into discovering new
classes of antibacterial agents, as well as toward modifying
existing types of antibiotics[2] in order to check the latest
bacterial moves. In the preceding paper[3] we described the
synthesis and biological evaluation of a series of monomeric
vancomycin derivatives, some of which exhibited good to
excellent activity against vancomycin-intermediate suscepti-
ble Staphylococcus aureus (VISA) and vancomycin-resistant
Enterococci (VRE). In this article we describe the synthesis
and biological evaluation of dimeric derivatives of vancomy-
OH
HO
HO
H2N
OH
O
O
O
O
Cl
O
O
O
OH
HO
Cl
O
H
N
O
H
H
NH
N
N
N
N
H
H
H
O
O
H
O
O
NH
O
HO
NH2
OH
OH
HO
1 : vancomycin
[a] Prof. Dr. K. C. Nicolaou, R. Hughes, Dr. S. Y. Cho, Dr. N. Winssinger
Department of Chemistry
Figure 1. Chemical structure of vancomycin (1).
and The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 North Torrey Pines Road
La Jolla, California 92037 (USA)
and
A number of glycopeptide antibiotics form non-covalent
dimers[4] and the tendency of these compounds to dimerize
has been correlated with their ability to eradicate bacteria.[5]
The reasons for this phenomenon are manifold. First, there is
a benefit from the multivalency.[6] Second, in the case of head-
to-tail, back-to-back dimer formation, the hydrogen bonds
that form the dimer interface are mediated by the same amide
units which are responsible for binding to the terminal d-Ala-
Department of Chemistry and Biochemistry
University of California, San Diego
9500 Gilman Drive, La Jolla, California 92093 (USA)
Fax : (‡1)858784-2469
E-mail: kcn@scripps.edu
[b] Dr. H. Labischinski, Dr. R. Endermann
Bayer AG, PH-R-Antiinfectives, 42096 Wuppertal (Germany)
3824
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3824 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

3824±3843
d-Ala, vancomycinꢀs binding site, through a different network
of hydrogen bonds[7] (see Figure 2). The net result of this
cooperative interaction is that the dimer has a higher affinity
for the d-Ala-d-Ala ligand than the monomer, and ligand-
bound monomer has a higher propensity to dimerize than the
free monomer. In the case of vancomycin, the dimerization
units; and c) the type of reaction to be used for the ligation.
On the basis of the expected ligand binding enhancement, due
to cooperative effects of the head-to-tail, back-to-back dimer,
we reasoned that the saccharide domain, specifically the
vancosamine nitrogen, would be the most appropriate site at
which to construct the bridge between the two vancomycin
units. It should be noted that dimerization from the C-termi-
nus to the N-terminus, as has been accomplished by Wil-
liams,[15] could also generate a back-to-back dimer capable of
similar cooperative effects. However, as we envisioned a one-
step dimerization process, this arrangement seemed overly
complicated. With the location of the tether decided upon, the
more challenging problem of selecting the tether itself that
would maximize the cooperative binding to the ligand needed
to be addressed. To this end, we envisioned a target-
accelerated combinatorial synthesis (TACS) strategy for the
construction of vancomycin dimers to simultaneously eval-
uate the effect of tether length and binding pocket modifica-
tions on the overall activity of vancomycin dimers (see
below). This approach required a ligation method that was
compatible with vancomycinꢀs polyfunctional structure, able
to operate at ambient temperature and in aqueous solution,
and reversible, at least as long as the TACS reaction was in
progress, so as to allow for dynamic equilibration toward the
thermodynamically most stable assemblies. It appeared that
both the classical formation of a disulfide bond and the more
contemporary olefin metathesis reaction could meet these
stringent requirements.
1
constant[8] for free vancomycin (1) is about 700m , while the
dimerization constant for the vancomycin/ligand complex[7] is
1
about 104 m . Given these considerations, we designed a
strategy for the generation and identification of highly potent
vancomycin dimers employing target-accelerated combinato-
rial synthesis (TACS). Similar dynamic combinatorial strat-
egies have been proposed and advanced in different contexts
by Lehn,[9] Sanders,[10] Benner,[11] and others.[12]
Attempting to harness the benefits of multivalency and
cooperativity for the generation of potent vancomycin-
derived antibiotics with activity against vancomycin-resistant
strains, several groups have synthesized dimeric derivatives of
vancomycin. Griffin and co-workers[13] synthesized covalent
dimers linked through the C-terminus of vancomycin (Fig-
ure 3). Significantly, in this first example of covalent dimeri-
zation of this antibiotic, compounds were produced which
displayed respectable antibacterial activity against vancomy-
cin-resistant strains. Additionally, the Whitesides group con-
structed head-to-head dimers.[14] Perhaps, however, of more
biological relevance are the head-to-tail dimers. Constructs of
this type were synthesized by the Williams group,[15] linking
from the C-terminus to the N-terminus, and by groups at
Abbott[16] and Eli Lilly,[17] both bridging their monomeric
units through the vancosamine nitrogen. Finally, a vancomy-
cin-derived trimer[18] has been reported as well as an
oligomeric system,[19] the latter exhibiting improved bacter-
iacidal activity against vancomycin-resistant strains.
Disulfide dimers: During our initial investigation of the two
chosen dimerization processes (disulfide formation and olefin
metathesis) several dimeric compounds were synthesized and
biologically evaluated. As depicted in Scheme 1, saponifica-
tion of the vancomycin-derived thioacetates (2, prepared as
described in the preceding article)[3] with NaOH (10.0 equiv)
in H2O at 238C proceeded smoothly to give, after sponta-
neous aerial oxidation, disulfides 3. Special care was necessary
in this procedure in order to balance the increased rate of
reaction under basic conditions with the decomposition which
appeared to take place at high pH. The antibacterial activities
against a variety of vancomycin-susceptible, vancomycin-
Results and Discussion
Having decided to explore the compelling dimer approach to
improving vancomycinꢀs biological activity, we faced the
following issues: a) choice of the site of dimerization;
b) choice of the appropriate bridge to join the two monomeric
Figure 2. Hydrogen-bond network of vancomycin head-to-tail, back-to-back dimer (left) and its wire frame representation (right, one vancomycin unit is
gray and the other is black).[7]
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3825 $ 17.50+.50/0
3825

K. C. Nicolaou et al.
FULL PAPER
OH
HO
OH
O
a.
OH
NH2
HO
H2N
HO
HO
OH
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
OH
OH
O
OH
HO
Cl
Cl
O
O
O
H
O
O
H
H
N
H
N
H
H
NH
NH
N
N
N
N
N
H
N
H
N
N
H
H
H
H
O
O
O
O
H
H
O
O
O
HN
O
O
NH
O
Linker
N
N
NH2
NH2
H
H
HO
OH
OH
OH
OH
HO
OH
O
OH
HO
b.
HO
H2N
HO
HO
H2N
OH
OH
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
O
OH
HO
HO
Cl
Cl
O
O
O
H
O
O
H
H
N
H
H
H
NH
N
N
N
N
N
N
N
H
N
H
N
N
H
H
H
H
O
O
O
H
O
H
O
O
O
NH
O
O
NH
O
N
HO
NH2
NH2
H
OH
OH
OH
OH
HO
HO
OH
HO
c.
OH
O
OH
HO
NH
HO
Linker
NH
HO
OH
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
O
OH
HO
Cl
HO
Cl
O
O
H
O
H
H
O
H
H
H
NH
N
N
NH
N
N
N
N
N
H
N
N
N
H
H
H
H
H
O
O
H
O
O
H
O
O
O
NH
O
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
HO
HO
Figure 3. Types of previously synthesized vancomycin dimers: a) Griffin[13] and Whitesides[14] (head-to-head); b) Williams[15] (head-to-tail), c) Abbott[16] and
Eli Lilly[17] groups (back-to-back).
intermediate resistant, and vancomycin-resistant bacteria for
these disulfide dimers were determined and are presented in
Tables1 and 2. For the non-substituted dimers 3a ± d (Table 1)
and 3n ± o (Table 2) the activity against vancomycin-resistant
strains is far superior to that exhibited by vancomycin. Indeed,
compounds 3a and 3b (Table 1) showed MIC values against
VRE strain L4001 of 1 mgmL and 2 mgmL , respectively.
Also, apparent from these data is a dependence on tether
length of antibacterial activity. This trend is summarized in
Figure 4 and reflects a requirement of a fifteen to twenty atom
bridge between the vancosamine nitrogens of the two joined
vancomycin units for optimal activity. Efforts to further
enhance the activity of such vancomycin dimers through
introduction of certain lipophilic groups as membrane
anchoring devices, as in compounds 3 f ± m (Table 1), failed
to produce the desired higher potency. Not only did such
groups fail to endow the dimers with enhanced biological
activity, but in fact, as the lipophilicity of the side chain
increased the activity of the compounds decreased. For
instance, 3 h and 3i were found to be less active than 3 f and
3g; and 3l and 3m exhibited decreased activity as compared
to 3j and 3k. In further studies, some of the most active
1
1
3826
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3826 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
OH
HO
R
HO
H
N
OH
O
O
O
O
O
AcS
n
Ligation via disulfide formation
Cl
O
O
O
OH
HO
O
Cl
O
NaOH, air (O2), H2O
(50 − 90 %)
H
N
O
H
H
N
NH
N
N
H
N
H
H
O
H
O
O
NH
O
HO
NH2
OH
OH
2
HO
OH
OH
OH
HO
R
R
OH
NH
HO
H
HO
OH
N
O
O
O
O
n
O
O
O
O
O
S
S
n
O
Cl
Cl
O
O
O
Cl
O
O
O
OH
OH
HO
HO
Cl
O
O
O
H
N
H
H
N
O
H
H
H
N
NH
N
NH
N
N
N
N
N
N
H
H
H
H
H
H
O
H
O
O
O
O
H
O
O
O
NH
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
3
HO
HO
Scheme 1. Dimerization of vancomycin analogues 2 to vancomycin dimers 3 through disulfide bond formation. NaOH (10.0 equiv), H2O, 23 8C, 48 h, 50 ±
90%. See Tables 1 and 2 for definitions of R and n. Ac ˆ acetate.
compounds were screened against additional strains of
vancomycin-resistant organisms (Table 3). We were pleased
to find that some of these compounds (3b ± d, n, o, q)
exhibited uniformly excellent activity against this broader
range of resistant bacterial strains.
ralº homodimer (entry 3b, 3c, 3n; Table 3) in activity.
However, it appears that the d configuration of amino acid
1 endows the dimer with higher biological activity than the
corresponding l-amino acid does. For example, compound 6-
(d-Ala) exhibited excellent antibacterial activity while 6-
(l-Ile) and 6-(l-Val) showed less antibiotic activity.[20] Inter-
estingly, compound 6-(H) (lacking an amino acid) proved to
be still reasonably active. Additionally, amino acid substitu-
tions bearing a heteroatom such as in 6-Ser and 6-Thr resulted
in improved activity.
The ease by which disulfide dimers could be formed
encouraged us to further consider such constructs. A hetero-
dimer with one vancomycin unit capable of binding the
natural d-Ala-d-Ala segment of the cell-wall and the other
vancomycin moiety poised to bind its mutant counterpart (d-
Ala-d-Lac) was contemplated as an attractive proposition for
the development of uniquely active and broad spectrum
agents effective against drug-resistant pathogens. In order to
explore this possibility, a number of heterodimers with the
optimum tether length was designed and targeted for syn-
thesis. Thus, as shown in Scheme 2, vancomycin-derived
thioacetate 2a was converted, in 80% yield, to the mixed
disulfide 4 by the action of NaOMe and pyS-Spy in MeOH.
Treatment of a 1:1 mixture of this mixed disulfide 4 and a
binding pocket modified thioacetate (5, R ˆ various amino
acid residues, see Table 4) synthesized as previously de-
scribed[3] with NaOMe in MeOH gave smoothly, and in 65 ±
90% yield, the desired heterodimers 6. The latter compounds
6 were screened against a panel of vancomycin-resistant
bacteria revealing a number of interesting data (Table 4).
Thus, several of the compounds synthesized rival the ªnatu-
Olefinic dimers: Encouraged by the success of the disulfide
dimers we sought to develop the second ligation method that
was based on the olefin metathesis[21] and by which olefinic
(mixture of cis and trans) dimers were expected to be formed.
As this work was in progress Arimoto and co-workers[18]
reported the application of the ring opening metathesis
polymerization (ROMP) reaction in methanolic solution to
the construction of vancomycin-based oligomers. However,
our circumstances dictated the requirement of conducting the
dimerization in aqueous media. Initial experiments using the
emulsion (H2O/CH2Cl2) conditions reported by Grubbs[22]
were encouraging as we were able to obtain dimeric
compounds. As applied to the formation of vancomycin
dimers, however, this protocol required heating to 508C and
the amount of CH2Cl2 was too high for our purposes. After
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3827 $ 17.50+.50/0
3827

K. C. Nicolaou et al.
FULL PAPER
Table 1. Antibacterial activity (MIC: mgmL 1) of vancomycin-derived disulfide dimers (3a ± m) against vancomycin-susceptible, vancomycin-intermediate
resistant and vancomycin-resistant bacteria.
HO
HO
OH
OH
OH
HO
H
H
OH
N
HO
N
R2
R2
O
O
O
O
O
O
O
R1
O
O
Cl
Cl
O
O
O
O
O
OH
OH
HO
Cl
HO
Cl
O
O
O
H
H
H
H
O
O
H
H
N
N
NH
N
N
NH
N
N
N
H
N
H
N
N
H
H
H
H
O
O
O
O
H
O
O
H
O
NH
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
3
HO
HO
Compound
R1
vancomycin
R2
Sa8250[a] Sp670[a] 27266[b] 4002[b]
27261[b] 48N[C]
25701[C] LO3[C]
133[C]
0.25
MU50[d] 4001[e]
0.5
0.25
2
1
0.5
2
1
1
8
1
8
0.5
8
4
8
>16
1
1
<0.03
<0.03
1
1
S
O
2
H
H
3a
O
S
S
S
2
S
3
O
3
3b
<0.03
<0.03
0.06
2
4
1
2
1
2
4
8
8
8
8
1
2
8
8
2
2
O
S
4
O
4
0.125
16
3c
H
H
H
O
S
5
O
5
3d
0.5
>16
1
8
8
8
16
16
16
8
16
8
O
S
S
S
9
O
9
3e
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
O
3f
S
2
O
2
O
O
2
2
2
4
4
8
4
4
8
8
8
8
8
8
4
8
8
8
8
8
4
4
O
S
S
3g
S
3
O
3
O
O
3h
S
2
O
2
16
8
16
16
16
>16
>16
>16
>16
16
16
>16
>16
>16
>16
O
S
S
3
3
S
3
O
3
>16
>16
>16
>16
>16
>16
>16
3i
O
O
S
2
O
2
3j
4
4
2
4
8
8
4
4
8
8
8
16
8
>16
>16
8
>16
>16
8
16
16
>16
>16
4
4
O
S
S
Ph
Ph
O
O
S
3
O
3
3k
O
PhpCl
PhpCl
>16
>16
>16
>16
>16
>16
>16
>16
3l
S
2
O
2
O
O
S
S
O
3
3m
O
>16
16
>16
>16
>16
>16
>16
>16
>16
>16
>16
O
S
3
[a] Vancomycin-susceptible strains of Streptococcus pneumoniae. [b] Vancomycin-susceptible strains of Enterococcus faecalis. [c] Vancomycin-susceptible
strains of Staphylococcus aureus. [d] Vancomycin-intermediate resistant Staphylococcus aureus. [e] Vancomycin-resistant (van A) Enterococcus faecium.
considerable experimentation, we found that by employing
thoroughly degassed H2O and a phase transfer catalyst
(C12H25NMe3Br) and a trace of CH2Cl2 (<5%), the reaction
proceeded smoothly at 238C. Furthermore, dimeric vanco-
mycin derivatives were cleanly obtained as the only product
under these new conditions. Indeed, the fact that this reaction
proceeds so well in H2O with such polyfunctional substrates as
vancomycin derivatives is a testament to the robust nature of
the Grubbꢀs catalyst [(PCy3)2Ru(CHPh)Cl2] system. Inciden-
tally, the recently reported water soluble ruthenium-based
catalyst[23] was also assayed but failed to yield significant
amounts of product. The successful olefin metathesis based
ligation reaction is summarized in Scheme 3. Thus, mono-
meric olefins 7 were subjected to the olefin metathesis
conditions {[(PCy3)2Ru(CHPh)Cl2] (0.2 equiv), C12H25NMe3Br
(2.2 equiv), H2O/CH2Cl2 (>95:5), 238C] to give dimers 8
in good yields. The dimers shown in Table 5 and Table 6
were synthesized by this method and tested for antibac-
terial activity against a variety of bacterial strains. As with the
disulfides, a strong relationship between the length of the
tether and the antibacterial action was noted (Figure 4).
The dependence of activity on tether length is more pro-
nounced for the olefins than for the disulfides, with the
optimal length of sixteen atoms between the nitrogens of
the vancosamine moieties. Furthermore, activity decreased
when additional lipophilic branching was introduced to the
dimers (compounds 8g and 8 h, as compared to 8c and 8d,
respectively, Table 5). However, this effect is less dramatic
than in the case of the disulfides. In addition, we have
synthesized eleven dimeric olefins with modified binding
pockets (amino acid 1 variations, see Scheme 4). The anti-
bacterial activity of these compounds against a range
of bacterial strains is summarized in Table 7. Several of
these compounds (the most potent being 10c, 10e, 10 h)
rival the activity of natural vancomycin-derived homodimers
(see Table 5 and Table 6). Interestingly, the inclusion of a
C-terminal valine (compound 10k, Table 7) reduced the
activity of the dimers by some two to four-fold over the
3828
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3828 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
Table 2. Antibacterial activity (MIC: mgmL 1) of vancomycin-derived disulfide dimers (3n ± s) against vancomycin-susceptible, vancomycin-intermediate
resistant and vancomycin-resistant bacteria.
OH
OH
HO
HO
OH
HO
H
H
N
N
HO
OH
R
O
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
OH
HO
Cl
HO
Cl
O
O
O
H
O
H
H
H
O
H
H
N
N
N
NH
N
NH
N
N
N
H
N
H
N
N
H
H
H
H
O
O
H
O
O
O
O
O
H
O
NH
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
3
HO
HO
Compound
R
Sa8250[a] Sp670[a]
27266[b]
4002[b]
0.5
27261[b] 48N[C]
25701[C] LO3[C]
133[c]
MU50[d] 4001[e]
1
vancomycin
0.5
0.25
2
2
2
1
1
4
1
8
0.5
0.25
4
8
>16
2
S
2
O
3n
<0.03
<0.03
1
16
1
O
O
O
O
S
S
S
S
2
S
O
3
3o
3q
3r
0.5
8
0.25
4
8
4
4
8
8
>16
>16
4
8
4
3
S
O
4
16
16
16
>16
>16
>16
>16
>16
4
S
O
5
8
4
16
16
16
>16
>16
>16
>16
>16
>16
16
>16
>16
>16
>16
5
S
O
9
3s
>16
>16
>16
>16
>16
>16
O
S
9
[a] Vancomycin-susceptible strains of Streptococcus pneumoniae. [b] Vancomycin-susceptible strains of Enterococcus faecalis. [c] Vancomycin-susceptible
strains of Staphylococcus aureus. [d] Vancomycin-intermediate resistant Staphylococcus aureus. [e] Vancomycin-resistant (van A) Enterococcus faecium.
unsubstituted variety (compound 8c, Table 5). In reflecting on
the above findings and considering the potential metabolic
instability of the disulfides, we opted to further pursue only
the olefinic dimers.
Target-accelerated combinatorial synthesis of vancomycin
dimers: In order to more rapidly access potent vancomycin
dimers than possible by the parallel synthesis described above
we contemplated a combinatorial approach to such systems.
At the onset of our studies we were aware of the work of Lehn
and colleagues[9] who have advanced the concepts of dynamic
combinatorial libraries. Similar concepts have also been
reported by Sanders,[10] Benner,[11] and others.[12, 24] This
strategy, whereby the building blocks of a potential combina-
torial library are allowed to pre-organize and react in the
presence of a target (or a host) (see Figure 5), should
generate, preferentially, products with the highest affinity
for the target out of all ªvirtuallyº possible library members.
The vancomycin d-Ala-d-Ala system appeared to us an ideal
Figure 4. Correlation of tether length and antibacterial activity against
vancomycin-susceptible and vancomycin-resistant strains of disulfide
dimers (a) and vancomycin olefin dimers (b). Tether length is the number
of atoms between neighboring vancosamine nitrogens. The vancomycin-
susceptible data is an average of nine strains [Streptococcus pneumoniea
(Sa8250 and Sp670), Enterococcus faecalis (27266, 4002, 27261), Staph-
ylococcus aureus (48N, 25701, LO3, 133)]. See Table 2 and Scheme 1 for
structures. The disulfide dimers with tether lengths of 12 and 16 atoms were
prepared from thioacetates bearing
a
2-(CH2CH2SAc) or
a
4-(CH2CH2SAc) substituted benzyl appendage to the vancosamine nitro-
gen, respectively.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3829 $ 17.50+.50/0
3829

K. C. Nicolaou et al.
FULL PAPER
OH
HO
OH
HO
HO
HN
HO
HN
OH
OH
O
O
O
O
3
O
O
O
O
O
O
3
AcS
N
S
S
Cl
Cl
O
O
O
O
O
a. NaOMe, MeOH
O
OH
OH
O
HO
Cl
HO
Cl
O
H
N
O
H
H
H
N
O
H
H
N
S
S
N
NH
N
NH
N
N
N
N
N
H
N
N
(80 %)
H
H
H
H
H
O
H
O
O
O
H
O
O
O
NH
O
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
HO
HO
2a
OH
HO
4
HO
H
N
OH
O
O
O
O
O
AcS
Cl
O
O
O
OH
b. NaOMe, MeOH
HO
Cl
(55 − 90 %)
O
H
N
H
H
R
N
N
N
N
H
H
H
O
H
O
O
O
NH
O
HO
N
H
NH2
OH
OH
O
5
HO
OH
HO
OH
HO
OH
HO
HN
H
N
HO
OH
O
O
O
O
O
O
O
O
S
S
O
O
Cl
Cl
O
O
O
O
O
OH
OH
HO
Cl
HO
Cl
O
O
O
H
N
H
H
O
H
H
H
R
N
NH
N
N
N
N
N
N
N
N
H
H
H
H
H
H
O
O
H
NH
O
O
O
H
O
O
O
O
NH
O
HO
N
H
HO
NH2
NH2
OH
OH
OH
OH
O
HO
HO
6
Scheme 2. Formation of the vancomycin-derived heterodimers 6. a) NaOMe (10.0 equiv), pyS-Spy (5.0 equiv), MeOH, 238C, 30 min, 80%; b) 4 (1.0 equiv),
5 (1.1 equiv), NaOMe (10.0 equiv), MeOH, 238C, 45 min, 55 ± 90%. See Table 4 for definitions of R. Compounds are designated and referred to as 6-(xxx)
where xxx is the three letter code for the amino acid represented by R. py ˆ pyridine.
thesis (TACS) would facilitate the generation of dimeric
vancomycin derivatives with optimized biological properties,
since both binding to d-Ala-d-Ala and the non-covalent
dimerization were implicated in the mechanism of action of
this antibiotic. According to this plan, a library of vancomycin
monomers bearing appropriately reactive functionality would
be allowed to pre-organize themselves onto d-Ala-d-Ala.
Figure 5. Schematic representation of target-accelerated combinatorial
synthesis. A target (red) is incubated with a library of building blocks
Supramolecular complexes that adopt the most stable bond-
ing arrangement should be selectively ligated to form covalent
dimers. As mentioned, vancomycin has a higher dimerization
constant in the presence of its target and, as such, we reasoned
that TACS would be able to select for both the proper tether
(blue). The assembly with the highest affinity to the target should be
formed preferentially upon ligation under dynamic conditions.
case to apply such a strategy. Specifically, we hypothesized
that a target (d-Ala-d-Ala)-accelerated combinatorial syn-
3830
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3830 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
Table 3. Antibacterial activity (MIC: mgmL 1) of selected vancomycin-derived disulfide dimers (3b ± d, n, o, and q) against vancomycin-susceptible,
vancomycin-intermediate resistant, and vancomycin-resistant bacteria. See Tables 1 and 2 for structures.
Compound
MU50[a]
133[a] 4002[b]
1528[c] 2689[c] 2741[c] 2781[C] 2805[C] 4001[d] 1669[e] 2671[e] 2823[e] 1803[f] 1924[f] 1944[f]
3.13
8
0.39
1
0.39 >100
50
1
>100
1
100
2
25
0.25
>100
2
100
0.5
50
0.5
100
1
50
0.5
25
0.25
50
0.5
vancomycin
1
3b
1
1
1
3c
3d
8
2
2
1
8
2
8
2
8
2
4
2
8
0.5
8
1
1
4
1
0.125
1
1
16
8
2
4
1
8
4
4
4
3n
16
1
1
1
1
0.5
2
0.125 0.25
0.5
0.13
0.06
0.25
3o
3q
8
4
8
2
8
2
4
2
8
4
8
2
8
>16
2
8
1
2
2
1
2
4
0.5
2
1
4
16
8
2
4
[a] Vancomycin-intermediate resistant Staphylococcus aureus. [b] Vancomycin-susceptible Enterococcus faecalis. [c] Vancomycin-resistant Enterococcus
faecalis. [d] Vancomycin-resistant Enterococcus faecium. [e] Vancomycin-resistant (van A) and Synercid-resistant (sat G) Enterococcus faecium.
[f] Vancomycin-resistant (van A) and Synercid-resistant (sat A) Enterococcus faecium.
length and for optimum amino acid 1 substitution which is
postulated to perturb vancomycinꢀs binding pocket (Figure 6).
In order to determine whether or not the dimerization of
vancomycin monomers would be effected with accelerated
rates in the presence of vancomycinꢀs target (Ac-d-Ala-d-Ala
or Ac2-l-Lys-d-Ala-d-Ala), we subjected olefinic monomers
7-(LeuNMe)C2 and 7-(LeuNMe)C4 separately to the olefin
metathesis conditions {olefin (550mm), C12H25NMe3Br
(2.75mm), [(PCy3)2Ru(CHPh)Cl2] (110mm), H2O/CH2Cl2
(>95:5), 238C} in the presence of 110mm Ac-d-Ala-d-Ala or
Ac2-l-Lys-d-Ala-d-Ala. The reaction progress was monitored
by HPLC analysis of aliquots taken at regular time intervals.
As shown in Figure 7, a pronounced rate enhancement for
dimerization was observed, particularly in the case of Ac2-l-
Lys-d-Ala-d-Ala as expected since this peptide binds to
vancomycin more tightly[25] and induces its dimerization to a
greater extent[9] than Ac-d-Ala-d-Ala.
Having demonstrated the validity of the target-accelerated
synthesis of vancomycin dimers from individual monomers,
we then proceeded to investigate a combinatorial version of
the dimerization process. To this end, three vancomycin
analogues 7-(LeuNMe)C2, 7-(LeuNMe)C3, and 7-(LeuN-
R2
OH
HO
HO
H
N
OH
O
O
n
O
O
Ligation via olefin metathesis
O
Cl
O
O
O
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
(30 − 80 %)
OH
R1
HO
Cl
H
N
O
H
H
N
N
N
N
H
H
H
O
O
H
O
O
NH
O
HO
NH2
OH
OH
OH
HO
OH
HO
R2
R2
HO
7
OH
HO
H
H
N
HO
N
OH
O
O
O
O
O
O
O
O
n
n
O
O
Cl
Cl
O
O
O
O
O
OH
R1
OH
HO
HO
Cl
Cl
O
O
H
N
O
H
N
H
H
H
N
H
R1
N
N
N
N
N
N
N
H
H
H
H
H
H
O
O
O
H
O
H
O
O
O
NH
O
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
8
HO
HO
Scheme 3. Dimerization of vancomycin analogues 7 to vancomycin dimers 8 through olefin metathesis. [(PCy3)2Ru(CHPh)Cl2] (0.2 equiv), C12H25NMe3Br
(2.2 equiv), H2O/CH2Cl2 (>95:5), 238C, 30 ± 80%. See Tables 5 and 6 for definitions of R1, R2 and n. Cy ˆ cyclohexyl.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3831 $ 17.50+.50/0
3831

K. C. Nicolaou et al.
FULL PAPER
Table 4. Antibacterial activity (MIC: mgmL 1) of vancomycin-derived disulfide heterodimers (6) against vancomycin-susceptible, vancomycin-intermediate
resistant, and vancomycin-resistant bacteria.
OH
OH
HO
HO
HO
OH
H
H
OH
HO
N
N
O
O
O
O
O
O
O
O
O
O
S
S
Cl
Cl
O
O
O
O
O
OH
OH
NHR
HO
Cl
HO
Cl
O
O
H
H
O
H
O
H
H
H
N
NH
N
N
N
N
N
N
H
N
N
H
H
H
H
O
O
O
O
H
O
O
H
O
NH
O
O
NH
O
HO
NH
NH2
NH2
OH
OH
O
OH
OH
6
OH
HO
HO
Compound
1
R
MU50[a] 133[a] 4002[b] 1528[c] 2689[c] 2741[c] 2781[C] 2805[C] 4001[d] 1669[e] 2671[e] 2823[e] 1803[f] 1924[f] 1944[f]
vancomycin
3.13
>16
0.39
0.39 >100
50
8
>100
100
16
8
25
8
>100
16
100
8
50
4
100
4
50
2
25
1
50
4
4
4
2
4
4
2
1
H
6-(H)
Gly
>16
>16
8
16
4
2
1
8
2
4
0.5
1
6-(Gly)
6-(D-Ala)
2
8
4
8
8
4
2
1
4
2
2
8
8
2
2
1
1
1
2
1
L-Ala
β-Ala
4
4
2
4
2
8
16
4
16
8
8
4
8
8
2
8
8
2
4
1
4
6-(β-Ala)
6-(Sar)[g]
6-(γ-Abu)
>16
16
4
2
8
2
1
1
Sar
4
0.5
1
1
>16
γ-Abu
>16
>16
>16
2
4
0.5
4
2
6-(ε-Ahx)[g]
6-(Ile)
4
2
2
8
ε-Ahx
>16
>16
>16
>16
>16
16
4
8
4
0.5
4
1
2
8
8
16
2
8
16
16
L-Ile
16
16
16
8
16
4
4
L-Val
L-Cha
L-Leu
L-Ser
>16
8
>16
4
>16
>16
>16
2
6-(Val)
>16
4
>16
8
>16
4
>16
2
>16
8
>16
4
>16
2
>16
8
>16
4
6-(Cha)[g]
6-(Leu)
6-(Ser)
6-(Thr)
6-(Met)
6-(Phe)
6-(Tyr)
1
2
>16
2
>16
4
>16
4
>16
>16
2
>16
2
>16
>16
8
>16
2
>16
8
>16
2
>16
2
>16
0.25
>16
2
1
0.5
L-Thr
L-Met
4
8
4
8
4
8
8
16
>16
>16
2
1
1
0.5
>16
2
2
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
L-Phe
L-Tyr
>16
4
>16
2
>16
>16
>16
4
>16
2
>16
8
>16
8
>16
4
>16
>16
4
>16
4
>16
>16
8
>16
1
1
1
1
6-(Thi)
>16
4
>16
8
>16
16
>16
2
>16
4
>16
2
>16
4
>16
16
>16
8
>16
16
>16
8
>16
16
>16
8
>16
8
>16
2
L-Thi
6-(Orn)[g]
L-Orn
L-Lys
L-Cit
6-(Lys)
>16
>16
>16
8
>16
16
>16
4
>16
2
>16
2
>16
4
>16
16
>16
1
>16
16
>16
16
>16
8
>16
8
>16
>16
4
6-(Cit)[g]
1
R[h]
R[l]
>16
>16
16
16
>16
16
6-(Asp(OtBu))
6-(Glu-(OtBu))
8
8
8
4
4
4
4
4
4
16
16
4
4
1
8
8
16
16
8
8
4
4
16
2
[a] Vancomycin-intermediate resistant Staphylococcus aureus. [b] Vancomycin-susceptible Enterococcus faecalis. [c] Vancomycin-resistant (van A) Enter-
ococcus faecalis. [d] Vancomycin-resistant (van A) Enterococcus faecium. [e] Vancomycin-resistant (van A) and Synercid-resistant (sat G) Enterococcus
faecium. [f] Vancomycin-resistant (van A) and Synercid-resistant (sat A) Enterococcus faecium. [g] Sar is sarcosine; e-Ahx is 6-aminocaproic acid; Cha is b-
cyclohexylalanine; Orn is ornithine; Cit is citrulline. [h] R ˆ Asp(OtBu). [i] R ˆ Glu(OtBu).
Me)C4 bearing tethers of differing length were mixed and
subjected to target-accelerated covalent dimerization by
olefin metathesis (Scheme 5). The experiment was monitored
by mass spectrometric analysis performed, periodically,
directly on the reaction mixture. In the absence of the target
(Figure 8),[26] the expected statistical mixture (1:2:3:2:1; two
permutations are degenerate) of the six possible dimers [8-
(LeuNMe)C2-(LeuNMe)C2, 8-(LeuNMe)C2-(LeuNMe)C3, 8-
(LeuNMe)C2-(LeuNMe)C4, 8-(LeuNMe)C3-(LeuNMe)C3, 8-
(LeuNMe)C3-(LeuNMe)C4, 8-(LeuNMe)C4-(LeuNMe)C4]
was observed, whereas in the presence of the target (Ac2-l-
Lys-d-Ala-d-Ala) a clear preference for the dimers with
shorter tethers [8-(LeuNMe)C2-(LeuNMe)C2, 8-(LeuN-
Me)C2-(LeuNMe)C3] was evident (Figure 8). From evaluating
the biological activity of the pure dimers (Table 5), it was
apparent that a strong correlation between target-induced
3832
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3832 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
Table 5. Antibacterial activity (MIC: mgmL 1) of vancomycin-derived olefinic dimers (8a ± w) against vancomycin-susceptible, vancomycin-intermediate
resistant and vancomycin-resistant bacteria. Unbranched compounds (8a ± f and 8o ± w) are referred to in a short-hand to convey tether length and position
one amino acid substitution. Thus, compound 8w is designated as 8-(H)C4-(H)C4. This designation implies that 8-(H)C4-(H)C4 was formed by joining two 7-
(H)C4 molecules (see Scheme 3). The olefin bridge is ꢀ1:1 mixture of cis:trans isomers.
HO
HO
OH
OH
HO
HN
OH
H
OH
HO
N
R2
R2
O
O
O
O
O
O
R1
O
O
Cl
Cl
O
O
O
O
O
OH
R3
HO
Cl
OH
R3
HO
Cl
O
H
O
H
H
H
O
H
H
N
N
N
N
N
N
N
H
N
N
H
N
H
H
H
H
O
O
O
H
O
O
O
H
O
NH
O
O
NH
O
HO
NH2
HO
NH2
OH
OH
OH
OH
8
HO
HO
Compound
R1
R2
R3
Sa8250[a] Sp670[a] 27266[b] 4002[b] 27261[b] 48N[C] 25701[c] LO3[c] 133[c] MU50[d] 4001[e]
1
vancomycin
0.5
1
0.25
1
2
0.5
>16
2
1
1
0.5
>16
0.25
>16
4
>16
>16
O
8a
D-NMeLeu
H
16
>16
>16
>16
>16
O
8b
2
2
8
H
0.06
<0.03
2
4
4
4
8
D-NMeLeu
D-NMeLeu
D-NMeLeu
O
O
2
O
8c
H
<0.03
<0.03
<0.03
<0.03
O
0.125
0.25
1
0.25
0.25
0.25
0.25
2
0.25
1
1
8
2
4
O
2
8d
H
4
4
1
1
O
2
3
O
3
8e
H
D-NMeLeu
4
8
>16
16
16
16
>16
16
16
>16
>16
O
O
7
8f
H
D-NMeLeu
D-NMeLeu
>16
>16
8
>16
4
>16
8
>16
8
>16
8
>16
8
>16
8
>16
16
O
>16
>16
8
7
O
O
O
8g
1
0.5
O
O
2
8h
D-NMeLeu
D-NMeLeu
D-NMeLeu
4
4
2
4
O
16
8
8
8
16
16
16
16
16
16
16
8
16
8
16
16
16
8
2
O
O
O
O
8i
O
3
3
8j
>16
4
>16
2
>16
4
>16
4
>16
4
>16
4
>16
16
>16
8
>16
>16
>16
O
2
2
2
2
O
D-NMeLeu
D-NMeLeu
D-NMeLeu
2
8k
Ph
Ph
0.5
2
4
8
O
O
O
O
2
8l
4
4
O
8
8
8
8
8
16
16
8
>16
>16
8
8
>16
>16
8
16
>16
>16
8m
O
PhpCl
4
>16
16
16
O
O
>16
O
2
8n
D-NMeLeu
PhpCl
>16
8
>16
>16
O
4
8
O
O
8o
L-Asn
L-Asn
2
4
4
8
H
16
16
16
8
16
16
>16
16
16
16
16
8
>16
>16
>16
>16
O
O
2
>16
16
8p
H
H
H
H
O
2
O
8q
β-Ala
β-Ala
0.25
1
0.25
0.5
O
1
4
0.25
1
2
2
16
8
>16
2
4
4
>16
8
8r
O
2
O
2
4
>16
16
2
O
8s
O
γ-Abu
L-Phe
L-Arg
0.125
0.5
0.25
1
1
4
0.5
1
8
2
8
8
4
0.5
1
8
>16
>16
O
O
8t
16
1
2
2
16
H
H
O
8u
0.5
1
1
0.5
1
16
2
1
16
>16
O
O
O
8v
H
H
H
H
8
8
>16
>16
16
>16
>16
16
>16
16
>16
16
>16
16
>16
8
>16
>16
>16
>16
O
8w
4
O
8
2
2
[a] Vancomycin-susceptible strains of Streptococcus pneumoniae. [b] Vancomycin-susceptible strains of Enterococcus faecalis. [c] Vancomycin-
susceptible strains of Staphylococcus aureus. [d] Vancomycin-intermediate resistant Staphylococcus aureus. [e] Vancomycin-resistant (van A)
Enterococcus faecium.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3833 $ 17.50+.50/0
3833

K. C. Nicolaou et al.
FULL PAPER
Table 6. Antibacterial activity (MIC: mgmL 1) of vancomycin-derived olefinic dimers (8x ± z) against vancomycin-susceptible, vancomycin-intermediate
resistant and vancomycin-resistant bacteria. The olefin bridge is ꢀ1:1 mixture of cis:trans isomers.
OH
OH
HO
HO
OH
NH
HO
H
HO
N
OH
O
O
R
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
OH
HO
Cl
HO
Cl
O
O
O
H
H
H
O
O
H
H
H
N
N
N
NH
N
NH
N
N
N
H
N
H
N
H
N
H
H
H
O
H
O
O
O
O
H
O
O
NH
O
O
NH
O
HO
HO
NH2
NH2
OH
OH
OH
OH
8
HO
HO
Compound
R
Sa8250[a]
Sp670[a]
27266[b]
4002[b]
0.5
1
27261[b]
48N[c]
25701[c]
LO3[c]
0.5
2
133[c]
0.25
MU50[d]
4001[e]
>16
4
1
vancomycin
0.5
0.25
2
2
2
2
1
2
1
2
4
8
O
8x
0.125
0.125
0.125
0.125
1
O
O
2
2
3
7
O
3
8y
8z
1
0.5
1
2
2
1
0.5
>16
4
2
O
7
>16
>16
>16
>16
>16
>16
>16
>16
>16
>16
O
[a] Vancomycin-susceptible strains of Streptococcus pneumoniae. [b] Vancomycin-susceptible strains of Enterococcus faecalis. [c] Vancomycin-susceptible
strains of Staphylococcus aureus. [d] Vancomycin-intermediate resistant Staphylococcus aureus. [e] Vancomycin-resistant (van A) Enterococcus faecium.
Table 7. Antibacterial activity (MIC: mgmL 1) of vancomycin derived olefinic dimers (10a ± k) against vancomycin-susceptible, vancomycin-intermediate
resistant and vancomycin-resistant bacteria. The olefin bridge is ca. 1:1 mixture of cis:trans isomers.
OH
OH
HO
HO
OH
NH
HO
HN
HO
OH
O
O
O
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
R2
OH
R2
HO
Cl
HO
Cl
O
H
H
H
O
O
H
H
H
N
N
N
N
N
N
N
H
N
N
N
H
H
H
H
H
O
O
O
O
H
O
H
O
O
NH
O
O
NH
O
R1
R1
NH2
NH2
OH
OH
OH
OH
10
HO
HO
Compound R1
R2
MU50[a] 133[a] 4002[b] 1528[c] 2689[c] 2741[c] 2781[c] 2805[c] 4001[d] 1669[e] 2671[e] 2823[e] 1803[f] 1924[f] 1944[f]
vancomycin
3.13 0.39
0.39
>16
>100
>16
50
>100
>16
100
>16
25
>100
>16
100
50
100
>16
>16
50
25
>16
>16
50
1
10a
>16
>16
>16
>16
>16
>16
>16 >16
>16 >16
>16
>16
>16
Gly
H
Gly
Gly
Gly
Gly
L-Met
L-Ser
>16
2
>16
1
>16
>16
2
>16
>16
1
>16
4
>16
>16
>16
10b
10c
0.125
>16
0.25
1
1
0.25
>16
0.25
4
>16
8
>16
0.5
0.5
0.5
>16
0.5
0.5
>16
0.5
0.25
>16
0.25
L-Cha[g]
L-Lys
>16
2
>16
1
>16
2
>16
4
>16
4
>16
1
10d
10e
1
1
2
>16
>16
2
Gly
>16
>16
2
>16
8
>16
4
>16
4
>16
>16
2
>16
2
>16
>16
4
>16
16
2
>16
>16
>16 >16
>16 >16
>16
>16
2
>16
>16
>16
>16
10f
10g
10h
10i
L-Phe
L-Val
L-Thi[g]
L-Ile
Gly
Gly
0.25
1
2
1
0.5
8
0.5
1
0.25
Gly
>16
>16
16
8
>16
2
>16
16
2
>16
>16
2
>16
>16
16
>16
>16
2
>16
>16
8
>16
>16
4
>16
>16
8
>16 >16
>16 >16
>16
>16
4
>16
>16
>16
>16
>16
>16
>16
4
L-Val L-Cha[g]
L-Val D-NMeLeu
10j
8
4
1
10k
[a] Vancomycin-intermediate resistant Staphylococcus aureus. [b] Vancomycin-susceptible Enterococcus faecalis. [c] Vancomycin-resistant (van A) Enter-
ococcus faecalis. [d] Vancomycin-resistant (van A) Enterococcus faecium. [e] Vancomycin-resistant (van A) and Synercid-resistant (sat G) Enterococcus
faecium. [f] Vancomycin-resistant (van A) and Synercid-resistant (sat A) Enterococcus faecium. [g] Cha is b-cyclohexylalanine; Thi is b-(2-thienyl)alanine.
3834
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3834 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
Figure 7. Proof of principle of target-accelerated combinatorial synthesis
of covalently linked vancomycin dimers through the olefin metathesis
reaction in the absence and presence of the target. a) 7-(LeuNMe)C2 ! 8-
(LeuNMe)C2-(LeuNMe)C2; b) 7-(LeuNMe)C4 ! 8-(LeuNMe)C4-(LeuN-
Me)C4. Vancomycin analogue (550 mm), C12H25NMe3Br (2.75mm), target
(110 mm), [(PCy3)2Ru(CHPh)Cl2] (110 mm), H2O/CH2Cl2 (>95:5), 238C
(see Table 5 for structures of compounds).
Figure 6. Schematic representation of dynamic target-accelerated syn-
thesis of vancomycin dimers. A library of vancomycin analogues (V, blue)
are incubated with their target, Ac2-l-Lys-d-Ala-d-Ala (T, red). Since the
dimerization constant of vancomycin (Kd ꢀ7 Â 102 m 1) is greater in the
presence of its target (Kd ꢀ104 m 1), it is expected that the target-bound
dimeric assemblies (III) should ligate preferentially over their non target-
bound counterparts.
ligand (vancomycin scaffold) on the rate of dimerization. A
most direct and potentially more productive way of modulat-
ing this affinity is to vary the first amino acid residue of
vancomycin. Thus, a mixture of two vancomycin analogues [7-
(LeuNMe)C2 and 7-(b-Ala)C2] was treated under the olefin
metathesis conditions described above in the absence and
presence of the Ac2-l-Lys-d-Ala-d-Ala (Scheme 6), and
product formation was again monitored by mass spectroscopy
(see Figrue 9).
rate acceleration and biological potency (Figure 4) does exist.
Most significantly, dimers from both the olefinic and disulfide
classes with optimal tether length (16 ± 18 atoms between the
two nitrogens), predicted by the TACS experiment described
above, exhibited potent antibacterial activities, particularly
against VRE.
With the optimum length of the tethering bridge deter-
mined, we then turned our attention to the influence of the
binding affinity of the target (Ac2-l-Lys-d-Ala-d-Ala) to the
It was interesting to note that, in the absence of the target,
the homodimer of the b-Ala-substituted vancomycin [8-(b-
OH
HO
HO
H
N
OH
O
O
O
O
Amino acid 1 modified
dimers
O
Cl
O
O
O
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
(40 − 75 %)
OH
R2
HO
Cl
H
N
O
H
H
N
N
N
N
H
H
H
O
O
H
O
O
NH
O
R1
N
NH2
H
OH
OH
HO
OH
OH
HO
HO
9
HO
OH
H
H
N
OH
N
HO
O
O
O
O
O
O
O
O
O
O
Cl
Cl
O
O
O
O
O
OH
R2
OH
R2
HO
Cl
HO
Cl
O
O
H
N
H
N
H
O
H
H
N
H
N
N
N
N
N
N
N
H
H
H
H
H
H
O
O
O
H
O
O
H
O
O
NH
O
O
NH
O
R1
R1
N
N
NH2
NH2
H
H
OH
OH
OH
OH
10
HO
HO
Scheme 4. Dimerization of vancomycin-derived olefins 9, with modifications at the amino acid 1-position, to vancomycin dimers 10. [(PCy3)2Ru(CHPh)Cl2]
(0.2 equiv), C12H25NMe3Br (2.2 equiv), H2O/CH2Cl2 (> 95:5), 238C, 40 ± 75%. See Table 7 for definitions of R1 and R2.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3835 $ 17.50+.50/0
3835

K. C. Nicolaou et al.
FULL PAPER
substituted vancomycin mono-
mer to dimerize in the absence
of the target. In the presence of
the target, however, the parent
vancomycin homodimer [8-
(LeuNMe)C2-(LeuNMe)C2]
OH
O
HO
HN
HO
OH
O
O
O
n
Olefin metathesis driven TACS
O
Cl
O
O
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
was obtained preferentially as
expected. The preferences ob-
served in this experiment are
again consistent with the poten-
cies exhibited by these com-
pounds. The results were also
consistent with the notion that
stronger affinity for the target
translates into higher dimeriza-
tion rates through monomer
selection. The observed en-
hancement results also estab-
lished that the background re-
action is relatively inconse-
quential. Table 8 summarizes
OH
HO
Cl
H
N
O
O
with and without
Ac2-L-Lys-D-Ala-D-Ala
H
O
H
H
N
N
N
N
N
H
H
H
O
O
O
H
O
NH
O
n= 2, 3, 4
HO
NH2
OH
OH
7
HO
3 Monomers
6 Dimers
n = 2 : 7-(LeuNMe)C2 ; n = 3 : 7-(LeuNMe)C3 ; n = 4 : 7-(LeuNMe)C4
Scheme 5. Target-accelerated combinatorial synthesis (TACS) of vancomycin dimers: the effect of tether length.
An equimolar mixture of 7-(LeuNMe)C2, 7-(LeuNMe)C3, and 7-(LeuNMe)C4 (200mm each) was treated with the
metathesis catalyst, [(PCy3)2Ru(CHPh)Cl2] (120mm), in the presence of C12H25NMe3Br (2.75 mm) in H2O/CH2Cl2
(>95:5) at 238C, in the presence and absence of the target (Ac2-l-Lys-d-Ala-d-Ala, 120mm).
the results of additional two-component target-accelerated
dimerization experiments. In all cases the results of the TACS
experiments faithfully predict the trend in biological poten-
cies.
Figure 8. Mass spectrometic analysis of the TACS experiment described in
Scheme 5. In the absence of the target (a) a statistical mixture [1:2:3:2:1] of
products is observed whereas in the presence of the target (b) the shorter
tether dimers 8-(LeuNMe)C2-(LeuNMe)C2 and 8-(LeuNMe)C2-(LeuN-
Me)C3 are preferentially formed.
Figure 9. Mass spectrometric analysis of the vancomycin dimer mixture
formed as described in Scheme 6. Note that in the absence of the target (a)
less of the dimer containing the (LeuNMe) is formed. However, in the
presence of the target (b), 8-(LeuNMe)C2-(LeuNMe)C2 is formed in
preference to 8-(b-Ala)C2-(b-Ala)C2.
Ala)C2-(b-Ala)C2] was formed preferentially over the parent
vancomycin homodimer [8-(LeuNMe)C2-(LeuNMe)C2]. This
outcome may be attributed to a higher tendency of the b-Ala-
3836
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3836 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
In an effort to examine the preferred orientation of
vancomycin monomers during dimerization, we designed
and synthesized olefin derivatives 11 and 12, which are
functionalized at the N-terminus and C-terminus, respec-
tively. In combination these derivatives were subjected
to TACS under the olefin metathesis conditions in the
absence and presence of Ac2-l-Lys-d-Ala-d-Ala as shown in
Scheme 7.
In both instances, mass spectroscopic analysis (Figure 10)
revealed the heterodimer 11-12 as the major product; but in
the presence of the target, the heterodimer strongly predomi-
nated in the ratio of 1:4.7:1.1 (11-11:11-12:12-12) as opposed
to the essentially statistical distribution of dimers ob-
served in the absence of the target [ratio of 1:2.6:1.2 (11-
11:11-12:12-12)]. These observations are in line with
the proposition, as supported by modeling, that only the
heterodimer (11-12) can adopt a head-to-tail, back-to-back
orientation, and thus enjoy the maximum benefit of the
hydrogen-bonding network associated with such an arrange-
ment.
OH
O
HO
HN
HO
OH
O
O
O
Olefin metathesis driven TACS
O
Cl
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
O
O
OH
HO
Cl
O
O
H
H
H
N
N
N
R
with and without
Ac2-L-Lys-D-Ala-D-Ala
N
N
H
H
H
O
O
H
O
O
NH
O
R = D-NMeLeu, β-Ala
HO
NH2
7
OH
OH
HO
2 Monomers
3 Dimers
R = LeuNMe : 7-(LeuNMe)C2 ; R = β-Ala : 7-(β-Ala)C2
Scheme 6. Target-accelerated combinatorial synthesis (TACS) of vanco-
mycin dimers: the effect of modulating binding affinity. An equimolar
mixture of 7-(LeuNMe)C2 and 7-(b-Ala)C2 (275mm each) was treated with
the metathesis catalyst, [(PCy3)2Ru(CHPh)Cl2] (110mm), in the presence of
C12H25NMe3Br (2.75mm) in H2O/CH2Cl2 (>95:5) at 238C, in the presence
and absence of the target (Ac2-l-Lys-d-Ala-d-Ala, 110mm).
Table 8. Binary dimerization experiments. Reactions were conducted under the same conditions as indicated in
the legend of Scheme 6.
Entry
Time [h]
Reactants
Additive
Product Ratio[a]
7-(H)C4 + 7-(D-LeuNMe)C4
7-(H)C4 + 7-(D-LeuNMe)C4
7-(H)C2 + 7-(D-LeuNMe)C2
7-(H)C2 + 7-(D-LeuNMe)C2
1
2
3
4
91
91
Ac2-L-Lys-D-Ala-D-Ala
1.0 : 6.1 : 3.3
1.0 : 1.7 : 2.1
None
120
120
Ac2-L-Lys-D-Ala-D-Ala
1.0 : 6.0 : 4.3
1.0 : 7.5 : 2.7
None
5
91
91
72
72
91
1.0 : 2.1 : 1.2
3.8 : 4.5 : 1.0
7-(β-Ala)C4 + 7-(D-LeuNMe)C4
7-(β-Ala)C4 + 7-(D-LeuNMe)C4
Ac2-L-Lys-D-Ala-D-Ala
6
7
None
1.0 : 2.5 : 1.4
7-(β-Ala)C2 + 7-(D-LeuNMe)C2
7-(β-Ala)C2 + 7-(D-LeuNMe)C2
Ac2-L-Lys-D-Ala-D-Ala
8
9
None
3.7 : 7.5 : 1.0
1.0 : 6.0 : 4.8
Ac2-L-Lys-D-Ala-D-Ala
7-(Asn)C4 + 7-(D-LeuNMe)C4
7-(Asn)C4 + 7-(D-LeuNMe)C4
10
1.0 : 1.8 : 1.1
91
None
[a] The product ratio is listed as 8-(AA)Cx-(AA)Cx:8-(AA)Cx-(d-NMeLeu)Cx:8-(d-NMeLeu)Cx-(d-NMe-
Leu)Cx as determined by mass spectroscopic analysis of the indicated mixture at the appropriate time. The
antibacterial activities of the homodimers formed in the above reactions are reported in Table 5 [7(H)C4 ! 8w,
7(H)C2 ! 8v, 7(d-LeuNMe)C4 ! 8d, 7(d-LeuNMe)C2 ! 8c, 7(b-Ala)C4 ! 8r, 7(b-Ala)C2 ! 8q, 7(Asn)C4 !
8p, and 7(b-Ala)C2 ! 8o].
OH
OH
HO
H2N
HO
HO
H2N
HO
OH
OH
O
O
O
O
O
O
Olefin metathesis driven TACS
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
O
O
Cl
Cl
O
O
O
O
O
+
with and without
Ac2-L-Lys-D-Ala-D-Ala
OH
OH
HO
Cl
HO
Cl
O
O
O
O
H
O
H
H
H
O
H
H
N
N
N
N
N
N
N
N
N
N
N
H
N
H
H
H
H
H
H
O
O
H
O
O
H
O
O
O
NH
O
O
NH
O
HO
N
NH2
NH2
O
H
11-11 : Homodimer
11-12 : Heterodimer
12-12 : Homodimer
OH
OH
OH
OH
HO
HO
11
12
Scheme 7. Target-accelerated combinatorial synthesis (TACS) of vancomycin dimers: Examining the effects of orientation. An equimolar mixture of 11 and
12 (275mm each) was treated with the metathesis catalyst, (PCy3)2Ru(CHPh)Cl2 (110mm), in the presence of C12H25NMe3Br (2.75 mm) in H2O/CH2Cl2
(>95:5) at 238C, in the presence and absence of the target (Ac2-l-Lys-d-Ala-d-Ala, 110mm). Compound 12-12 is the dimer formed from the union, through
olefin metathesis, of two molecules of 12. Likewise, 11-11, is the dimer formed from two units of 11 and 11-12 is the corresponding heterodimer.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3837 $ 17.50+.50/0
3837

K. C. Nicolaou et al.
FULL PAPER
revealing. Thus, in the absence of the target, mass spectro-
scopic analysis (Figure 11) showed a ratio of 0:2.7:1 for 13-
13:13-7-(LeuNMe)C4:8-LeuNMe)C4-LeuNMe)C4 as opposed
to a ratio of 0.1:1:1.1 for this same reaction mixture in the
Figure 10. Mass spectrometric analysis of the experiment depicted in
Scheme 7. Note that in the presence of the target (b) there is a greater
preference for the heterodimer 11-12 which can, apparently, adopt a head-
to-tail, back-to-back orientation. Furthermore this mode of dimerization
appears to be enhanced in the presence of the target (b).
Figure 11. Mass spectrometric analysis of the experiment depicted in
Scheme 8. Note that in the presence of the target (b) homodimer 7-
(LeuNMe)C4-(LeuNMe)C4 is favored, apparently, due to its enhanced
ability to adopt head-to-tail, back-to-back orientation.
An additional experiment performed with vancomycin
monomers 7-(LeuNMe)C4 and 13 in the absence and presence
of the target Ac2-l-Lys-d-Ala-d-Ala (Scheme 8) was also
Scheme 8. Target-accelerated combinatorial synthesis (TACS) of vancomycin dimers: Examining the effects of orientation. An equimolar mixture of 13 and
7-(LeuNMe)C4 (275mm each) was treated with the metathesis catalyst, [(PCy3)2Ru(CHPh)Cl2], (110mm), in the presence of C12H25NMe3Br (2.75 mm) in H2O/
CH2Cl2 (>95:5) at 238C, with or without the target (Ac2-l-Lys-d-Ala-d-Ala, 110mm). Compound 13-13 is the dimer formed from the union through olefin
metathesis of two molecules of 13. Likewise, 8-(LeuNMe)C4-(LeuNMe)C4, is the dimer formed from two units of 7-(LeuNMe)C4 and 13-7-(LeuNMe)C4 is
the corresponding heterodimer.
3838
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3838 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
presence of Ac2-l-Lys-d-Ala-d-Ala. The increased abundance
of 8-(LeuNMe)C4-(LeuNMe)C4 observed, relative to the
other dimers, in the presence of vancomycinꢀs target is not
surprising in view of the fact that this is the only dimer from
this set that can readily adopt the head-to-tail, back-to-back
orientation. Thus, these TACS experiments suggest the
preferred supramolecular structure, in solution, of the dimeric
vancomycin-derived complexes.[26]
the thirty-six members expected, only thirty could be ob-
served as distinct peaks by mass spectrometry as a conse-
quence of the degeneracy of the dimers. The vertical bars in
Figure 12 reflect the observed relative abundance of each of
the thirty distinct (with regard to their mass) vancomycin
dimers after adjustment to account for the expected statistical
occurrence[28] (average values for three experiments). Fig-
ure 13 graphically exhibits the antibacterial activities of ten
Having demonstrated that the target-accelerated dimeriza-
tion of vancomycin analogues selects for both the adequacy of
the tether and the affinity for the target, we proceeded to
perform an eight-component [7-(LeuNMe)C2, 7-(LeuN-
Me)C4, 7-(Asn)C2, 7-(Asn)C4, 7-(b-Ala)C2, 7-(b-Ala)C4, 7-
(H)C2, and 7-(H)C4] target-accelerated (Ac2-l-Lys-d-Ala-d-
Ala) combinatorial synthesis experiment employing the olefin
metathesis reaction as a means of ligation (Scheme 9). From
OH
HO
HN
HO
OH
O
O
O
O
Olefin metathesis driven TACS
n
Figure 13. Correlation of relative abundance of vancomycin dimers (a ± j,
Figure 12 and Table 5), with antibacterial activity (MIC, average of the
vancomycin susceptible strains listed in Table 5).
O
Cl
[(PCy3)2Ru(CHPh)Cl2], C12H25NMe3Br
H2O/CH2Cl2 (>95:5)
O
O
OH
HO
Cl
O
with and without
H
N
O
H
H
Ac2-L-Lys-D-Ala-D-Ala
N
N
individually prepared compounds. Gratifyingly, the target-
accelerated dimerization strategy predicted quite reliably the
overall trend of the observed biological potencies of the
library members, with only relatively minor deviations. The
identification and correct ranking of compounds [a: 8-
(LeuNMe)C2-(LeuNMe)C2, b: 8-(LeuNMe)C2-(b-Ala)C2, c:
8-(LeuNMe)C2-(LeuNMe)C4] as highly potent antibiotics
effective against both vancomycin-susceptible and vancomy-
cin-resistant strains is highly significant. Since six out of the
seven most abundant com-
R
N
N
H
H
H
O
O
H
O
D-NMeLeu, β-Ala
O
NH
O
R =
H, L-Asn
HO
NH2
OH
OH
n = 2, 4
7
HO
8 Monomers
36 Dimers
Scheme 9. Target-accelerated combinatorial synthesis (TACS) of 36
vancomycin dimers: simultaneously selecting for binding affinity and
tether length.
pounds contain LeuNMe, the
results also underscore the im-
portance of the amino acid
residue at position one for
strong binding and presumed
(observed in some cases) bio-
logical activity.
Conclusion
In this and the preceding paper
we have synthesized a series of
vancomycin derivatives starting
from vancomycin itself and
subjected a number of them to
target-accelerated combinato-
rial synthesis, a strategy that
facilitated the rapid discovery
of potent dimeric vancomycin-
derived antibiotics. Active
against vancomycin-susceptible
and vancomycin-resistant bac-
teria, some of these compounds
Figure 12. Target-accelerated combinatorial synthesis of a 36-member vancomycin dimer library. The vertical
bars represent the relative abundance of vancomycin dimers as compared to that of a): 8-(LeuNMe)C2-
(LeuNMe)C2 after the appropriate statistical adjustment. The data above represent the average of three
experiments. Compounds are labeled as follows: 1: 8-(H)C2-(H)C2, 2: 8-(H)C2-(H)C4, 3: 8-(H)C4-(H)C4, 4: 8-
(H)C2-(b-Ala)C2, 5: 8-(b-Ala)C2-(H)C4, 6: 8-(H)C2-(Asn)C2, 7: 8-(H)C2-(LeuNMe)C2, 8: 8-(H)C4-(b-Ala)C4, 9: 8-
(Asn)C2-(H)C4, 10: 8-(b-Ala)C2-(b-Ala)C2, 11: 8-(LeuNMe)C2-(H)C4, 12: 8-(b-Ala)C2-(b-Ala)C4, 13: 8-(Asn)C4-
(H)C4, 14: 8-(b-Ala)C2-(Asn)C2, 15: 8-(LeuNMe)C4-(H)C4, 16: 8-(b-Ala)C4-(b-Ala)C4, 17: 8-(LeuNMe)C2-(b-
Ala)C2, 18: 8-(b-Ala)C2-(Asn)C4, 19: 8-(b-Ala)C2-(LeuNMe)C4, 20: 8-(Asn)C2-(Asn)C2, 21: 8-(Asn)C2-(b-
Ala)C4, 22: 8-(LeuNMe)C2-(Asn)C2, 23: 8-(b-Ala)C4-(LeuNMe)C4, 24: 8-(LeuNMe)C2-(LeuNMe)C2, 25: 8-
(Asn)C2-(Asn)C4, 26: 8-(Asn)C2-(LeuNMe)C4, 27: 8-(LeuNMe)C2-(LeuNMe)C4, 28: 8-(Asn)C4-(Asn)C4, 29: 8-
(Asn)C4-(LeuNMe)C4, 30: 8-(LeuNMe)C4-(LeuNMe)C4 (see Table 5 for structures of compounds).
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3839 $ 17.50+.50/0
3839

K. C. Nicolaou et al.
FULL PAPER
Table 9. Antibacterial activity (MIC: mgmL 1) of selected vancomycin-derived disulfide and olefinic dimers (3b, 3c, 3n, 10c, 10e, and 10 h) against
vancomycin-susceptible, vancomycin-intermediate resistant, and vancomycin-resistant bacteria. See Tables 1, 2, and 7 for structures.
Compound
MU50[a] 133[a] 4002[b] 1528[c] 2689[c] 2741[c] 2781[c] 2805[c] 4001[d] 1669[e] 2671[e] 2823[e] 1803[f] 1924[f] 1944[f]
vancomycin 3.13
0.39
1
0.39 >100
50
1
>100 100
25
0.25
100
0.5
50
0.5
100
50
0.5
25
0.25
50
0.5
>100
2
1
3b
8
1
1
1
2
1
3c
3n
8
2
2
2
1
1
1
1
1
2
2
1
2
2
2
0.5
1
1
1
0.125
0.06
1
16
1
0.5
0.125
0.25 0.5
>16
0.5
0.5
0.13
0.25
10c
10e
2
2
0.125
0.25
1
1
1
2
2
4
>16
1
4
4
1
1
0.25
4
8
0.5 >16
0.5
0.5
0.25
0.25
0.25
0.5
0.5
2
2
2
10h
2
0.25
1
2
2
1
4
2
8
1
0.25
[a] Vancomycin-intermediate resistant Staphylococcus aureus. [b] Vancomycin-susceptible Enterococcus faecalis. [c] Vancomycin-resistant Enterococcus
faecalis. [d] Vancomycin-resistant Enterococcus faecium. [e] Vancomycin-resistant (van A) and Synercid-resistant (sat G) Enterococcus faecium.
[f] Vancomycin-resistant (van A) and Synercid-resistant (sat A) Enterococcus faecium.
3 f: tR ˆ 4.0 min (gradient over 7 min); LCMS (ES): calcd for
rival (see Table 9) or exceed, in potency, the most active
antibacterial agents known today. The concept of target-
accelerated combinatorial synthesis (dynamic combinatorial
synthesis) has been validated in a sophisticated system and in
a practical way leading to the discovery of potential drug
candidates. Further application of the principles involved and
the ligation reactions used in this study (disulfide bond
formation and olefin metathesis) in other situations are
envisioned. These may include RNA and DNA binding
constructs as well as protein and saccharide-type receptor
ligand systems.
‡
‡
C160H188Cl4N18O52S2 [M‡2H] : 1702.0, found 1702.8; [M‡3H] : 1134.5,
found 1134.4.
3g: tR ˆ 4.2 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C
162H192Cl4N18O52S2 [M‡2H] : 1715.5, found 1715.5; [M‡3H] : 1144.2,
found 1144.2.
3 h: tR ˆ 4.4 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C
166H200Cl4N18O51S2 [M‡2H] : 1743.2, found 1744.5; [M‡3H] : 1162.5,
found 1162.1.
3i: tR ˆ 4.5 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C168H204Cl4N18O51S2 [M‡2H] : 1757.7, found 1758.2; [M‡3H] : 1172.1,
found 1172.8.
3j: tR ˆ 7.6 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
C166H184Cl4N18O52S2 [M‡2H] : 1735.8, found 1735.6; [M‡3H] : 1157.2,
found 1157.4.
Experimental Section
3k: tR ˆ 4.1 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C168H188Cl4N18O52S2 [M‡2H] : 1749.5, found 1750.7; [M‡3H] : 1166.2,
General: See paper one in this sequence.[3]
found 1166.5.
General procedure for the conversion of thioacetates 2 to disulfides 3
(Scheme 1): A solution of thioacetate 2 (10 mg, ꢀ6.2 mmol) in water
(1.0 mL) was treated with NaOH (2.5 mg, 10.0 equiv, 62.5 mmol) and stirred
for 24 ± 48 h at ambient temperature. Purification by reverse-phase HPLC
(VYDAC C18, 25 mm  250 mm, flow rate 6.5 mLmin 1, 0 ! 100%
CH3CN (0.05% TFA) in H2O (0.05% TFA) over 30 min) provided the
desired disulfide 3 in yields ranging from 50 ± 90% (analytical HPLC given
below for LiChrospher C18, 6 mm  250 mm, flow rate 1.0 mLmin 1, 0 !
100% CH3CN (0.05% TFA) in H2O (0.05% TFA) over the time
indicated).
3l: tR ˆ 4.3 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C166H182Cl6N18O52S2 [M‡2H] : 1770.2, found 1770.7; [M‡3H] : 1180.9,
found 1180.4.
3m: tR ˆ 4.3 min (gradient over 7 min); LCMS (ES): calcd for
‡
‡
C
168H186Cl6N18O52S2 [M‡2H] : 1785.2, found 1785.7; [M‡3H] : 1190.2,
found 1190.6.
3n: tR ˆ 6.9 min (gradient over 10 min); LCMS (ES): calcd for
‡
‡
C152H172Cl4N18O50S2 [M‡2H] : 1628.5, found 1628.7; [M‡3H] : 1086.0,
found 1086.1.
Representative 1H NMR spectral data for compounds 3
3o: tR ˆ 6.9 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
3a: 1H NMR (500 MHz, CD3OD, 330 K): d ˆ 7.76 ± 7.71 (m, 8H), 7.40 ± 7.38
(m, 8H), 7.15 ± 7.09 (m, 10H), 6.54 ± 6.51 (m, 4H), 5.57 ± 5.37 (m, 10H),
4.85 ± 4.80 (m, 6H), 4.25 ± 3.87 (m, 30H), 3.13 ± 3.11 (m, 2H), 2.96 ± 2.90 (m,
10H), 2.88 ± 2.84 (m, 5H), 2.38 ± 2.22 (m, 2H), 2.11 ± 2.09 (m, 5H), 1.81 ±
1.74 (m, 2H), 1.62 ± 1.50 (m, 6H), 1.39 ± 1.34 (m, 6H), 1.05 ± 1.03 (m, 6H).
C154H176Cl4N18O50S2 [M‡2H] : 1643.5, found 1643.5; [M‡3H] : 1096.2,
found 1096.2.
3p: tR ˆ 7.5 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
C
156H180Cl4N18O50S2 [M‡2H] : 1657.7, found 1657.2; [M‡3H] : 1105.4,
found 1105.0.
3a: tR ˆ 7.0 min (gradient over 10 min); LCMS (ES): calcd for
3q: tR ˆ 8.0 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
‡
‡
C152H172Cl4N18O50S2 [M‡2H] : 1628.5, found 1628.8; [M‡3H] : 1086.0,
C158H184Cl4N18O50S2 [M‡2H] : 1671.5, found 1671.7; [M‡3H] : 1114.3,
found 1086.1.
found 1114.4.
3b: tR ˆ 6.8 min (gradient over 8 min); LCMS (ES): calcd for
3r: tR ˆ 8.5 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
‡
‡
C154H176Cl4N18O50S2 [M‡2H] : 1643.5, found 1643.2; [M‡3H] : 1096.2,
C166H200Cl4N18O50S2 [M‡2H] : 1727.8, found 1727.9; [M‡3H] : 1151.2,
found 1095.8.
found 1151.9.
3c: tR ˆ 7.2 min (gradient over 10 min); LCMS (ES): calcd for
3s: tR ˆ 8.7 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
‡
‡
C
156H180Cl4N18O50S2 [M‡2H] : 1657.7, found 1657.2; [M‡3H] : 1105.4,
C166H200Cl4N18O50S2 [M‡2H] : 1727.8, found 1727.9; [M‡3H] : 1151.2,
found 1105.2.
found 1151.1.
3d: tR ˆ 9.0 min (gradient over 10 min); LCMS (ES): calcd for
Preparation of thiopyridyl vancomycin derivative 4:
A mixture of
‡
‡
C158H184Cl4N18O50S2 [M‡2H] : 1671.5, found 1672.4; [M‡3H] : 1114.3,
thioacetate 2a (120 mg, 72 mmol) and dipyridyldisulfide (160 mg,
10.0 equiv, 720 mmol) was dissolved in MeOH (5.0 mL). To this mixture
was added a solution of NaOH (28 mg, 10.0 equiv, 720 mmol in 3.5 mL
MeOH). After 30 min, the reaction mixture was purified by reverse-phase
HPLC (VYDAC C18, 25 mm  250 mm, flow rate 6.5 mLmin 1, 0 ! 100%
found 1114.4.
3e: tR ˆ 8.6 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
C166H200Cl4N18O50S2 [M‡2H] : 1727.8, found 1727.7; [M‡3H] : 1151.2,
found 1151.7.
3840
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3840 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
CH3CN (0.05% TFA) in H2O (0.05% TFA) over 30 min) to give 4 (124 mg,
80%). Analytical HPLC: tR ˆ 8.2 min (LiChrospher C18, 6 mm  250 mm,
flow rate 1.0 mLmin 1, 0 ! 100% CH3CN (0.05% TFA) in H2O (0.05%
TFA) over 10 min); 1H NMR (500 MHz, CD3OD, 330 K): d ˆ 7.88 ± 7.71 (m,
5H), 7.54 ± 7.45 (m, 4H), 7.26 ± 7.20 (m, 2H), 7.14 ± 7.05 (s, 2H), 6.78 (brs,
2H), 6.63 (s, 1H), 6.50 (m, 2H), 6.08 (brs, 2H), 5.67 (d, J ˆ 7.6 Hz, 1H), 5.58
(d, J ˆ 4.4 Hz, 1H), 5.46 (d, J ˆ 13 Hz, 2H), 5.35 (d, J ˆ 3.9 Hz, 1H), 4.90
(m, 1H), 4.82 (s, 1H), 4.69 (s, 1H), 4.36 (d, J ˆ 8.8 Hz, 1H), 4.20 (s, 1H),
4.14 (t, J ˆ 6.3 Hz, 2H), 4.04 ± 3.88 (m, 4H), 3.70 (d, J ˆ 7.9 Hz, 2H), 3.49
(m, 1H), 3.03 ± 3.00 (m, 3H), 2.41 (s, 3H), 2.17 ± 2.13(m, 4H), 2.01 (d, J ˆ
13.5 Hz, 1H), 1.83 ± 1.74 (m, 6H), 1.36 (m, 2H), 1.20 (d, J ˆ 6.3 Hz, 3H),
6-Glu(OtBu): tR ˆ 8.1 min; LCMS (ES): calcd for C156H177Cl4N19O53S2
‡
‡
[M‡2H] : 1687.5, found 1686.8; [M‡3H] : 1124.2, found 1122.8.
General procedure for dimerization of 7 to dimers 8 through olefin
metathesis (Scheme 3): C12H25NMe3Br (4.2 mg, 14.0 equiv, 86.0 mmol) was
added under argon to a solution of olefin 7 (10 mg, ꢀ6.25 mmol) in
degassed water (1.0 mL). To this vigorously stirred solution (ambient
temperature) was added, dropwise,
a solution of Grubbꢀs catalyst
[(Cy3P)2Ru(CHPh)Cl2] (1.0 mg, 0.2 equiv, 1.25 mmol) in CH2Cl2 (200 mL).
After complete addition, argon was purged through the system until only a
trace of CH2Cl2 remained (ꢀ20 min). After stirring vigorously for 48 ± 96 h
at ambient temperature, the desired olefinic dimer 8 was isolated, in pure
form, after purification by reverse-phase HPLC (VYDAC C18, 25 mm Â
250 mm, flow rate 6.5 mLmin 1, 0 ! 100% CH3CN (0.05% TFA) in H2O
(0.05% TFA) over 30 min). Analytical HPLC given below for LiChrospher
C18, 6 mm  250 mm, flow rate 1.0 mLmin 1, 0 ! 100% CH3CN (0.05%
TFA) in H2O (0.05% TFA) over 10 min.
‡
1.03 ± 1.00 (m, 6H); LCMS (ES): calcd for C81H91Cl2N10O25S2 [M‡H] :
1739.5, found 1739.3.
General procedure for the synthesis of disulfide heterodimers
6
(Scheme 2): A 1:1 mixture (ꢀ2 mg, 1.2 mmol) of thiopyridyl vancomycin
derivative 4 and thioacetate 5 was dissolved in MeOH (0.8 mL) and NaOH
(1.6 mg, 10.0 equiv, 12.0 mmol) was added. The reaction mixture was stirred
for 45 min at ambient temperature and the product was purified by reverse-
phase HPLC (VYDAC C18, 10 mm  250 mm, flow rate 3.5 mLmin 1, 0 !
100% CH3CN (0.05% TFA) in H2O (0.05% TFA) over 20 min) to give
Representative 1H NMR spectroscopic data for compounds 8
8c: 1H NMR (500 MHz, CD3OD, 330 K): d ˆ 7.80 ± 7.70 (m, 10H), 7.45 ±
7.35 (m, 7H), 7.13 ± 7.06 (m, 10H), 6.6 ± 6.51 (m, 6H), 5.55 ± 5.35 (m, 12H),
4.83 ± 4.81 (m, 8H), 4.26 ± 3.82 (m, 25H), 3.23 ± 3.21 (m, 4H), 2.96 ± 2.89 (m,
11H), 2.88 ± 2.83 (m, 6H), 2.13 ± 2.09 (m, 5H), 1.80 ± 1.74 (m, 2H), 1.65 ±
1.55 (m, 6H), 1.37 ± 1.32 (m, 6H), 1.07 ± 1.05 (m, 6H).
pure heterodimer 6 in yields ranging from 55 ± 80% (analytical HPLC data
1
given below for LiChrospher C18, 6 mm  250 mm, flow rate 1.0 mLmin
,
0 ! 100% CH3CN (0.05% TFA) in H2O (0.05% TFA) over 10 min).
‡
8a: tR ˆ 7.6 min; LCMS (ES): calcd for C150H166Cl4N18O50 [M‡2H] : 1581.2,
‡
6-H: tR ˆ 7.7 min; LCMS (ES): calcd for C147H162Cl4N18O50S2 [M‡2H] :
‡
found 1581.3; [M‡3H] : 1055.2, found 1055.2.
‡
1594.8, found 1594.1; [M‡3H] : 1063.5, found 1063.6.
‡
8b: tR ˆ 7.7 min; LCMS (ES): calcd for C152H170Cl4N18O50 [M‡2H] : 1596.3,
‡
6-Gly: tR ˆ 7.8 min; LCMS (ES): calcd for C149H165Cl4N19O51S2 [M‡2H] :
‡
found 1595.8; [M‡3H] : 1064.2, found 1064.2.
‡
1622.8, found 1621.4; [M‡3H] : 1082.5, found 1082.8.
‡
8c: tR ˆ 8.9 min; LCMS (ES): calcd for C154H174Cl4N18O5o [M‡2H] : 1610.0,
‡
6-Ala: tR ˆ 7.9 min; LCMS (ES): calcd for C150H167Cl4N19O51S2 [M‡2H] :
‡
found 1610.3; [M‡3H] : 1073.6, found 1073.8.
‡
1629.5, found 1629.0; [M‡3H] : 1086.7, found 1086.4.
‡
8d: tR ˆ 8.4 min; LCMS (ES): calcd for C156H178Cl4N18O50 [M‡2H] : 1623.2,
‡
6-b-Ala: tR ˆ 7.8 min; LCMS (ES): calcd for C150H167Cl4N19O51S2 [M‡2H] :
‡
found 1624.3; [M‡3H] : 1083.5, found 1083.7.
‡
1629.5, found 1629.5; [M‡3H] : 1086.7, found 1086.8.
‡
8e: tR ˆ 11.5 min; LCMS (ES): calcd for C164H1194Cl4N18O50 [M‡2H] :
‡
6-Sar: tR ˆ 7.9 min; LCMS (ES): calcd for C150H167Cl4N19O51S2 [M‡2H] :
‡
1679.5, found 1679.8; [M‡3H] : 1119.6, found 1119.5.
‡
1629.5, found 1629.8; [M‡3H] : 1087.5, found 1087.1.
‡
8 f: tR ˆ 11.5 min; LCMS (ES): calcd for C164H194Cl4N18O50 [M‡2H] :
‡
6-g-Abu: tR ˆ 7.7 min; LCMS (ES): calcd for C151H169Cl4N19O51S2 [M‡2H] :
‡
1679.5, found 1679.3; [M‡3H] : 1119.6, found 1120.1.
‡
1636.5, found 1636.2; [M‡3H] : 1091.3, found 1091.8.
‡
8g: tR ˆ 9.2 min; LCMS (ES): calcd for C160H194Cl4N18O52 [M‡2H] : 1668.5,
‡
6-e-Ahx: tR ˆ 7.8 min; LCMS (ES): calcd for C153H173Cl4N19O51S2 [M‡2H] :
‡
found 1668.6; [M‡3H] : 1112.6, found 1112.4.
‡
1650.2, found 1650.1; [M‡3H] : 1100.5, found 1100.0.
‡
8 h: tR ˆ 9.5 min; LCMS (ES): calcd for C162H190Cl4N18O52 [M‡2H] :
‡
6-Ile: tR ˆ 8.0 min; LCMS (ES): calcd for C153H173Cl4N19O51S2 [M‡2H] :
‡
1682.5, found 1682.4; [M‡3H] : 1122.2, found 1121.4.
‡
1650.3, found 1649.3; [M‡3H] : 1101.5, found 1101.0.
‡
8i: tR ˆ 10.3 min; LCMS (ES): calcd for C166H198Cl4N18O52 [M‡2H] :
‡
6-Val: tR ˆ 7.8 min; LCMS (ES): calcd for C152H171Cl4N19O51S2 [M‡2H] :
‡
1710.5, found 1710.2; [M‡3H] : 1122.2, found 1122.7.
‡
1643.5, found 1643.1; [M‡3H] : 1096.3, found 1096.4.
‡
8j: tR ˆ 11.0 min; LCMS (ES): calcd for C168H202Cl4N18O52 [M‡2H] :
‡
6-Cha: tR ˆ 8.2 min; LCMS (ES): calcd for C156H177Cl4N19O51S2 [M‡2H] :
‡
1724.5, found 1724.8; [M‡3H] : 1150.2, found 1150.1.
‡
1670.0, found 1669.8; [M‡3H] : 1113.5, found 1113.2.
8k: tR ˆ 7.8 min (gradient over 8 min); LCMS (ES): calcd for
‡
6-Leu: tR ˆ 7.9 min; LCMS (ES): calcd for C153H173Cl4N19O51S2 [M‡2H] :
‡
‡
C
166H182Cl4N18O52 [M‡2H] : 1702.5, found 1701.8; [M‡3H] : 1135.3, found
‡
1650.3, found 1651.5; [M‡3H] : 1101.5, found 1100.4.
1135.2.
‡
6-Ser: tR ˆ 7.8min; LCMS (ES): calcd for C150H167Cl4N19O52S2 [M‡2H] :
‡
8l: tR ˆ 9.5 min; LCMS (ES): calcd for C168H186Cl4N18O52 [M‡2H] : 1716.5,
‡
1637.5, found 1637.5; [M‡3H] : 1091.8, found 1092.2.
‡
found 1716.2; [M‡3H] : 1144.6, found 1144.4.
‡
6-Thr: tR ˆ 7.9min; LCMS (ES): calcd for C151H169Cl4N19O52S2 [M‡2H] :
‡
8m: tR ˆ 9.8 min; LCMS (ES): calcd for C166H180Cl6N18O52 [M‡2H] :
‡
1645.2, found 1644.8; [M‡3H] : 1097.5, found 1097.5.
‡
1737.1, found 1737.2; [M‡3H] : 1158.3, found 1158.6.
‡
6-Met: tR ˆ 8.00 min; LCMS (ES): calcd for C152H171Cl4N19O51S3 [M‡2H] :
8n: tR ˆ 10.2 min (gradient over 8 min); LCMS (ES): calcd for
‡
1659.8, found 1659.3; [M‡3H] : 1107.5, found 1106.7.
‡
‡
C168H184Cl6N18O52 [M‡2H] : 1751.0, found 1750.6; [M‡3H] : 1167.7, found
‡
6-Phe: tR ˆ 8.1 min; LCMS (ES): calcd for C156H171Cl4N19O51S2 [M‡2H] :
1167.7.
‡
1668.6, found 1667.2; [M‡3H] : 1112.5, found 1112.8.
‡
8o: tR ˆ 7.8 min; LCMS (ES): calcd for C146H156Cl4N20O52 [M‡2H] : 1582.4,
‡
‡
6-Tyr: tR ˆ 8.1 min; LCMS (ES): calcd for C156H171Cl4N19O52S2 [M‡2H] :
found 1582.4; [M‡3H] : 1056.4, found 1056.3.
‡
1676.3, found 1675.4; [M‡3H] : 1117.5, found 1117.2.
‡
8p: tR ˆ 8.2 min; LCMS (ES): calcd for C150H164Cl4N20O52 [M‡2H] : 1611.4,
‡
‡
6-Thi: tR ˆ 8.0 min; LCMS (ES): calcd for C154H169Cl4N19O51S2 [M‡2H] :
found 1611.4; [M‡3H] : 1074.6, found 1074.6.
‡
1671.5, found 1671.4; [M‡3H] : 1114.2, found 1114.5.
‡
8q: tR ˆ 7.8 min; LCMS (ES): calcd for C144H154Cl4N18O50 [M‡2H] : 1540.2,
‡
‡
6-Orn: tR ˆ 7.9 min; LCMS (ES): calcd for C152H172Cl4N20O51S2 [M‡2H] :
found 1540.3; [M‡3H] : 1027.8, found 1027.9.
‡
1651.2, found 1651.6; [M‡3H] : 1100.2, found 1100.4.
‡
8r: tR ˆ 8.3 min; LCMS (ES): calcd for C148H162Cl4N18O50 [M‡2H] : 1568.3,
‡
‡
6-Lys: tR ˆ 7.7 min; LCMS (ES): calcd for C153H174Cl4N20O51S2 [M‡2H] :
found 1569.0; [M‡3H] : 1045.9, found 1046.8.
‡
1658.2, found 1658.3; [M‡3H] : 1105.2, found 1105.6.
‡
8s: tR ˆ 6.8 min; LCMS (ES): calcd for C146H158Cl4N18O50 [M‡2H] : 1554.2,
‡
‡
6-Cit: tR ˆ 7.8 min; LCMS (ES): calcd for C153H173Cl4N21O52S2 [M‡2H] :
found 1554.2; [M‡3H] : 1036.3, found 1036.6.
‡
1672.2, found 1672.2; [M‡3H] : 1115.3, found 1115.0.
8t: tR ˆ 7.5 min (gradient over 8 min); LCMS (ES): calcd for
‡
‡
6-Asp(OtBu): tR ˆ 8.8 min; LCMS (ES): calcd for C155H167Cl4N19O51S2
C156H162Cl4N18O50 [M‡2H] : 1616.4, found 1616.2; [M‡3H] : 1078.8, found
‡
‡
[M‡2H] : 1679.2, found 1679.6; [M‡3H] : 1120.2, found 1120.5.
1077.8.
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3841 $ 17.50+.50/0
3841

K. C. Nicolaou et al.
FULL PAPER
8u: tR ˆ 6.7 min (gradient over 8 min); LCMS (ES): calcd for
(LeuNMe)C4 (655 mg, 0.4 mmol) was dissolved in degassed H2O (1.8 mL).
To this solution was added C12H25NMe3Br (1.7 mg, 5.56 mmol). This
reaction mixture was split into two equal parts (900 mL each). To one part
was added H2O (100 mL) and to the other was added a solution of Ac2-l-
Lys-d-Ala-d-Ala (45 mg in 100 mL of H2O). To these individual, vigorously
stirred solutions was added, dropwise, a solution of Grubbsꢀ catalyst
[(PCy3)2Ru(CHPh)Cl2] (98 mg) in CH2Cl2 (200 mL) at ambient temper-
ature. After complete addition, argon was purged through the system until
only a trace of CH2Cl2 remained (ꢀ20 min). The final concentrations of the
components in the reaction mixture were as follows: 7-(LeuNMe)C2
‡
‡
C150H168Cl4N24O50 [M‡2H] : 1625.4, found 1625.4; [M‡3H] : 1083.9, found
1083.9.
‡
8v: tR ˆ 7.5 min; LCMS (ES): calcd for C138H144Cl4N16O48 [M‡2H] : 1469.2,
‡
found 1469.2, [M‡3H] : 979.2, found 979.3.
‡
8w: tR ˆ 8.31 min; LCMS (ES): calcd for C142H154Cl4N16O48 [M‡2H] :
‡
1497.3, found 1498.4; [M‡3H] : 998.5, found 999.3.
‡
8x: tR ˆ 8.64 min; LCMS (ES): calcd for C154H174Cl4N18O5o [M‡2H] :
‡
1609.7, found 1610.3; [M‡3H] : 1073.6, found 1073.8.
‡
(200mm),
7-(LeuNMe)C3
(200mm),
7-(LeuNMe)C4
(200mm),
8y: tR ˆ 9.7 min; LCMS (ES): calcd for C156H178Cl4N18O50 [M‡2H] : 1623.2,
‡
C12H25NMe3Br (2.75mm), ligand (120mm), and [(PCy3)2Ru(CHPh)Cl2]
(120mm). After stirring for 24 h, the reaction mixtures were examined by
MS (Hewlett Packard 1100 Series, electrospray). The z‡3 region of the mass
spectrum is depicted in Figure 8.
found 1623.9; [M‡3H] : 1083.5, found 1082.9.
‡
8z: tR ˆ 11.4 min; LCMS (ES): calcd for C164H194Cl4N18O50 [M‡2H] :
‡
1679.5, found 1679.9; [M‡3H] : 1119.6, found 1120.4.
Dimerization of vancomycin derivatives 9 to dimers 10 through olefin
metathesis (Scheme 4): The dimerization of compound 9[3] to dimeric
vancomycin derivative 10 (Scheme 4) through olefin metathesis was carried
out in the same manner as described above for the conversion of
compounds 7 to 8 (analytical HPLC given below for LiChrospher C18,
6 mm  250 mm, flow rate 1.0 mLmin 1, 0 ! 100% CH3CN (0.05% TFA)
in H2O (0.05% TFA) over 10 min).
Target-accelerated combinatorial synthesis (binding selective dimerization,
Scheme 6): An equimolar mixture of two vancomycin analogues 7-
(LeuNMe)C2 (866 mg, 0.55 mmol) and 7-(b-Ala)C2 (855 mg, 0.55 mmol)
was dissolved in degassed H2O (1.8 mL). To this solution was added
C12H25NMe3Br (1.7 mg, 5.56 mmol). This reaction mixture was split into two
equal parts (900 mL each). To one part was added H2O (100 mL) and to the
other was added a solution of Ac2-l-Lys-d-Ala-d-Ala (41 mg in 100 mL of
H2O). To these individual, vigorously stirred solutions was added, drop-
wise, a solution of Grubbsꢀ catalyst [(PCy3)2Ru(CHPh)Cl2] (90 mg) in
CH2Cl2 (200 mL) at ambient temperature. After complete addition, argon
was purged through the system until only a trace of CH2Cl2 remained
(ꢀ20 min). The final concentrations of the components in the reaction
mixture were 7-(LeuNMe)C2 (275mm), 7-(b-Ala)C2 (275mm),
C12H25NMe3Br (2.75mm), ligand (110mm), and [(PCy3)2Ru(CHPh)Cl2]
(110mm). After stirring for 24 h, the reaction mixtures were examined by
MS (Hewlett Packard 1100 Series, electrospray). The z‡3 region of the mass
spectrum is depicted in Figure 9. The experiments summarized in Table 8
were conducted under identical conditions.
‡
10a: tR ˆ 7.1 min; LCMS (ES): calcd for C142H150Cl4N18O50 [M‡2H] :
‡
1526.3, found 1526.4; [M‡3H] : 1017.8, found 1017.0.
‡
10b: tR ˆ 7.3 min; LCMS (ES): calcd for C160H184Cl4N20O54S2 [M‡2H] :
‡
1729.6, found 1729.6; [M‡3H] : 1153.4, found 1153.9.
‡
10c: tR ˆ 7.1 min; LCMS (ES): calcd for C156H176Cl4N20O52 [M‡2H] :
‡
1685.4, found 1685.4; [M‡3H] : 1123.9, found 1123.8.
‡
10d: tR ˆ 8.0 min; LCMS (ES): calcd for C160H180Cl4N20O52 [M‡2H] :
‡
1679.5, found 1680.2; [M‡3H] : 1120.0, found 1119.3.
‡
10e: tR ˆ 6.9 min; LCMS (ES): calcd for C154H174Cl4N22O52 [M‡3H] :
1103.3, found 1103.2.
‡
10 f: tR ˆ 7.3 min; LCMS (ES): calcd for C150H168Cl4N20O52 [M‡2H] :
Target-accelerated combinatorial synthesis (orientation selective dimeri-
zation, Scheme 7): An equimolar mixture of the two vancomycin analogues
11 (901 mg, 0.55 mmol) and 12 (885 mg, 0.55 mmol) was dissolved in degassed
H2O (1.8 mL). To this solution was added C12H25NMe3Br (1.7 mg,
5.56 mmol). This reaction mixture was split into two equal parts (900 mL
each). To one part was added H2O (100 mL) and to the other was added a
solution of Ac2-l-Lys-d-Ala-d-Ala (41 mg in 100 mL H2O). To these
individual, vigorously stirred solutions was added, dropwise, a solution of
Grubbsꢀ catalyst [(PCy3)2Ru(CHPh)Cl2] (90 mg) in CH2Cl2 (200 mL) at
ambient temperature. After complete addition, argon was purged through
the system until only a trace of CH2Cl2 remained (ꢀ20 min). The final
concentrations of the components in the reaction mixture were as follows:
11 (275mm), 12 (275mm), C12H25NMe3Br (2.75mm), ligand (110mm), and
[(PCy3)2Ru(CHPh)Cl2] (110mm). After stirring for 24 h, the reaction
mixtures were examined by MS (Hewlett Packard 1100 Series, electro-
spray). The z‡3 region of the mass spectrum is depicted in Figure 10.
‡
1673.4, found 1673.6; [M‡3H] : 1115.9, found 1116.1.
‡
10g: tR ˆ 7.1 min; LCMS (ES): calcd for C152H168Cl4N20O52 [M‡3H] :
1083.9, found 1083.8.
‡
10 h: tR ˆ 7.3 min; LCMS (ES): calcd for C164H180Cl4N20O54S2 [M‡2H] :
‡
1751.6, found 1751.7; [M‡3H] : 1167.0, found 1167.2.
‡
10i: tR ˆ 6.8 min; LCMS (ES): calcd for C154H172Cl4N20O52 [M‡2H] :
‡
1639.4, found 1638.3; [M‡3H] : 1093.3, found 1093.3.
‡
10j: tR ˆ 8.0 min; LCMS (ES): calcd for C166H192Cl4N20O52 [M‡2H] :
‡
1721.6, found 1720.8; [M‡3H] : 1148.0, found 1148.2.
‡
10k: tR ˆ 7.8 min; LCMS (ES): calcd for C160H184Cl4N20O52 [M‡2H] :
‡
1681.5, found 1681.7; [M‡3H] : 1121.3, found 1121.5.
Kinetics of dimerization through olefin metathesis in the presence and
absence of ligand (Figure 7): Olefinic vancomycin analogue 7-(LeuNMe)C2
(6.0 mg, 3.72 mmol) was dissolved in degassed H2O (6.0 mL). To this
solution was added C12H25NMe3Br (5.5 mg, 18.2 mmol). This solution was
split into three equal parts (2.0 mL each) and to each solution was added
either H2O (200 mL), a solution of Ac-d-Ala-d-Ala (49 mg in 200 mL of
H2O), or a solution of Ac2-l-Lys-d-Ala-d-Ala (90 mg in 200 mL of H2O). To
these individual, vigorously stirred solutions was added, dropwise, a
solution of Grubbsꢀ catalyst [(PCy3)2Ru(CHPh)Cl2] (181 mg) in CH2Cl2
(200 mL) at ambient temperature. After complete addition, argon was
Target-accelerated combinatorial synthesis (orientation selective dimeri-
zation, Scheme 8): An equimolar mixture of the two vancomycin analogues
13 (885 mg, 0.55 mmol) and 7-(LeuNMe)C4 (901 mg, 0.55 mmol) was
dissolved in degassed H2O (1.8 mL). To this solution was added
C12H25NMe3Br (1.7 mg, 5.56 mmol). This reaction mixture was split into
two equal parts (900 mL each). To one part was added H2O (100 mL) and to
the other was added a solution of Ac2-l-Lys-d-Ala-d-Ala (41 mg in 100 mL
of H2O). To these individual, vigorously stirred solutions was added,
dropwise, a solution of Grubbsꢀ catalyst [(PCy3)2Ru(CHPh)Cl2] (90 mg) in
CH2Cl2 (200 mL) at ambient temperature. After complete addition, argon
was purged through the system until only a trace of CH2Cl2 remained
(ꢀ20 min). The final concentrations of the components in the reaction
mixture were as follows: 13 (275mm), 7-(LeuNMe)C4 (275mm),
C12H25NMe3Br (2.75mm), ligand (110mm), and [(PCy3)2Ru(CHPh)Cl2]
(110mm). After stirring for 24 h, the reaction mixtures were examined by
MS (Hewlett Packard 1100 Series, electrospray). The z‡3 region of the mass
spectrum is depicted in Figure 11.
purged through the system until only
a trace of CH2Cl2 remained
(ꢀ20 min). The final concentrations of the components in the reaction
mixture were as follows: 7-(LeuNMe)C2 (550mm), C12H25NMe3Br
(2.75mm), ligand (110mm), and [(PCy3)2Ru(CHPh)Cl2] (110mm). The
reactions were monitored by HPLC (LiChrospher C18, 6 mm  250 mm,
flow rate 1.0 mLmin 1, 0 ! 100% CH3CN (0.05% TFA) in H2O (0.05%
TFA) over 10 min, with detection at 254 nm) at the following time
intervals: 12, 36, 72, and 144 h. The percentage of dimer formed was plotted
against time as shown in Figure 7. The kinetic experiment with vancomycin
analogue 7-(LeuNMe)C4 (Figure 7b) was performed under identical
conditions and the analysis performed at the following time intervals: 12,
24, 36, and 60 h.
Target-accelerated combinatorial synthesis (eight component competition,
Scheme 9): An equimolar mixture of eight vancomycin analogues [7-
(LeuNMe)C2 (708 mg, 0.44 mmol), 7-(LeuNMe)C4 (721 mg, 0.44 mmol), 7-(b-
Ala)C2 (684 mg, 0.44 mmol), 7-(b-Ala)C4 (696 mg, 0.44 mmol), 7-(Asn)C2
(703 mg, 0.44 mmol), 7-(Asn)C4 (715 mg, 0.44 mmol), 7-(H)C2 (652 mg,
Target-accelerated combinatorial synthesis (length selective dimerization,
Scheme 5): An equimolar mixture of three vancomycin analogues 7-
(LeuNMe)C2 (644 mg, 0.4 mmol), 7-(LeuNMe)C3 (650 mg, 0.4 mmol), and 7-
3842
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3842 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 17

Vancomycin Dimers
3824±3843
0.44 mmol), and 7-(H)C4 (664 mg, 0.44 mmol)] was dissolved in degassed
H2O (3.6 mL). To this solution was added C12H25NMe3Br (3.3 mg,
10.9 mmol). This reaction mixture was split into two equal parts (1.8 mL
each). To one part was added H2O (200 mL) and to the other was added a
solution of Ac2-l-Lys-d-Ala-d-Ala (261 mg in 200 mL H2O). To these
individual, vigorously stirred solutions was added, dropwise, a solution of
Grubbsꢀ catalyst [(PCy3)2Ru(CHPh)Cl2] (289 mg) in CH2Cl2 (200 mL) at
ambient temperature. After complete addition, argon was purged through
the system until only a trace of CH2Cl2 remained (ꢀ20 min). The final
concentrations of the components in the reaction mixture were as follows:
7-(LeuNMe)C2 (110mm), 7-(LeuNMe)C4 (110mm), 7-(b-Ala)C2 (110mm), 7-
(b-Ala)C4 (110mm), 7-(Asn)C2 (110mm), 7-(Asn)C4 (110mm), 7-(H)C2
(110mm), and 7-(H)C4 (110mm), C12H25NMe3Br (2.75mm), ligand
(176mm), and [(PCy3)2Ru(CHPh)Cl2] (176mm). After stirring for 24 h, the
reaction mixtures were examined by LCMS [LiChrospher C18, 6 mm Â
250 mm, flow rate 1.0 mLmin 1, 0 ! 100% CH3CN (0.05% TFA) in H2O
(0.05% TFA) over 10 min, with detection at 254 nm] specifically observing
Lehn, Chem. Eur. J. 1999, 5, 2455 ± 2463; d) V. Berl, I. Huc, J.-M. Lehn,
A. DeCian, J. Fischer, Eur. J. Org. Chem. 1999, 3089 ± 3094; e) O.
Ramstorm, J.-M. Lehn, ChemBioChem 2000, 1, 41 ± 48; f) B. Hasen-
knopf, J.-M. Lehn, B. O. Kneisel, G. Baum, D. Fenske, Angew. Chem.
1996, 108, 1987 ± 1990; Angew. Chem. Int. Ed. Engl. 1996, 35, 1838 ±
1840; g) B. Hasenknopf, J.-M. Lehn, N. Boumediene, A. Doupot-
Gervais, A. Van Dorsselaer, B. Kneisel, D. Fenske, J. Am. Chem. Soc.
1997, 119, 10956 ± 10962.
[10] a) S. Otto, R. L. E. Furlan, J. K. M. Sanders, J. Am. Chem. Soc. 2000,
122, 12063 ± 12064; b) G. R. L. Cousins, S.-A. Poulsen, J. K. M.
Sanders, Curr. Opin. Chem. Biol. 2000, 4, 270 ± 279; c) D. Hamilton,
N. Feeder, S. Teat, J. K. M. Sanders, New J. Chem. 1998, 22, 1019 ± 1021
d) P. A. Brady, J. K. M. Sanders, J. Chem. Soc. Perkin Trans. 1 1997,
3237 ± 3253; e) G. R. L. Cousins, S.-A. Poulsen, J. K. M. Sanders,
Chem. Commun. 1999, 1575 ± 1576; f) G. R. L. Cousins, R. L. E.
Furlan, Y.-F. Ng, J. E. Redman, J. K. M. Sanders, Angew. Chem.
2001, 113, 437 ± 442; Angew. Chem. Int. Ed. 2001, 40, 423 ± 428.
[11] a) J. P. Nawrocki, M. Wigger, S. Audetat, S. A. Benner, Synlett 1998,
688 ± 691; b) J. P. Nawrocki, M. Wigger, C. H. Watson, T. W. Hayes,
M. W. Senko, S. A. Benner, J. R. Eyler, Rapid Commun. Mass
Spectrom. 1996, 10, 1860 ± 1864; c) M. Wigger, J. P. Nawrocki, C. H.
Watson, J. R. Eyler, S. A. Benner, Rapid Commun. Mass Spectrom.
1997, 11, 1749 ± 1752.
[12] a) B. Klekota, M. H. Hammond, B. L. Miller, Tetrahedron Lett. 1997,
38, 8639 ± 8642; b) M. Albrecht, O. Blau, R. Fröhlich, Chem. Eur. J.
1999, 5, 48 ± 56; c) S. B. Lee, S. Hwang, D. S. Chung, J.-I. Hong,
Tetrahedron Lett. 1998, 39, 873 ± 876; d) M. C. Calama, R. Hulst, R.
Fokkens, N. M. M. Nibbering, P. Timmerman, D. N. Reinhoudt, Chem.
Commun. 1998, 1021 ± 1022; e) H. Hioki, W. C. Still, J. Org. Chem.
1998, 63, 904 ± 905.
[13] U. M. Sundram, J. H. Griffin, J. Am. Chem. Soc. 1996, 118, 13107 ±
13108.
[14] J. Rao, G. M. Whitesides, J. Am. Chem. Soc. 1997, 119, 10286 ±
12090.
[15] T. Staroske, D. H. Williams, Tetrahedron Lett. 1998, 39, 4917 ± 4920.
[16] M. Adamczyk, J. A. Moore, S. D. Rege, Z. Yu, Bioorg. Med. Chem.
Lett. 1999, 9, 2437 ± 2440.
[17] D. R. Stack, R. G. Thompson, EP 0801075A1, 1997, CAN 127:
359109.
‡3
the z region (Hewlett Packard 1100 Series, electrospray). From these
data, the relative amounts of each dimer were determined taking into
account the expected statistical abundance for each compound [i.e., the
abundance of heterodimers was halved due to the occurrence of two
equivalent combinations for each, e.g. 8-(LeuNMe)C2-(LeuNMe)C4 is
equivalent in mass to 8-(LeuNMe)C4-(LeuNMe)C2]. The average result of
three experiments are plotted in Figure 12.
Acknowledgement
We thank Dr. D. H. Huang and Dr. G. Siuzdak for NMR and mass
spectroscopic assistance, respectively. Financial support for this work was
provided by The Skaggs Institute for Chemical Biology and the National
Institutes of Health (USA), fellowships from The Skaggs Institute for
Research (to R.H.), the American Chemical Society Division of Organic
Chemistry (sponsored by Hoechst Marion Roussel to R.H.) and the
George Hewitt Foundation (to N.W.), and grants from Amgen, Bayer AG,
Boehringer-Ingelheim, Glaxo, Hoffmann-La Roche, DuPont, Merck,
Novartis, Pfizer, and Schering Plough.
[18] a) J. Rao, J. Lahiri, R. M. Weis, G. M. Whitesides, J. Am. Chem. Soc.
2000, 122, 2698 ± 2710; b) J. Rao, J. Lahiri, L. Isaacs, R. M. Weis, G. M.
Whitesides, Science 1998, 280, 708 ± 711.
[19] H. Arimoto, K. Nishimura, T. Kinumi, I. Hayakawa, D. Uemura,
Chem. Commun. 1999, 1361 ± 1362.
[1] R. Leclerq, E. Derlot, J. Duval, P. Courvalin, N. Engl. J. Med. 1998,
319, 157 ± 161; b) K. Hiramatsu, Drug Resist. Updates 1998, 1, 135 ±
150.
[2] a) T. J. Dougherty, J. F. Barrett, Curr. Opin. Anti-Infect. Invest. Drugs
1999, 1, 18 ± 25; b) J. Zhu, Expert Opin. Ther. Patents 1999, 9, 1005 ±
1019; c) M. B. Schmid, Curr. Opin. Chem. Biol. 1998, 2, 529 ± 534;
d) K. Bush, Curr. Opin. Chem. Biol. 1997, 1, 169 ± 175.
[3] K. C. Nicolaou, S. Y. Cho, R. Hughes, N. Winssinger, C. Smethurst, H.
Labischinski, R. Endermann, Chem. Eur. J. 2001, 7, 3798 ± 3823,
preceding paper in this issue.
[4] a) J. P. Mackay, U. Gerhard, D. A. Beauregard, R. A. Maplestone,
D. H. Williams, J. Am. Chem. Soc. 1994, 116, 4573 ± 4580; b) P. J. Loll,
A. E. Bevivino, B. D. Korty, P. H. Axelsen, J. Am. Chem. Soc. 1997,
119, 1516 ± 152; c) J. P. Waltho, D. H. Williams, J. Am. Chem. Soc.
1989, 111, 2475 ± 2480; d) T. Staroske, D. P. OꢀBrien, T. J. D. Jorgensen,
P. Roepstorff, D. H. Williams, A. J. R. Heck, Chem. Eur. J. 2000, 6,
504 ± 509.
[5] a) M. A. Cooper, D. H. Williams, Chem. Biol. 1999, 6, 891 ± 899;
b) B. S. Westwell, B. Bardsley, R. J. Dancer, A. C. Try, D. H. Williams,
Chem. Commun. 1999, 589 ± 590; c) D. A. Beauregad, D. H. Williams,
M. N. Gwynn, D. J. C. Knowles, Antimicrob. Agents Chemother. 1995,
39, 781 ± 785.
[20] Malabarba, T. I. Nicas, R. C. Thompson, Med. Res. Rev. 1997, 17, 69 ±
137.
[21] For reviews of olefin methathesis, see: a) R. H. Grubbs, S. Chang,
Tetrahedron 1998, 54, 4413 ± 4450; b) A. Fürstner, Top. Organomet.
Chem. 1998, 1, 37 ± 72; c) M. Schuster, S. Blechert, Angew. Chem. 1997,
109, 2150 ± 2152; Angew. Chem. Int. Ed. Engl. 1997, 36, 2036 ± 2056;
d) A. Fürstner, Angew. Chem. 2000, 112, 3140 ± 3172; Angew. Chem.
Int. Ed. 2000, 39, 3012 ± 3043.
[22] D. M. Lynn, S. Kanaoka, R. H. Grubbs, J. Am. Chem. Soc. 1996, 118,
784 ± 790.
[23] B. Mohr, D. M. Lynn, R. H. Grubbs, Organometallics 1996, 15, 4317 ±
4325.
[24] a) S. B. Shuker, P. J. Hajduk, R. P. Meadows, S. W. Fesik, Science 1996,
274, 1531 ± 1534; b) D. J. Maly, I. C. Choong, J. A. Ellman, Proc. Natl.
Acad. Sci. USA 2000, 97, 2419 ± 2424; c) D. A. Erlanson, A. C.
Braisted, D. R. Raphael, M. Randal, R. M. Stoud, E. M. Gordon,
J. A. Wells, Proc. Natl. Acad. Sci. USA 2000, 97, 9367 ± 9372.
[25] M. Nieto, H. R. Perkins, Biochem. J. 1971, 124, 845 ± 852.
[26] a) T. J. D. Jorgensen, P. Roepstroff, A. J. R. Heck, Anal. Chem. 1998,
70, 4427 ± 4432; b) T. J. D. Jorgensen, T. Staroske, P. Roepstorff, D. H.
Williams, A. J. R. Heck, J. Chem. Soc. Perkin Trans. 2 1999, 1859 ±
1863.
[6] M. Mammen, S.-K. Choi, G. M. Whitesides, Angew. Chem. 1998, 110,
2908 ± 2953; Angew. Chem. Int. Ed. 1998, 37, 2754 ± 2794.
[7] D. H. Williams, A. J. Maguire, W. Tsuzki, M. S. Westall, Science 1998,
280, 711 ± 714.
[8] a) U. Gerhard, J. P. Mackay, R. A. Maplestone, D. H. Williams, J. Am.
Chem. Soc. 1993, 115, 232 ± 237; b) J. P. Mackay, U. Gerhard, D. A.
Beauregard, M. S. Westall, M. S. Searle, D. H. Williams, J. Am. Chem.
Soc. 1994, 116, 4581 ± 4590.
[27] P. J. Loll, R. Miller, C. M. Weeks, P. H. Axelsen, Chem. Biol. 1998, 5,
293 ± 298.
[9] a) J.-M. Lehn, A. V. Eliseev, Science 2001, 291, 2331 ± 2332; b) I. Huc,
J.-M. Lehn, Proc. Natl. Acad. Sci. USA 1997, 94, 2106 ± 2110; c) J.-M.
Received: April 9, 2001 [F3190]
Chem. Eur. J. 2001, 7, No. 17
ꢁ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0717-3843 $ 17.50+.50/0
3843