Towards peptide isostere libraries: Aqueous aldol reactions on hydrophilic solid supports

Article abstract of DOI:10.1039/a909246c

Polar polyoxyethylene-polyoxypropylene (POEPOP) resin was derivatised with a 4-hydroxymethyl phenoxy linker and used as a solid support for lanthanide triflate catalysed Mukaiyama type solid phase aldol reactions. The conditions were optimised using resin bound 4-alkoxybenzaldehyde and it was shown that use of an aqueous solvent was crucial. Also, the reactions were performed with an N-terminal peptide aldehyde substrate in very high yields. POEPOP resins prepared with different monomer chain lengths and commercially available PEG-grafted NovaSyn TG resin were compared in the reactions.

Full text of DOI:10.1039/a909246c

View Article Online / Journal Homepage / Table of Contents for this issue
Towards peptide isostere libraries: aqueous aldol reactions on
hydrophilic solid supports
Anette Graven, Morten Grøtli and Morten Meldal*
1
Centre for Solid Phase Organic Combinatorial Chemistry (SPOCC),
Department of Chemistry, Carlsberg Laboratory, Gamle Carlsberg Vej 10, DK-2500 Valby,
Denmark
Received (in Cambridge, UK) 23rd November 1999, Accepted 22nd December 1999
Polar polyoxyethylene-polyoxypropylene (POEPOP) resin was derivatised with a 4-hydroxymethyl phenoxy
linker and used as a solid support for lanthanide triflate catalysed Mukaiyama type solid phase aldol reactions.
The conditions were optimised using resin bound 4-alkoxybenzaldehyde and it was shown that use of an aqueous
solvent was crucial. Also, the reactions were performed with an N-terminal peptide aldehyde substrate in very high
yields. POEPOP resins prepared with different monomer chain lengths and commercially available PEG-grafted
NovaSyn TG resin were compared in the reactions.
to the interior of the resins, were previously developed in
Introduction
this laboratory.30–35 Thus, polyoxyethylene–polyoxypropylene
(POEPOP) resin, containing only ether and hydroxy function-
alities, was designed as a chemically inert hydrophilic resin
compatible with the aqueous conditions used in enzymatic reac-
tions and screening and also with various conditions generally
used in organic synthesis. In the context of performing organic
reactions on peptide derived substrates attached to a solid
support, the use of polar, hydrophilic resins and water tolerant
catalysts is a very attractive approach, which greatly facilitates
the ease and robustness of the operation since time-consuming
steps to dry the hygroscopic solid supports are eliminated.
This paper describes model studies for performing lanthanide
triflate catalysed aldol reactions on hydrophilic solid sup-
ports in aqueous media, including the first examples of
solid phase aldol reactions of an N-terminal peptide alde-
hyde. Also, a comparison of POEPOP resin with commercially
available hydrophilic PEG-grafted NovaSyn TG resin is
made.
In recent years, solid phase synthesis has emerged as a powerful
tool for creating large numbers of highly diverse compounds for
utilisation in various screening protocols, and libraries of
organic molecules and peptides have proven to be highly effi-
cient for the discovery process of new therapeutic lead com-
pounds.1 Linear peptides are generally not suited for use as
drugs because of their poor oral availability and fast metabol-
ism in the body, however, it is possible to improve resistance
towards proteases through several types of modifications.
Biologically active dipeptide mimetics as well as linear pep-
tides in which an amide bond has been replaced by a non-
hydrolysable bond or by a functionality which mimics the
transition state of hydrolysis, have been described as protease
inhibitors.2–7 Currently, work is progressing in this laboratory to
expand the concept of preparing combinatorial libraries of
peptides to include the synthesis of peptide libraries containing
amide bond isosteres formed in a variety of organic reactions
on the solid phase, for on-bead screening for endoprotease
inhibitors.
The aldol reaction is one of the most powerful organic reac-
tions for the formation of carbon–carbon bonds,8–10 and several
examples of solid phase aldol reactions have appeared in the
literature.11–18 Among the myriad of variants of the aldol reac-
tion, the Mukaiyama-directed cross aldol reaction19–22 has been
extensively utilized. In this transformation, silyl enol ethers
derived from ketones, aldehydes, esters, thioesters or amides are
reacted with aldehydes in the presence of a catalyst in either
stoichiometric or sub-stoichiometric amounts. By far, most of
the common catalysts for this reaction are Lewis acids, and
chiral induction has been successfully demonstrated in the aldol
reaction by use of chiral Lewis acids.23–26 However, as most
Lewis acids are very sensitive to water, the reactions normally
require strictly anhydrous conditions as even small amounts
of water significantly lower the yields due to decomposition of
the catalyst and hydrolysis of the silyl enol ethers. Thus, the
discovery of lanthanide trifluoromethanesulfonates (triflates) as
water tolerant Lewis acids27–29 has greatly expanded the scope
of the reaction, especially from an industrial point of view.
Also, the use of scandium triflate as an efficient catalyst in solid
phase aldol reactions has been reported, albeit in non-aqueous
media.14
Results and discussion
Derivatisation of resins
Preparation of POEPOP–HMP resins. To enable a com-
parison of the synthetic utility of POEPOP resin in the work
presented here, one of the most widely used commercially avail-
able hydrophilic resins, NovaSyn TG–HMP, was selected as a
reference system. In order to make a direct comparison it was
necessary to derivatise POEPOP, made with different PEG-
monomer lengths, with the 4-hydroxymethylphenoxy (HMP)
linker moiety. Initially, POEPOP was per-mesylated and
reacted in an SN2 reaction with several equivalents of HMP
overnight in DMF at temperatures ranging from 50–90 ЊC in
the presence of Cs2CO3. Since this strategy did not satisfactorily
incorporate HMP into the resin, probably due to the presence
of secondary alcohol functionalities in POEPOP, attention was
turned towards Mitsunobu chemistry.36 Several examples of
successful solid phase Mitsunobu ether formations have been
published.37–43 As HMP is a bifunctional compound with
respect to Mitsunobu couplings, the benzylic alcohol was
protected with the dimethoxytrityl (DMT) group to avoid
polymerisation. DMT was selected as it allows simple and
direct monitoring and for incorporation to be quantified by
cleavage of a known amount of resin by treatment with mild
Several novel solid supports compatible with conditions
used in enzymatic reactions and allowing enzymes to penetrate
DOI: 10.1039/a909246c
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962
This journal is © The Royal Society of Chemistry 2000
955

View Article Online
Table 1 Mitsunobu reactions on POEPOP and NovaSyn TG-OH
Entry
Resin
Conditions
Yield(%)a
14
1
2
POEPOP400
POEPOP400
POEPOP400
POEPOP400
POEPOP400
POEPOP1500
POEPOP1500
POEPOP1500
POEPOP1500
POEPOP1500
POEPOP900
NovaSyn TG-OH
DEADb–PPh3–1 (1:1:1) (5 eq.)
0.12 M in THF
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.12 M in NEM
<10
3
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.12 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.25 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:0.6) (5 eq.)
0.25 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.05 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.1 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (10 eq.)
0.2 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (10 eq.)
0.2 M in THF–DCM (1:1) × 2
2:1 (1:1) (5 eq.)
68
>95
17
4
5
6
<10
47
7
8
49
9
45
10
11
12
51
0.08 M in DCM
DEAD–PPh3–1 (1:1:1) (5 eq.)
0.1 M in THF–DCM (1:1)
DEAD–PPh3–1 (1:1:1) (5eq.)
0.25 M in THF–DCM (1:1)
53
21
a Determined spectrophotometrically at 498 nm. b DEAD: diethyl azodicarboxylate.
acid followed by spectrophotometric quantitation of the
released DMT cations. The results are summarised in Table 1.
Initially, literature procedures38,41,42 were tested (entries 1–3).
One of the most widely used solvents in Mitsunobu chemistry
is THF. However, using this solvent gave unsatisfactory results
in the coupling between 1 and POEPOP400 resin (entry 1). The
swelling of POEPOP is very low in THF, possibly preventing
good solvation and accessibility of resin bound reactants. In
contrast, when 50% THF in DCM was used, the swelling of
POEPOP was improved, and good yields were obtained (entry
3). The yield could be further improved by increasing the con-
centration of reagents (entry 4). It was not possible to obtain
the reaction using N-ethyl morpholine as the solvent (entry 2).
The importance of using equimolar amounts of reagents is
demonstrated in entry 5. Changing to POEPOP1500 led to a
significant drop in coupling efficiency (entry 6); this may be
explained by the fact that POEPOP1500 has a much higher
swelling and also a lower loading compared with POEPOP400
,
leading to lowered concentrations of both reagents and resin
bound reactant. To counteract this dilution, care was taken to
diminish the amount of solvent used to increase the concen-
tration of reagents, leading to a somewhat improved yield
(entry 7). However, increasing the concentration of reagents
further to the level optimal for POEPOP400 had no effect on the
yield (entry 8). Surprisingly, double coupling did not increase
the level of incorporation (entry 9). A Mitsunobu-like process
has been reported by Castro et al.44 using the triphenyl-
phosphine-cyclic sulfamide betaine 2, and the reaction has been
successfully applied on solid phase.45–47 However, no improve-
ment was obtained using
2 in the coupling between
POEPOP1500 and 1 (entry 10). Similar results were obtained
with POEPOP900 (entry 11). For comparison, 1 was also reacted
with NovaSyn TG-OH resin under the same conditions, leading
to an incorporation of only 21% (entry 12).
To eliminate any interference from any residual free hydroxy
groups remaining on POEPOP900 and POEPOP1500 after deriv-
atisation, the resins were capped using acetic anhydride in
pyridine prior to deprotection of the linker.
tion of sodium 4-phenoxybenzaldehyde to provide polymer
supported 4-alkoxybenzaldehydes 4a–d (Scheme 1). The yields
were quantified by NMR of the crude product by cleaving a
known amount of resin against an added internal standard
(EtOAc). The reactions worked equally well for all resins, as the
loadings of 4-alkoxybenzaldehyde were always above 80%
compared with initial loading of linker. Products 4a–d exhib-
Preparation of polymer supported 4-alkoxybenzaldehyde.
POEPOP–HMP resins or NovaSyn TG–HMP, obtained from a
commercial source, were mesylated and added to a DMF solu-
ited a typical carbonyl IR absorbance at 1698 cmϪ1
.
956
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962

View Article Online
Table 2 Solution phase aldol reactions of 4-hydroxybenzaldehyde using 10 mol% of lanthanide triflates as catalysts at rt
Silyl enol
ether
Yield
(%)b
Entry
Catalyst
Solvent (ratio)
Producta
1
2
3
4
Yb(OTf)3
Y(OTf)3
Yb(OTf)3
Y(OTf)3
5
5
5
5
THF–H2O (4:1)
THF–H2O (4:1)
THF–DCM (2:3)
THF–DCM (2:3)c
8a
20
25
40
65
35
75
30
8a
8b
8a
8c
5
6
Yb(OTf)3
Y(OTf)3
6
6
THF–H2O (4:1)
7ad
7ce
THF–DCM (2:3) c
a Products identified by NMR of crude mixture after aqueous work-up. b Yield determined by NMR of crude mixture by comparing signals of
products and anisaldehyde. c Performed at 0 ЊC. d syn:anti: 75:25. e 50:50.
Scheme 1 Derivatisation of resins 3a–d with 4-hydroxybenzaldehyde.
Aldol reactions
Lanthanide triflate catalysed aldol reactions in aqueous media
were initially reported by Kobayashi and Hachiya,27,28 enabling
the facile use of aqueous formaldehyde solutions in the reaction
as well as several other aldehydes. In the work presented here,
initial solution experiments were conducted with model sub-
strates to optimise the conditions. 4-Methoxybenzaldehyde was
selected as the aldehyde component along with silyl enol ethers
5 and 6 representative of silyl enol ethers reacting from primary
or secondary positions, respectively. The results are summar-
ized in Table 2. The conditions established by Kobayashi et al.
for the ytterbium triflate (Yb(OTf)3) catalysed reaction of
4-methoxybenzaldehyde with 6, as given in entry 5, lead to the
aldol product 7a in 75% yield. However, reaction of silyl ether 5
only gave 20% of product 8a under these conditions, due to
rapid hydrolysis of 5 (entry 1). Consequently, the catalyst was
substituted with the weaker Lewis acid yttrium triflate
(Y(OTf)3). This did not lead to any significant improvement
(entry 2). However, when the aqueous solvent was substituted
with an organic mixture of THF–DCM, the yields were
increased. Using Yb(OTf)3 under these conditions led to the
formation of the dehydrated product 8b (entry 3), whereas
using Y(OTf)3 as the catalyst gave a mixture of the aldol
product 8a and the silylated aldol product 8c in quantitative
yield after aqueous work-up (entry 4). Reaction of 6 under
the latter conditions led to only a 30% yield of silylated aldol
product 7c (entry 6).
Next, the reactions were attempted on a solid phase. The
aldol reactions were studied using a stoichiometric amount of
catalyst in different solvents. The products were cleaved from
the resins by treating with TFA:H2O (95:5) for 2 h at ambient
temperature and analysed by NMR. Under the cleavage
conditions, the aldol products eliminated water to form the
α,β-unsaturated compounds 9 and 10. This was also shown in
control experiments in which 7a and 8a were submitted to the
cleavage conditions in solution and were shown to be converted
into 7b and 8b, respectively. No retro-aldol reaction was
observed. In the case of the reaction of 4d with 6, it was
demonstrated that the primary product formed on the solid
phase, as in solution, was the aldol product 11 (Scheme 2).
On-bead 1D-1H- and 2D-1H-NMR spectroscopy (COSY,
TOCSY) were aquired with a resin sample from the reaction
Scheme 2 Solid phase aldol reaction of 4d and 6.
using nanoprobe MAS-NMR spectroscopy. From these
spectra, the structure of the expected aldol product (11) was
assigned. Also, the on-bead 1D-spectrum showed the syn:anti
ratio of 79:21 obtained on the solid phase to be close to that
obtained in solution (75:25) (Fig. 1). By analogy with this
experiment and the solution phase studies, it is most likely that
the primary products formed on the solid phase are the aldol
products. The results of the solid phase aldol reactions are
summarised in Table 3.
Initially, the optimal solution conditions determined for the
reaction of 4-methoxybenzaldehyde with 5 were tested with
resin 4a, but in contrast to the solution reaction no product was
obtained. It has been shown that when running Yb(OTf)3
catalysed aldol reactions in THF in solution, the use of an
additive increased the yields, water being the most efficient.27
Thus, when 50 equivalents of water relative to the catalyst were
added, the highest yields were obtained in the reactions. Con-
sequently, these experimental conditions were attempted on the
solid phase by adding 50 equivalents of water to the organic
solvent mixture; however, this only led to a yield of less than
10% (entry 2). It was suspected that the catalyst was not
sufficiently dissolved to enter the interior of the resin, hence, the
Yb(OTf)3 catalysed aldol reaction using 6 was performed in an
organic solvent mixture of CH3CN–DCM in which Yb(OTf)3
is completely soluble (entry 3), yet no reaction was observed.
These findings are in contrast to reported solid phase aldol
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962
957

View Article Online
Table 3 Solid phase aldol reactions of resin bound 4-hydroxybenzaldehyde using stoichiometric amounts of lanthanide triflates as catalysts at rt
Silyl enol
ether
Yield
(%)b
Entry
Resin
Catalyst
Solvent (ratio)
Producta
1
2
3
4
5
6
7
8
9
10
11
12
13
4a
Y(OTf)3
Y(OTf)3
Yb(OTf)3
Y(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3
Yb(OTf)3 g
Yb(OTf)3 g
Yb(OTf)3
5
5
6
5
5
6
6
6
6
5
6
5
6
THF–DCM (2:3)c
THF–DCM (2:3)d
CH3CN–DCM (3:2)
THF–H2O (4:1)
THF–H2O (4:1)
THF–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
CH3CN–H2O (4:1)
—
<10
4a
9
4a
—
27
35
75
81
91e
10
28
80
32
83
4a
9
9
10
10
10
10
9
10
9
10
4a
4a
4a
4a
4bf
4cf
4cf
4df
4df
a Products obtained after cleavage from solid supports as identified by NMR. b Yield is determined by NMR of the crude cleavage mixture by
comparing signals from starting materials and products. All recoveries, as determined by integrating signals from starting materials and product
against an added internal standard (EtOAc), are above 60% compared with initial loading. c Performed at 0 ЊC. d 50 Equivalents of water relative to
the catalyst were added. e The resin was reacted twice in the aldol reaction. f Resin was re-capped prior to use. g Performed with 0.2 equivalents of
catalyst relative to aldehyde.
when POEPOP400 was used as the solid support, a dramatic
decrease in the yield of the reaction was observed, whereas
POEPOP900 and POEPOP1500 led to identical results to those
obtained using NovaSyn TG as solid support as shown in
entries 10–13 (Table 3). However, it was found that with
POEPOP resin, it was important to re-cap any free hydroxy
groups that may have been released on the resin backbone
under the hydrolytic conditions during the washing of the resins
after alkylation, as the yields were otherwise lowered. It has
been previously reported that salicylaldehyde is a substrate for
the lanthanide triflate catalysed aldol reaction,27 our results,
however, indicate that this tolerance to a free hydroxy group in
the reaction is probably the exception rather than the rule. As
shown for POEPOP1500 in entries 12 and 13 (Table 3), the reac-
tions performed equally well when only a catalytic amount
(20 mol% relative to aldehyde) of Yb(OTf)3 was used.
1
To explore the utility of the reaction in the context of prepar-
ing peptide isostere libraries for solid phase screening, the
reaction conditions were tested on a peptide like substrate.
Peptide 12 (Scheme 3) was selected as an N-terminal aldehyde
model substrate. The regularly spaced amide and ester
carbonyls could potentially act as bidentate ligands to the cata-
lyst due to the oxyphilic nature of lanthanides, and thereby
influence the results of the reactions. As polyethyleneglycol
(PEG) resins based on PEG1500 are biocompatible with enzyme
reactions and the resins of choice for preparing libraries for
solid phase screening, it was appropriate to prepare 12 on
POEPOP1500 only. Using the optimised conditions determined
for the reactions of 4a–d, peptide 12 was shown to undergo
nearly quantitative reactions with 5 as well as 6 in the presence
of 0.2 equivalents of catalyst relative to the aldehyde as shown
in Table 4. These results indicate that even in the presence of
several amide and ester carbonyls, Yb(OTf)3 shows a strong
preference for coordinating to the aldehyde moiety, facilitating
sufficiently rapid aldol reactions to efficiently compete with
hydrolysis of the silyl enol ethers, resulting in very high yields of
aldol product using only catalytic amounts of catalyst.
Fig. 1 (a) H-NMR spectrum of the diastereomeric mixture of aldol
products obtained from the reaction of anisaldehyde and 6. The signals
at δ 4.68 and 5.25 correspond to the protons α to the hydroxy groups
(syn and anti isomers, respectively). (b) Nanoprobe MAS 1H-NMR
spectrum of the diastereomeric mixture of aldol products obtained
from the reaction of 4d and 6. The signals at δ 4.72 and 5.28 correspond
to the protons α to the hydroxy groups (syn and anti isomers, respect-
ively), proving that the primary products obtained on the solid phase
are the aldol products, and that elimination takes place during cleavage
from the resin. The signals at δ 4.95, 6.91 and 7.19 belong to the HMP-
linker. Also, the spectrum displays two CDCl3 signals, one from solvent
inside and outside the bead, respectively. The signals from δ 2.91 to 4.55
are residual resin peaks.
reactions in which insoluble Sc(OTf)3 was used as a catalyst in
DCM for reactions run on a polystyrene support.14 These com-
bined results suggest that the nature of the resins used play an
important role in all these reactions, and that addition of water
is crucial when performing the reactions on polar, hydrophilic
supports, possibly due to low mobility of the catalyst caused by
chelation of resin polyethyleneglycol in the absence of water. In
contrast, running the Y(OTf)3 catalysed aldol reaction of 5
with an aqueous solvent increased the yield to 27% (entry 4),
paralleling the yield obtained using these conditions in solu-
tion; this was further increased to 35% upon changing the
catalyst to Yb(OTf)3, thus exceeding the yield obtained in solu-
tion under similar conditions. The optimal conditions found for
the reaction of 6 in solution gave equally good results on the
solid phase with 4a. As the POEPOP polymers may not be well
solvated in THF, an alternative solvent mixture in which THF
was exchanged for acetonitrile was tested in the reaction of 4a
with 6 leading to an improved yield of 81%, which could be
further improved to 91% upon double coupling. Surprisingly,
The results are promising for expanding the concept of per-
forming aldol reactions on N-terminal peptide aldehydes to
form combinatorially generated libraries of protease inhibitors,
and currently work is being carried out towards this aim.
Conclusion
In the work presented here, the first examples of solid phase
aldol reactions in aqueous media, including aldol reactions of
an N-terminal peptide aldehyde substrate, using a polar, hydro-
philic resin and catalytic amounts of Yb(OTf)3 are reported.
958
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962

View Article Online
Scheme 3 Solid phase aldol reactions of a model peptide substrate. Numberings refer to numbering used in interpretation of NMR data (see
Experimental section).
Table 4 Solid phase aldol reactions of peptide 12 using 0.2 equivalents
of Yb(OTf)3 in CH3CN–H2O (4:1) at rt
Experimental
Novasyn TG–HMP resin, Novasyn TG–OH resin, 1-
(mesitylene-2-sulfonyl)-3-nitro-1H-1,2,4-triazole (MSNT) and
Silyl enol
ether
Yield
(%)b
Producta
O-(benzotriazol-1-yl)-N,N,NЈ,NЈ-tetramethyluronium
tetra-
Entry
fluoroborate (TBTU) were purchased from Novabiochem
(Läufelfingen, Switzerland). N-(Fluoren-9-ylmethoxycarbonyl)
(Fmoc)-leucine and Fmoc-phenylalanine pentafluorophenyl
(Pfp) ester were obtained from Bachem (Bubendorf,
Switzerland). 1-Methylimidazole, PPh3, DEAD, 4-hydroxy-
1
2
5
6
13a, 13bc
14
>95
>95d
a Products obtained after cleavage from solid support as identified by
NMR of crude cleavage mixture. b Yield is determined by NMR of the
crude cleavage mixture by comparing signals from starting material and
product. c Ratio between 13a and 13b (E and Z isomers): 2:1:1. d E:Z:
initially 94:6, equilibrating to 50:50 over one week.
methylphenol,
methylsulfonyl(mesyl)chloride,
caesium
carbonate, 4-hydroxybenzaldehyde, Y(OTf)3, Yb(OTf)3,
1-phenyl-1-trimethylsilyloxyethylene and 1-(trimethylsilyloxy)-
cyclohexene were purchased from Fluka (Buchs, Switzerland).
p,pЈ-Dimethoxytrityl chloride and 4-carboxybenzaldehyde were
obtained from Aldrich (Steinheim, Germany). Trichloroacetic
acid was purchased from Merck (Darmstadt, Germany) and
trifluoroacetic acid from Merck-Schuchardt (Hohenbrun,
Germany). Piperidine was purchased from Riedel-de-Häen
(Seelze, Germany). All solvents were HPLC grade and were
purchased from LabScan (Dublin, Ireland). All reagents and
solvents were used as received without further purification.
POEPOP resins32 and compound 244 were prepared according
to literature procedures. Nanoprobe MAS NMR-spectra were
recorded on a Varian Unity Inova 500 MHz spectrometer
equipped with a 4 mm Nano NMR probe at 25 ЊC using a spin
rate of approximately 2000 Hz. Solution phase NMR spectra
were recorded on a Varian Unity Inova 500 MHz spectrometer
or a Bruker AMX 250 MHz spectrometer. All J values are given
in Hz.
The reactions proceed in high yields, which is important for the
methodology to be successfully adapted to synthesis in a com-
binatorial one-bead-one-compound format. In the course of
optimising the conditions for the reactions, it was discovered
that the use of an aqueous solvent is essential for the success of
the reaction on polar solid supports. This is the first observation
of this phenomenon. Hydrophilic POEPOP resins, compatible
with performing on-bead enzymatic reactions and screenings
and prepared using different monomer lengths, were derivatised
with the HMP linker using Mitsunobu chemistry. Also, the
POEPOP resins were compared with commercially available
NovaSyn TG-OH or NovaSyn TG–HMP in all reactions
reported. In the aldol reaction, identical results were obtained
using NovaSyn TG, POEPOP900 or POEPOP1500, whereas only
minor yields were obtained using POEPOP400. In the SN2-
reactions between mesylated 3a–d and 4-hydroxybenzaldehyde,
all resins performed equally well and high yields were obtained
in all cases. However, in the Mitsunobu reaction signifi-
cantly higher yields were obtained using the POEPOP resins,
particularly POEPOP400. The results obtained, comparing
POEPOP to NovaSyn TG, thus suggest that the synthetic
utilities of POEPOP900 and POEPOP1500 resin are similar to that
of NovaSyn TG, whereas POEPOP400 differs significantly from
both NovaSyn TG and POEPOP resins made from PEG-
monomer of longer chain lengths.
4-(p,pЈ-Dimethoxytrityl)oxymethylphenol (DMT–HMP, 1)
A solution of p,pЈ-dimethoxytrityl chloride (8.5 g, 25 mmol)
in pyridine (40 cm3) was added dropwise over 2 h to a stirred
solution of 4-hydroxybenzyl alcohol (3.1 g, 25 mmol) in pyrid-
ine (60 cm3) at 0 ЊC and stirring was continued for 18 h at
ambient temperature. The reaction was quenched by addition
of methanol (20 cm3) and the solvents were removed under
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962
959

View Article Online
reduced pressure. The residue was dissolved in DCM (50
cm3) and washed with saturated aqueous sodium hydrogen
carbonate (3 × 50 cm3) and brine (50 cm3). The organic
fraction was dried and evaporated under reduced pressure to
leave a brown oil, which was purified by flash column chrom-
atography on silica using ethyl acetate–n-heptane (15:85 to
40:60) containing 1% triethylamine as eluent to give 1 (10.2 g,
95%) as a light yellow foam. δH (250 MHz; CDCl3) 3.67 (6H, s,
OMe), 3.96 (2H, s, O–CH2–Ar), 6.68–7.49 (17H, Ar); δC (67.8
MHz; CDCl3) 55.21 (OMe), 65.23 (O–CH2–Ar), 99.58 (O–C–
Ar3), 113.09 (–CH–C–OMe), 115.14 (–CH–C–OH), 126.69
(Ph), 127.79 (Ph), 128.23 (Ph), 128.65 (–CH–CH–C–OH),
130.09 (–CH–CH–C–OMe), 131.36 (–C–(CH)2–C–OH), 136.44
(–C–(CH)2–C–OMe), 145.13 (Ph), 154.84 (C–OH), 158.40
(C–OMe).
washed with DCM (5×) and DMF (3×) and was added to the
preformed DMF solution of sodium 4-carboxyphenolate. The
reaction was stirred overnight at 90 ЊC. The resin was drained
and washed with DMF (5×), DMF:water (1:1) (5×), water
(5×), acetonitrile (5×) and DCM (5×) and dried in vacuo. The
incorporation level was determined after cleavage of 50 mg of
resin with 95% trifluoroacetic acid–H2O (95:5) (2 cm3) for 2 h at
ambient temperature. The cleavage mixture was drained from
the resin and the resin was rinsed with acetonitrile (5 × 2 cm3).
The cleavage mixture and the washings were combined and
evaporated under reduced pressure and the residue was quanti-
fied by 1H-NMR spectroscopy by integrating signals from
4-hydroxybenzaldehyde against a known amount of EtOAc as
an added internal standard. The incorporation was 0.22 mmol
g
Ϫ1 resin (85%).
Solid phase Mitsunobu reaction
Aqueous solution phase aldol reactions
A solution of PPh3 (590 mg, 2.25 mmol) in THF–DCM (1:1)
(0.5 cm3) was added dropwise to a stirred solution of DEAD
(350 mm3, 2.25 mmol) in THF–DCM (1:1) (0.5 cm3) in an
argon atmosphere at 0 ЊC. The bright red mixture was trans-
These were performed as described in ref. 27.
2-[Hydroxy(4-methoxyphenyl)methyl]cyclohexanone 7a. The
1
yield determined through H-NMR of the crude mixture by
ferred to the POEPOP1500 resin (1 g, loading: 0.45 mmol gϪ1
)
integrating signals from product and starting material, was
75%, syn:anti = 75:25. The NMR-spectrum of the product was
in accordance with published data.27
which was swollen in THF–DCM (1:1) (16 cm3) in an argon
atmosphere at 0 ЊC and mixed well. DMT–HMP (1) (960 mg,
2.25 mmol) was dissolved in THF–DCM (1:1) (2 cm3) and
added to the resin mixture. The reaction was allowed to run
overnight at ambient temperature. The solution was removed
by filtration and the resin was washed sequentially with THF–
DCM (1:1) (3×), DCM (5×), acetonitrile (5×) and DCM (3×)
and dried in vacuo. The substitution level determined spectro-
photometrically at 498 nm by cleavage of the DMT protecting
group from a known amount of resin using a 2% w/w solution
of trichloroacetic acid in DCM followed by calibration against
a standard curve was 0.21 mmol gϪ1 (47%). The remaining free
hydroxy groups on the resin were acetylated using acetic
anhydride in pyridine solution (1:1) (10 cm3) for 18 h. Depro-
tection of resin: the resin was washed with a 2% w/w solution of
trichloroacetic acid in DCM until the solution remains colour-
less. The resin was washed sequentially with DCM (5×), tetra-
hydrofuran (3×), acetonitrile (5×), water (5×), 10% aqueous
sodium hydrogen carbonate (3×), water (5×), acetone (5×) and
DCM (3×).
3-Hydroxy-3-(4-methoxyphenyl)-1-phenylpropan-1-one
8a.
1
The yield determined through H-NMR of the crude mixture
by integrating signals from product and starting material,
was 20% using Yb(OTf)3 as the catalyst and 25% using Y(OTf)3
as the catalyst. The NMR-spectrum of the product was in
accordance with published data.48
Non-aqueous solution phase aldol reactions
2-[Trimethylsilyloxy(4-methoxyphenyl)methyl]cyclohexanone
7c. 4-Hydroxybenzaldehyde (121 mm3, 1 mmol) and 1-(tri-
methylsilyloxy)cyclohexene (211 mm3, 1.1 mmol) were dis-
solved in THF (1 cm3) and added to a stirred suspension of
Y(OTf)3 in DCM (1.5 cm3) at 0 ЊC. After 1 h the reaction mix-
ture was allowed to come to ambient temperature and stirred
for another 1.5 h. At this point TLC showed no remaining silyl
enol ether. DCM (2 cm3) was added and the reaction mixture
was washed with water (3 × 5 cm3), dried and evaporated. The
1
crude mixture was analysed by H-NMR and the product was
Mitsunobu-like reactions using triphenylphosphine–cyclic
sulfamide betaine 2
identified in agreement with literature data.49 The yield,
determined through 1H-NMR of the crude mixture by integrat-
ing signals from product and starting material, was 30%,
syn:anti = 50:50.
POEPOP1500 (1 g, loading: 0.45 mmol gϪ1) was swollen in a
solution of DMT–HMP (1) (958 mg, 2.25 mmol) in DCM
(30 cm3) in an argon atmosphere. Triphenylphosphine–cyclic
sulfamide betaine 2 (1.04 g, 2.25 mmol) was added portionwise
over 15 min with careful mixing. After 48 h at ambient temper-
ature the resin was drained and washed with DCM (5×),
acetonitrile (5×) and DCM (3×). The level of incorporation
of the linker was determined as described in the typical
procedure for solid phase Mitsunobu reactions to be 0.23 mmol
gϪ1 (51%). The residual hydroxy functionalities on the resin
were acetylated and the linker was deprotected as described in
the procedure for solid phase Mitsunobu reactions.
The following examples were synthesised following the above
procedure.
3-(4-Methoxyphenyl)-1-phenylprop-2-en-1-one 8b. The yield
1
of 8b, determined through H-NMR of the crude mixture by
integrating signals from the product and starting material, was
40% using Yb(OTf)3 as the catalyst. The product was identified
by NMR and the spectrum was in accordance with
literature.50,51
3-Trimethylsilyloxy-3-(4-methoxyphenyl)-1-phenylpropan-1-
one 8c and 3-hydroxy-3-(4-methoxyphenyl)-1-phenylpropan-1-
one 8a. The yields, determined through H-NMR of the crude
Attachment of 4-hydroxybenzaldehyde to POEPOP–HMP
or NovaSyn TG–HMP was accomplished by the following
general method described for the preparation of resin 4a.
1
mixture by integrating signals from products (no detectable
starting material in crude product), were 65% 8a and 35% 8c
using Y(OTf)3 as the catalyst. The products were identified by
NMR and the spectra were in accordance with literature.48,49
Preparation of resin 4a
4-Hydroxybenzaldehyde (175 mg, 1.43 mmol) and sodium
hydroxide (52 mg, 1.3 mmol) were stirred in DMF (5 cm3) at
90 ЊC until all was dissolved. Mesyl chloride (405 mm3, 5.2
mmol) was added dropwise to a suspension of NovaSyn TG–
HMP (1 g, loading: 0.26 mmol gϪ1) in DCM:N,N-diisopropyl-
ethylamine (1:1) (10 cm3) at 0 ЊC and was stirred at 0 ЊC for 1 h
and at ambient temperature for 1 h. The resin was drained and
Typical procedure for solid phase aldol reactions
2-(4-Hydroxybenzylidene)cyclohexanone 10. Resin 4d (50 mg,
loading = 0.23 mmol gϪ1) was swollen in a solution of Yb(OTf)3
(1.4 mg, 0.0023 mmol) in acetonitrile–water (2:1) (375
960
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962

View Article Online
mm3) and allowed to stand for 1 h at ambient temperature.
1-(Trimethylsilyloxy)cyclohexene (22 mm3, 0.115 mmol) was
dissolved in acetonitrile (250 mm3) and added to the resin sus-
pension. The reaction was allowed to run overnight. The resin
was drained and washed with acetonitrile–water (4:1) (5×),
acetonitrile (5×) and DCM (5×). The product was cleaved from
the resin with 95% trifluoroacetic acid in water (3 cm3) for 2 h at
ambient temperature. The cleavage solution was drained from
the resin and the resin was washed with acetonitrile (5 × 3 cm3).
The cleavage solution and washings were combined and evap-
orated under reduced pressure. The residue was dissolved in
CDCl3 with a few drops of d6-DMSO added and analysed by
1H-NMR. The yield of 10 determined by integrating signals
from product and starting material was 83%. The product was
identified by NMR: δH (250 MHz; CDCl3) 1.71 (2H, m), 1.83
(2H, m), 2.44 (2H, m), 2.77 (2H, m), 6.80 (2H, d, J 8.8, Ar), 7.25
(2H, d, J 8.8, Ar), 7.41 (1H, br s, vinyl-H); δC (62.5 MHz;
CDCl3) 23.7, 24.3, 29.5, 40.8, 116.2 (Ar), 127.4 (Ar), 133.1 (Ar),
137.1 (vinyl-C), 158.8 (Ar).
139.95 (C-1), 166.59 (C-7), 171.25 (CO (Phe)), 174.72 (COOH
(Leu)), 192.16 (CHO); m/z (ESI) 411.3 [(M ϩ H)ϩ, 821.5
[(2M ϩ H)ϩ), 19], 843 [(2M ϩ Naϩ), 8].
Solid phase aldol reactions of peptide aldehyde 12 were per-
formed following the typical procedure described above using
0.2 equivalents of Yb(OTf)3 as the catalyst.
13a ؉ 13b. The yield of the combined products, determined
1
through H-NMR of the crude mixture by integrating signals
from products and starting material, was >95%. The product
was obtained as a mixture of 13a and the (E) and (Z) isomers
of 13b in a ratio of 2:1:1. Only a single diastereomer of 13a
was identified.
13a. δH (500 MHz; CDCl3) 0.61 (3H, d, J 6.1, Me (Leu)),
0.63 (3H, d, J 6.4, Me (Leu)), 1.32 (2H, m, β-H (Leu)), 1.39
(1H, m, γ-H (Leu)), 2.85 (1H, dd, J 8.3 and 14.3, β-H (Phe)),
2.95 (1H, dd, J 8.4 and 16.5, 9-H), 3.0 (1H, dd, J 5.1 and 14.3,
β-H (Phe)), 3.19 (1H, dd, J 4.1 and 16.5, 9-H), 4.20 (1H, m, α-H
(Leu)), 4.63 (1H, ddd, J 5.1, 8.0 and 8.3, α-H (Phe)), 5.07 (1H,
dd, J 4.1 and 8.4, 8-H), 6.9 (1H, t, J 7.3, Ph (Phe)), 6.96 (2H, dd,
J 7.3 and 7.9, Ph (Phe)), 7.0 (2H, d, J 7.9, Ph (Phe)), 7.19 (2H,
dd, J 7.5 and 7.9, 3Ј-H and 5Ј-H), 7.21 (2H, d, J 8.3, 2-H and
6-H), 7.30 (1H, t, J 7.5, 4Ј-H), 7.33 (1H, d, J 8.6, N–H (Leu)),
7.44 (2H, d, J 8.3, 3-H and 5-H), 7.46 (1H, d, J 8.0, N–H (Phe)),
7.67 (2H, d, J 7.9, 2Ј-H and 6Ј-H); δC (125 MHz, CDCl3) 21.94
(Me (Leu)), 23.06 (Me (Leu)), 24.83 (C-γ (Leu)), 37.78 (C-β
(Phe)), 41.18 (C-β (Leu)), 47.91 (C-9), 50.92 (C-α (Leu)), 54.66
(C-α (Phe)), 69.46 (C-8), 125.95 (C-2 and C-6), 126.61 (Ph
(Phe)), 127.48 (C-3 and C-5), 128.24 (C-2Ј and C-6Ј), 128.34
(Ph (Phe)), 128.72 (C-3Ј and C-5Ј), 129.56 (Ph (Phe)), 133.41
(C-1), 133.44 (C-4Ј), 137.02 (C-1Ј), 137.47 (Ph (Phe)), 148.25
(C-4), 167.09 (C-7), 171.24 (CO (Phe)), 174.38 (COOH), 198.67
(C-10); m/z (ESI) 531.2 [(M ϩ H)ϩ, 74%], 553.2 [(M ϩ Na)ϩ,
100].
3-(4-Hydroxyphenyl)-1-phenylprop-2-en-1-one 9. The reaction
between 4d and 5 was performed following the above procedure.
The yield, determined by integrating signals from the product
and starting material, was 32%. The product was identified in
accordance with published data.50
Preparation of peptide aldehyde 12
N-Fmoc-leucine (39 mg, 0.11 mmol), MSNT (33 mg, 0.11
mmol) and N-methylimidazole (8 mm3, 0.11 mmol) were com-
bined in DCM (3 cm3) and the mixture was added to POEPOP–
HMP resin (100 mg, loading = 0.22 mmol gϪ1). The reaction
was run at ambient temperature for 45 minutes and was then
repeated. The resin was washed with DCM (5×), DMF (5×)
and DCM (3×) and dried in a high vacuum. The loading,
measured by cleavage of a known amount of resin in 20%
piperidine in DMF solution and spectrophotometric analysis
of the released fluoren-9-ylpiperidine adduct against a standard
curve at 290 nm, was 0.22 mmol gϪ1. The resin bound N-Fmoc-
leucine was deprotected in 20% piperidine in DMF solution
and washed with DMF (6×). N-Fmoc-phenylalanine-Pfp ester
(61 mg, 0.11 mmol) was dissolved in DMF (2 cm3) and added to
the resin. The reaction was followed using the Kaiser test52 and
was washed with DMF (6×) when the reaction was complete.
The resin was deprotected as described above and washed with
DMF (6×). 4-Carboxybenzaldehyde (10 mg, 0.07 mmol) was
dissolved in DMF (2 cm3) and N-ethylmorpholine (14 mm3,
0.11 mmol) was added, followed by TBTU (20 mg, 0.06 mmol).
The mixture was left standing for 5 minutes before addition
to the resin and the coupling reaction was followed using the
Kaiser test. After coupling the resin was capped using acetic
anhydride in pyridine (1:1) for two hours at ambient tem-
perature and was washed with DCM (5×), acetone (5×),
water (5×), 10% aqueous sodium hydrogen carbonate (3×),
water (5×), acetonitrile (5×), DCM (5×) and dried in a high
vacuum. A resin sample was cleaved as described for the prep-
aration of 4a and peptide aldehyde 12 was analysed by NMR:
δH (500 MHz; CDCl3) 0.79 (3H, d, J 6.6, Me (Leu)), 0.81 (3H, d,
J 6.6, Me (Leu)), 1.48 (1H, m, β-H (Leu)), 1.57 (1H, m, γ-H
(Leu)), 1.59 (1H, m, β-H (Leu)), 3.07 (1H, dd, J 7.2 and 14, β-H
(Phe)), 3.18 (1H, dd, J 6.3 and 14, β-H (Phe)), 4.42 (1H, m, α-H
(Leu)), 4.83 (1H, m, α-H (Phe)), 7.09 (1H, d, J 8.2, N–H (Leu)),
7.09 (1H, t, J 7.3, Ph (Phe)), 7.15 (2H, dd, J 7.3 and 8.3, Ph
(Phe)), 7.19 (2H, d, J 8.3, Ph (Phe)), 7.60 (1H, d, J 7.7, N–H
(Phe)), 7.78 (2H, d, J 9.2, 2-H and 6-H), 7.81 (2H, d, J 9.2, 3-H
and 5-H), 9.9 (1H, s, CHO); δC (125 MHz; CDCl3) 22.43 (Me
(Leu)), 23.43 (Me (Leu)), 25.28 (C-γ (Leu)), 38.34 (C-β (Phe)),
41.77 (C-β (Leu)), 51.48 (C-α (Leu)), 55.27 (C-α (Phe)), 127.24
(Ph (Phe)), 128.45 (C-2 and C-6), 128.86 (Ph (Phe)), 129.98 (Ph
(Phe)), 130.13 (C-3 and C-5), 137.45 (Ph (Phe)), 138.68 (C-4),
13b. (E)-isomer: δH (500 MHz; CDCl3) 0.61 (3H, d, J 6.1, Me
(Leu)), 0.63 (3H, d, J 6.4, Me (Leu)), 1.32 (2H, m, β-H (Leu)),
1.39 (1H, m, γ-H (Leu)), 2.81 (1H, dd, J 8.3 and 14, β-H (Phe)),
2.97 (1H, dd, J 5.4 and 14, β-H (Phe)), 4.20 (1H, m, α-H (Leu)),
4.58 (1H, ddd, J 5.4, 8.1 and 8.3, α-H (Phe)), 6.9 (1H, t, J 7.3,
Ph (Phe)), 6.96 (2H, dd, J 7.3 and 7.9, Ph (Phe)), 7.0 (2H, d,
J 7.9, Ph (Phe)), 7.23 (2H, d, J 8.3, 2-H and 6-H), 7.25 (2H,
3Ј-H and 5Ј-H), 7.30 (1H, d, J 8.3, N–H (Leu)), 7.34 (1H, 4Ј-H),
7.35 (1H, d, J 15.87, 9-H), 7.49 (1H, d, J 15.87, 8-H), 7.51 (2H,
d, J 8.3, 3-H and 5-H), 7.53 (1H, d, J 8.1, N–H (Phe)), 7.76 (2H,
2Ј-H and 6Ј-H); δC (125 MHz, CDCl3) 21.94 (Me (Leu)), 23.06
(Me (Leu)), 24.83 (C-γ (Leu)), 37.72 (C-β (Phe)), 41.18 (C-β
(Leu)), 50.92 (C-α (Leu)), 54.72 (C-α (Phe)), 123.63 (C-9),
126.61 (Ph (Phe)), 126.78 (C-2 and C-6), 127.42 (C-8), 128.14
(C-3 and C-5), 128.34 (Ph (Phe)), 128.58 (C-2Ј and C-6Ј),
128.84 (C-3Ј and C-5Ј), 129.56 (Ph (Phe)), 133.18 (C-4Ј), 135.30
(C-1), 137.47 (Ph (Phe)), 137.93 (C-1Ј), 140.72 (C-4), 166.79
(C-7), 171.18 (CO (Phe)), 174.38 (COOH), 190.13 (C-10).
(Z)-isomer: δH (500 MHz; CDCl3) 0.61 (3H, d, J 6.1, Me
(Leu)), 0.63 (3H, d, J 6.4, Me (Leu)), 1.32 (2H, m, β-H (Leu)),
1.39 (1H, m, γ-H (Leu)), 2.81 (1H, dd, J 8.3 and 14, β-H (Phe)),
2.97 (1H, dd, J 5.4 and 14, β-H (Phe)), 4.20 (1H, m, α-H (Leu)),
4.58 (1H, ddd, J 5.4, 8.1 and 8.3, α-H (Phe)), 6.46 (1H, d, J 12.9,
9-H), 6.74 (1H, d, J 12.9, 8-H), 6.9 (1H, t, J 7.3, Ph (Phe)), 6.96
(2H, dd, J 7.3 and 7.9, Ph (Phe)), 7.0 (2H, d, J 7.9, Ph (Phe)),
7.21 (2H, 3Ј-H and 5Ј-H), 7.23 (2H, d, J 8.3, 2-H and 6-H), 7.30
(1H, d, J 8.3, N–H (Leu)), 7.34 (1H, 4Ј-H), 7.51 (2H, d, J 8.3,
3-H and 5-H), 7.53 (1H, d, J 8.1, N–H (Phe)), 7.67 (2H, 2Ј-H
and 6Ј-H); δC (125 MHz, CDCl3) 21.94 (Me (Leu)), 23.06 (Me
(Leu)), 24.83 (C-γ (Leu)), 37.72 (C-β (Phe)), 41.18 (C-β (Leu)),
50.92 (C-α (Leu)), 54.72 (C-α (Phe)), 126.61 (Ph (Phe)), 126.78
(C-2 and C-6), 128.14 (C-3 and C-5), 128.94 (C-2Ј and C-6Ј),
128.34 (Ph (Phe)), 128.37 (C-9), 128.90 (C-3Ј and C-5Ј), 129.56
(Ph (Phe)), 133.99 (C-4Ј), 135.30 (C-1), 136.03 (C-1Ј), 137.47
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962
961

View Article Online
8 C. H. Heathcock, in Comprehensive Organic Synthesis, ed. B. M.
Trost, I. Fleming and C. H. Heathcock, Pergamon Press, New York,
1991.
9 B. M. Kim, S. F. Williams and S. Masamune, in Comprehensive
Organic Synthesis, ed. B. M. Trost, I. Fleming and C. H. Heathcock,
Pergamon Press, New York, 1991.
(Ph (Phe)), 138.14 (C-8), 140.72 (C-4), 166.79 (C-7), 171.18 (CO
(Phe)), 174.38 (COOH), 194.21 (C-10).
m/z (ESI) 513.2 [(M ϩ H)ϩ, 37%], 535.2 ](M ϩ Na)ϩ, 69].
1
14. The yield of 14, determined through H-NMR of the
crude mixture by integrating signals from product and starting
material, was >95%. The product was obtained as the (E) iso-
mer, which rearranged to an E:Z 50:50 mixture of over one
week.
10 I. Paterson, in Comprehensive Organic Synthesis, ed. B. M. Trost,
I. Fleming and C. H. Heathcock, Pergamon Press, New York,
1991.
11 C. C. Leznoff and J. Y. Wong, Can. J. Chem., 1973, 51, 3756.
12 M. J. Kurth, L. A. A. Randall, C. Chen, C. Melander, R. B. Miller,
K. McAllister, G. Reitz, R. Kang, T. Nakatsu and C. Green, J. Org.
Chem., 1994, 59, 5862.
13 M. Reggelin and V. Brenig, Tetrahedron Lett., 1996, 37, 6851.
14 S. Kobayashi, I. Hachiya and M. Yasuda, Tetrahedron Lett., 1996,
37, 5569.
15 A. V. Purandare and S. Natarajan, Tetrahedron Lett., 1997, 38, 8777.
16 S. Kobayashi, T. Wakabayshi and M. Yasuda, J. Org. Chem., 1998,
63, 4868.
17 C. W. Phoon and C. Abell, Tetrahedron Lett., 1998, 39, 2655.
18 C. Gennari, S. Ceccarelli, U. Piarulli, K. Aboutayab, M. Donghi
and I. Paterson, Tetrahedron, 1998, 54, 14999.
19 T. Mukaiyama, K. Banno and K. Narasaka, J. Am. Chem. Soc.,
1974, 96, 7503.
20 R. Noyori, K. Yokoyama, J. Sakata, I. Kuwajima, E. Nakamura and
M. Shimizu, J. Am. Chem. Soc., 1977, 99, 1265.
21 S. Murata, M. Suzuki and R. Noyori, J. Am. Chem. Soc., 1980, 102,
3248.
22 S. E. Denmark, B. D. Griedel, D. M. Coe and M. E. Schnute, J. Am.
Chem. Soc., 1994, 116, 7026.
23 S. G. Nelson, Tetrahedron: Asymmetry, 1998, 9, 357.
24 H. Gröger, E. M. Vogl and M. Shibasaki, Chem. Eur. J., 1998, 4,
1137.
25 S. Saito and H. Yamamoto, Chem. Eur. J., 1999, 5, 1959.
26 N. Yoshikawa, Y. M. A. Yamada, J. Das, H. Sasai and M. Shibasaki,
J. Am. Chem. Soc., 1999, 121, 4168.
27 S. Kobayashi and I. Hachiya, J. Org. Chem., 1994, 59, 3590.
28 S. Kobayashi, Synlett, 1994, 689.
29 C. Le Roux, L. Ciliberti, H. Laurant-Robert, A. Laporterie and
J. Dubac, Synlett, 1998, 1249.
30 M. Meldal, Tetrahedron Lett., 1992, 33, 3077.
31 F.-I. Auzanneau, M. Meldal and K. Bock, J. Pept. Sci., 1995, 1, 31.
32 M. Renil and M. Meldal, Tetrahedron Lett., 1996, 37, 6185.
33 A. Schleyer, M. Meldal, M. Renil, H. Paulsen and K. Bock, Angew.
Chem., Int. Ed. Engl., 1997, 36, 1976.
(E)-isomer: δH (500 MHz; CDCl3) 0.72 (3H, s, Me (Leu)),
0.74 (3H, s, Me (Leu)), 1.40 (1H, m, β-H (Leu)), 1.49 (2H, m, β-
H (Leu) and γ-H (Leu), 1.61 (2H, m, 4Ј-H), 1.79 (2H, m, 5Ј-H),
2.36 (2H, m, 6Ј-H), 2.65 (2H, m, 3Ј-H), 2.99 (1H, m, β-H (Phe)),
3.10 (1H, m, β-H (Phe)), 4.33 (1H, m, α-H (Leu)), 4.74 (1H, m,
α-H (Phe)), 7.02–7.11 (5H, m, Ph (Phe)), 7.10 (1H, d, N–H
(Leu)), 7.23 (2H, d, 3-H and 5-H), 7.26 (1H, s, 8-H), 7.39 (1H,
d, N–H (Phe)), 7.58 (2H, d, 2-H and 6-H); δC (125 MHz;
CDCl3) 22.27 (Me (Leu)), 23.27 (Me (Leu)), 23.70 (C-5Ј), 24.20
(C-4Ј), 25.22 (C-γ (Leu)), 29.41 (C-3Ј), 38.22 (C-β (Phe)), 40.73
(C-6Ј), 41.69 (C-β (Leu)), 51.36 (C-α (Leu)), 54.95 (C-α (Phe)),
127.04 (Ph (Phe)), 127.67 (C-2 and C-6), 128.76 (Ph (Phe)),
129.85 (Ph (Phe)), 130.50 (C-2 and C-6), 134.34 (C-1), 134.48
(C-8), 137.43 (Ph (Phe)), 138.56 (C-2Ј), 139.03 (C-4), 167.04
(C-7), 171.31 (CO (Phe)), 174.69 (COOH (Leu)), 201.78 (C-1Ј).
(Z)-isomer: δH (500 MHz; CDCl3) 0.72 (3H, s, Me (Leu)),
0.74 (3H, s, Me (Leu)), 1.40 (1H, m, β-H (Leu)), 1.49 (2H, m,
β-H (Leu) and γ-H (Leu)), 1.75 (2H, m, 4Ј-H), 1.82 (2H, m,
5Ј-H), 2.38 (2H, m, 6Ј-H), 2.47 (2H, m, 3Ј-H), 2.99 (1H, m, β-H
(Phe)), 3.10 (1H, m, β-H (Phe)), 4.33 (1H, m, α-H (Leu)), 4.74
(1H, m, α-H (Phe)), 6.22 (1H, s, 8-H), 7.02–7.11 (5H, m, Ph
(Phe)), 7.10 (1H, d, N–H (Leu)), 7.15 (2H, d, 3-H and 5-H),
7.33 (1H, d, N–H (Phe)), 7.46 (2H, d, 2-H and 6-H); δC (125
MHz; CDCl3) 22.27 (Me (Leu)), 23.27 (Me (Leu)), 25.22 (C-γ
(Leu)), 26.87 (C-4Ј), 26.96 (C-5Ј), 38.22 (C-β (Phe)), 38.85
(C-3Ј), 41.69 (C-β (Leu)), 44.78 (C-6Ј), 51.36 (C-α (Leu)), 54.95
(C-α (Phe)), 127.04 (Ph (Phe)), 127.46 (C-2 and C-6), 128.76
(Ph (Phe)), 129.05 (C-3 and C-5), 129.85 (Ph (Phe)), 129.87
(C-8), 133.52 (C-1), 137.43 (Ph (Phe)), 139.32 (C-4), 142.84
(C-2Ј), 167.36 (C-7), 171.31 (CO (Phe)), 174.69 (COOH (Leu)),
209.16 (C-1Ј).
34 J. Rademann, M. Meldal and K. Bock, Chem. Eur. J., 1999, 5, 1218.
35 J. Rademann, M. Grøtli, M. Meldal and K. Bock, J. Am. Chem.
Soc., 1999, 121, 5459.
m/z (ESI) 491.6 [(M ϩ H)ϩ, 81%], 981 [(2M ϩ H)ϩ, 16],
1003.1 [(2M ϩ Na)ϩ, 5].
36 O. Mitsunobu, Synthesis, 1981, 1.
37 B. C. Hamper, D. R. Dukesherer and M. S. South, Tetrahedron
Lett., 1996, 37, 3671.
38 V. Krchnak, Z. Flegelova, A. S. Weichsel and M. Lebl, Tetrahedron
Lett., 1995, 36, 6193.
39 F. Wang and J. R. Hauske, Tetrahedron Lett., 1997, 38, 6529.
40 R. Devraj and M. Cushman, J. Org. Chem., 1996, 61, 9368.
41 T. A. Rano and K. T. Chapman, Tetrahedron Lett., 1995, 36,
3789.
42 L. S. Richter and T. R. Gadek, Tetrahedron Lett., 1994, 35, 4705.
43 S. Sarshar, D. Siev and A. M. M. Mjalli, Tetrahedron Lett., 1996, 37,
835.
Acknowledgements
The authors are grateful to Dr Axel Meissner, Carlsberg
Laboratory, for help with acquisition and interpretation of
high-field NMR data, and to Dr Charlotte H. Gotfredsen,
Carlsberg Laboratory, for acquisition of Nanoprobe MAS-
NMR spectra. Also, we thank the Danish National Research
foundation for financial support.
44 J. L. Castro, V. G. Matassa and R. G. Ball, J. Org. Chem., 1994, 59,
2289.
References
45 E. E. Swayze, Tetrahedron Lett., 1997, 38, 8465.
46 E. E. Swayze, Tetrahedron Lett., 1997, 38, 8643.
47 M. R. Pavia, M. P. Cohen, G. J. Dilley, G. R. Dubuc, T. L. Durgin,
F. W. Forman, M. E. Hediger, G. Milot, T. S. Powers, I. Sucholeiki,
S. Zhou and D. G. Hangauer, Bioorg. Med. Chem., 1996, 4, 659.
48 E. Hasegawa, K. Ishiyama, T. Horaguchi and T. Shimizu, J. Org.
Chem., 1991, 56, 1631.
49 N. Giuseppone, P. Van de Weghe, M. Mellah and J. Collin,
Tetrahedron, 1998, 54, 13129.
50 R. Lin, Y. Yu and Y. Zhang, Synth. Commun., 1993, 23, 271.
51 K. Hantawong and W. S. Murphy, J. Chem. Res. (M), 1988, 10,
2520.
1 Combinatorial Peptide and Non-Peptide Libraries: A Handbook, ed.
G. Jong, VCH, Weinheim, New York, Basel, Cambridge, Tokyo,
1996.
2 A. F. Spatola, in Chemistry and Biochemistry of Amino Acids,
Peptides and Proteins, ed. B. Weinstein, Marcel Dekker, New York,
1983.
3 D. H. Rich, in Comprehensive Medicinal Chemistry, ed. C. Hansch,
P. G. Sammes and J. B. Tailor, Pergamon Press, Oxford, 1989.
4 M. A. Gallop, R. W. Barret, W. J. Dower, S. P. A. Fodor and
E. M. Gordon, J. Med. Chem., 1994, 37, 1233.
5 E. M. Gordon, R. W. Barret, W. J. Dower, S. P. A. Fodor and
M. A. Gallop, J. Med. Chem., 1994, 37, 1385.
52 E. Kaiser, Anal. Biochem., 1970, 34, 595.
6 D. C. Horwell, Bioorg. Med. Chem., 1996, 4, 1573.
7 J. P. Vacca and J. H. Condra, Drug Discovery, 1997, 2, 261.
Paper a909246c
962
J. Chem. Soc., Perkin Trans. 1, 2000, 955–962