Esterolytic peptide dendrimers with a hydrophobic core and catalytic residues at the surface

Article abstract of DOI:10.1002/adsc.200404070

(3S)-4-(9-Fluorenylmethoxycarbonylamino)-3-methyl(allyloxycarbonyl) aminoethyloxyacetic acid (1) was prepared from (R)-3-aminopropane-1,2-diol and used as branching unit for the synthesis of second generation peptide dendrimers with six individually addressable variable amino acid positions. Three pairs of diastereomeric dendrimers were prepared bearing a common hydrophobic core and permutations of the catalytic triad amino acids aspartate, histidine and serine at the surface. Dendrimers with two surface histidine residues catalyzed the hydrolysis of fluorogenic 8-acyloxypyrene-1,3,6-trisulfonates in aqueous buffer pH 6.0 with rate enhancement k_cat/k_uncat in the 10 ³ range and Michaelis-Menten constants K_M in the 10 ^-4 M range. Substrate recognition involves electrostatic interactions, as shown by competitive inhibition of catalysis observed with pyrene-1,3,6,8-tetrasulfonate. The 4-fold to 7-fold lowering in K_M between the butyryl and nonanoyl esters in the most active dendrimers provides evidence for a hydrophobic component in substrate binding, which is absent in a closely related, less active diastereomeric peptide dendrimer.

Full text of DOI:10.1002/adsc.200404070

FULL PAPERS
Esterolytic Peptide Dendrimers with a Hydrophobic Core and
Catalytic Residues at the Surface
Anthony Clouet, Tamis Darbre, Jean-Louis Reymond*
University of Berne, Freiestrasse 3, CH-3012 Berne, Switzerland
Fax: (þ41)-31-631-8057, e-mail: jean-louis.reymond@ioc.unibe.ch
Received: March 8, 2004; Accepted: July 19, 2004
Supporting Information for this article is available on the WWW under http://asc.wiley-vch.de or from the author.
Abstract: (3S)-4-(9-Fluorenylmethoxycarbonylami- in the 103 range and Michaelis-Menten constants KM
no)-3-methyl(allyloxycarbonyl)aminoethyloxyacetic
in the 10À4 M range. Substrate recognition involves
acid (1) was prepared from (R)-3-aminopropane-1,2- electrostatic interactions, as shown by competitive in-
diol and used as branching unit for the synthesis of hibition of catalysis observed with pyrene-1,3,6,8-tet-
second generation peptide dendrimers with six indi- rasulfonate. The 4-fold to 7-fold lowering in KM be-
vidually addressable variable amino acid positions. tween the butyryl and nonanoyl esters in the most ac-
Three pairs of diastereomeric dendrimers were pre- tive dendrimers provides evidence for a hydrophobic
pared bearing a common hydrophobic core and per- component in substrate binding, which is absent in a
mutations of the catalytic triad amino acids aspartate, closely related, less active diastereomeric peptide
histidine and serine at the surface. Dendrimers with dendrimer.
two surface histidine residues catalyzed the hydrolysis
of fluorogenic 8-acyloxypyrene-1,3,6-trisulfonates in Keywords: dendrimers; enzyme models; ester hydrol-
aqueous buffer pH6.0 with rate enhancement kcat/kuncat ysis; peptides, solid-phase synthesis
Introduction
tween amino acids within the peptide dendrimer struc-
ture as a protein model. Peptide dendrimers are formed
Enzyme catalysis results from cooperative effects be- by alternating functional amino acids with a branching
tween amino acid side chains and cofactors that are diamino acid and can be prepared by standard solid
made possible through the ordered folding of linear pep- phase peptide synthesis. Esterolytic dendrimers were
tides.[1] In de novoenzyme design one attempts to use the obtained by combining the catalytic triad amino acids
knowledge of enzyme structure and function to build ar- aspartate, histidine and serine within the dendrimer
tificial enzymes from first principles.[2] Catalytic anti- structure.[9]
bodies[3] and catalytic peptides[4–7] have provided con-
Herein we report the preparation of catalytic peptide
vincing examples that catalysis on a protein basis is not dendrimers based on an orthogonally protected diamino
restricted to natural enzymes. The field of organic catal- acid branching unit 1. This strategy allows one to pre-
ysis has shown that efficient and selective catalytic cycles pare dendrimers with differently functionalized branch-
can be designed with small molecule catalysts of a much es, and was used to assemble different permutations of
lower complexity level than enzymes, such as simple catalytic amino acids at the dendrimer surface around
amino acids, suggesting that most of the protein struc- a common hydrophobic core. This arrangement follows
ture can, in fact, be disposed of for the pure purpose of the overall structural design of globular proteins where
chemistry.[8]
hydrophobic amino acids are usually found at the core,
Recently, we have started to explore de novo enzyme while catalytic sites are formed by pockets close to the
design on the basis of peptide dendrimers.[9] Dendrim- protein surface. Oursynthetic approach has not been ap-
ers are tree-like structures that adopt a globular or plied before to dendrimer synthesis and the dendrimers
disk-shape structure as a consequence of topology rath- reported here are among the very few examples of den-
er than folding.[10] The dendrimer strategy circumvents drimers with differentially functionalized layers.[12] The
the protein folding problem. While catalytic dendrim- peptide dendrimers obtained catalyze the hydrolysis of
ers have been assembled from pre-existing catalytic pyrene-trisulfonate esters with enzyme-like properties
subunits such as metal complexes and cofactors,[11] we of substrate binding and turnover in aqueous environ-
are investigating catalysis arising from the interplay be- ment.
Adv. Synth. Catal. 2004, 346, 1195–1204
DOI: 10.1002/adsc.200404070
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
1195

Anthony Clouet et al.
FULL PAPERS
Results and Discussion
Synthesis
Our aim was to prepare catalytic peptide dendrimers
with differentially functionalized layers on the basis of
a selective protection scheme. Our peptide dendrimers
are assembled as disulfide-bridged dimers of two sec-
ond-generation peptide dendrimers containing three
variable amino acids present in one, two, and four cop-
ies, separated by a common diamino acid branching
unit (Scheme 1). We envisioned to synthesize the mono-
meric peptide dendrimers following the established sol-
id-phase synthesis route, but employing a differentially
protected diamino acid building block 1 at the first layer
and at one of the second layer branching units, which
would enable us to differentiate six individual positions,
as opposed to only three when using only the symmetri-
cally protected diamino acid building block 2.
The realization of this strategy required the synthesis
of adiamino acid with two orthogonal protecting groups.
One amino group would be protected by the standard
Fmoc (¼9-fluorenylmethoxycarbonyl) protecting
group for immediate functionalization, while the second
would be blocked by the Alloc (¼allyloxycarbonyl)
protecting group, which is resistant to piperidine and
can be removed selectively under neutral conditions
with Pd(PPh3)4 and phenyltrihydrosilane (PhSiH3).[13]
The chiral branching diamino acid 1 was prepared in
eight steps from commercially available (R)-N-tert-bu-
tyloxycarbonyl-3-aminopropane-1,2-diol 3, in an overall
yield of 23% (Scheme 2). Thus, protection of the amino
group in 3 with Boc2O gave 4. The diol 4 was converted
to epoxide 5 by treatment with methanesulfonyl chlor-
ide in pyridine followed by NaOHin aqueous DMSO
(74%).[14] Chiral GC analysis gave an enantiomeric ex-
cess of 80% ee for epoxide 5. The partial racemization
was probably caused by the mesylation step due to an in-
complete selectivity for the primary alcohol. Nucleo-
philic epoxide opening with ammonia gave amino alco-
hol 6. Protection of the free amino group with allyl chlor-
oformate then gave the Boc-Alloc intermediate 7, which
was alkylated at the secondary alcohol with ethyl bro-
moacetate to yield 8. Ester hydrolysis with LiOHin
aqueous THF was followed by treatment with aqueous
HCl to remove the Boc protecting group, and finally re-
action of the free amine with FmocCl in aqueous diox-
ane gave the orthogonally protected branching amino
acid 1. The preparation of the corresponding symmetri-
cal bis-Fmoc protected (2-amino-1-aminomethyl-
Scheme 1. Structure of disulfide-bridged peptide dendrimers
with differential functionalization of layers. (A1 –A6, A’1 –
A’6 ¼variable amino acids, C¼cysteine).
Scheme 2. Synthesis of chiral branching unit 1. Conditions: a)
Boc2O, Et3N, CH2Cl2/MeOH, 2 h, 258C, 94%; b) MsCl, Pyr,
08C, then NaOH, DMSO, 08C, 74%; c) NH4OH, EtOH,
1.5 h, quant.; d) allyl chloroformate, EtOAc, NaHCO3, 4 h,
87%; e) BrCH2CO2Et, NaH, dry THF, 5 h, 63%; f) i)
LiOH, THF/H2O, 25 8C, overnight, ii) aqueous 3 N HCl,
608C, 3 h, iii) Fmoc-Cl, NaHCO3, H2O/dioxane, 258C, over-
night, 63%. Boc¼tert-butoxycarbonyl, THF¼tetrahydrofur-
an, Fmoc-Cl¼9-fluorenylmethoxycarbonyl chloride, MsCl¼
methanesulfonyl chloride.
ethoxy)-acetic acid branching unit 2 has been described standard coupling methodology with BOP as in situ cou-
previously.[9]
pling reagent.[16] Fmoc deprotection then allowed us to
The peptide dendrimers were synthesized on a Rink attach amino acid A2, followed by the branching unit
amide resin (Scheme 3).[15] The synthetic sequence start- 2, and two copies of amino acid A3. The first branch
ed with sequential attachment of the first amino acid A1, was completed by capping with acetic anhydride. The
cysteine, and the first branching unit 1 following the Alloc protecting group at the level of the first branching
1196
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1195–1204

Esterolytic Peptide Dendrimers with a Hydrophobic Core
FULL PAPERS
Preparation of Esterase Dendrimers with
Hydrophobic Core
The dendrimer synthesis was used to prepare a family of
peptide dendrimers bearing a hydrophobic core at posi-
tions A1, A2 and A4 and catalytic amino acids at the sur-
face positions A3, A5 and A6. All dendrimers were pre-
pared with a conserved core consisting of phenylalanine
and leucine at position A2 and A4. A histidine residue
was placed at position A1 to provide a positive charge
in acidic pHand facilitate purification of the monomer
by RP-HPLC, as well as to facilitate disulfide bridge for-
mation. Indeed, we have previously experienced that di-
sulfide bridge formation is very slow and sometimes im-
possible with dendrimers bearing hydrophobic residues
at position A1.[9b] Histidine has both polar and aromatic
character. As will be seen below, histidine at position A1
is not sufficient to induce catalysis of the hydrolytic reac-
tions studied. The surface positions A3, A5 and A6 were
used to install permutations of the catalytic triad amino
acids aspartate, histidine and serine, with the aim of
mimicking a hydrolase-type enzyme. Since exchanging
amino acids at position A5 and A6 gives diastereoisomer-
ic products at the asymmetric branching unit 1 posi-
tioned in the second layer, the planned synthesis result-
ed in three pairs of diastereoisomeric peptide dendrim-
ers A/iA (two serines at A3), B/iB (two aspartates at A3),
and C/iC (two histidines at A3). The dendrimers were
obtained in yields of 20–30% after purification by
semi-preparative HPLC, which is very satisfactory for
a total of nine sequential peptide coupling steps, show-
ing that the use of orthogonal protection with Alloc
was very efficient. The monomeric dendrimers were
then dimerized using aldrithiol[17] to give 21 combina-
tions of disulfide-bridged dimeric dendrimers (A-A to
iC-iC), which were isolated with 30–75% yields after
purification by semi-preparative RP-HPLC. All den-
drimers were purified by semi-preparative RP-HPLC
and characterized by ESI MS.
Scheme 3. Solid phase synthesis of peptide dendrimers. Con-
ditions: Deprotecting steps: i¼DMF, piperidine (4:1); ii¼
PhSiH3 (25 equivs.), CH2Cl2, 3 min, then Pd(PPh3)4 (0.1
equiv.). Coupling steps: iii¼BOP/DIEA. Capping steps:
iv¼Ac2O/CH2Cl2 (1:1). 1) i, then iii with FmocA1OH; 2) i,
then iii with FmocCys(Trt)OH; 3) i, then iii with 1; 4) i,
then iii with FmocA2OH; 5) i, then iii with 2; 6) i, then iii
with FmocA3OH; 7) i, then iv; 8) ii, then iii with FmocA4OH;
9) i, then iii with 1; 10) i, then iii with FmocA5OH; 11) i, then
iv; 12) ii, then iii with FmocA6OH; 13) i, then iv; 14) TFA
cleavage and HPLC purification; 15) disulfide coupling with
2,2’-dithiopyridine.
Esterolytic Activity of Peptide Dendrimers
We expected that an esterolytic activity might arise from
a cooperative effect between the catalytic triad amino
acids aspartate, histidine and serine at positions A3, A5
and A6 at the surface of the dendrimers. The hydropho-
bic core amino acids at positions A2 and A4 might assist
unit was removed by treatment with tetrakis(triphenyl- catalysis by providing enhanced substrate binding
phosphine)palladium(0) and phenylsilane.[13] The through hydrophobic interactions. The dendrimers (6
fourth amino acid A4 was introduced, followed again monomers and 21 dimers) were assayed for esterolytic
by the asymmetric branching unit 1. Fmoc deprotection activity with various fluorogenic esters. The substrates
then allowed us to couple a single copy of an amino acid (200 mM) were tested for hydrolysis in aqueous buffer
at position A5, which was acetylated. Finally, removal of at pH6.0 in the presence of 2.5 mol % (5 mM) dendrim-
the second Alloc protection as above and peptide cou- er catalyst in a microtiter-plate setup. There was no ac-
pling allowed us to install the last amino acid A6, which tivity with esters of 7-hydroxy-N-methylquinolinium
was acetylated.
salts, even though these were active with our previous
Adv. Synth. Catal. 2004, 346, 1195–1204
asc.wiley-vch.de
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
1197

Anthony Clouet et al.
FULL PAPERS
Scheme 4. The hydrolysis of pyrene 1,3,5-trisulfonate esters is
catalyzed by hydrophobic core peptide dendrimers with sur-
face catalytic residues.
Figure 1. Time-course graph for the hydrolysis of 10 to form
~
*
13 in the absence ( ) or presence of dendrimers iA-iC ( ),
*
or iA-C ( ).Conditions: 200 mM substrate 10, 5 mM dendrim-
Table 1. Hydrolysis of pyrenesulfonate ester 10 in the pre-
sence of peptide dendrimers. The sequence of the amino
acids on the surface are given in parenthesis.[a]
er, 268C, in 20 mM aqueous bis-Tris buffer pH6.0. The reac-
tion was followed using a SpectraMAX fluorescence plate
reader (lexc ¼460 nm, lem ¼530 nm). Fluorescence was con-
verted to product concentration using a calibration curve,
which was linear in the concentration range used.
Vnet/Vuncat
–
A
iA
B
iB
C
iC
A(SSHD)
4.0
3.5
1.5
0.5
11.0
6.5
4.5
2.5
1.0
1.5
3.0
3.0
/
/
/
1.5
0
3.0
3.5
/
/
/
1.0
5.5
5.0
/
/
/
/
/
/
/
/
/
iA (SSDH)
B (DDSH)
iB (DDHS)
C (HHSD)
iC (HHDS)
3.5
0.5
0.5
8.0
9.0
16.0
6.5
10.0
[a]
Conditions: 200 mM 10, 5 mM dendrimer, 258C, 20 mM
aqueous bis-Tris pH6.0. The ratio Vnet/Vuncat is reported
for monomeric dendrimers(1st column) and all combina-
tions of dimers. Vuncat is the apparent spontaneous rate of
hydrolysis of the butyrate ester 10 in buffer alone, and
Vnet ¼Vapp –Vuncat, where Vapp is the apparent rate of hydro-
lysis of 10 in the presence of dendrimer. The reactions
were run in 96-well polystyrene half area microtiter plates
and followed using a SpectraMAX fluorescence detector
with lexc ¼460 nm, lem ¼530 nm. Fluorescence was conver-
ted to product concentration using a calibration curve,
which was linear in the concentration range used.
series of peptide dendrimers,[9] or with aliphatic esters
and acyloxymethyl ethers of umbelliferone, which are
substrates suitable for esterolytic catalytic antibodies
and enzymes.[18] We next turned out attention to the acy-
loxypyrene-1,3,6-trisulfonates 9–12 (Scheme 4). In this
case, several dendrimers showed significant rate en-
hancements over the uncatalyzed background reaction.
Figure 2. Double reciprocal plot for hydrolysis of ester 10 cat-
*
alyzed by peptide dendrimers iC-iC (HHSD-HHSD) ( ) and
~
iA-iC (DDHS-HHDS) ( ).
The best apparent catalytic effects with peptide den- also observed with the diastereoisomeric pair of hetero-
drimers were observed with butyrate 10 and the two di- dimers iAC (Vnet/Vuncat ¼8) and iAiC (Vnet/Vuncat ¼9), as
astereoisomeric homodimers bearing two surface histi- well as with the monomeric dendrimers C (Vnet/Vuncat
¼
dine residues CC and iCiC, which showed apparent cat- 11). Similar results were obtained with esters 9 and 11
alytic effects of Vnet/Vuncat ¼16 and Vnet/Vuncat ¼10, re- as substrates. By contrast, there was no observable catal-
spectively (Table 1). Catalysis was confirmed by the ob- ysis with the dodecanoyloxypyrene derivative 12.
servation of multiple turnovers (Figure 2). Catalysis of
The kinetics of catalysis were characterized in detail
butyryloxypyrene-1,3,6-trisulfonate 10 hydrolysis was for the five most reactive dendrimers (Table 2, Fig-
1198
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1195–1204

Esterolytic Peptide Dendrimers with a Hydrophobic Core
FULL PAPERS
Table 2. Michaelis–Menten parameters for the most active dendrimers on pyrene substrates 9, 10 and 11.
9
10
11
4-MeIm
k2 (mMÀ1 minÀ1
)
1.1ꢀ10À3
8.1ꢀ10À4
8.2ꢀ10À4
kuncat (minÀ1
)
8.5ꢀ10À5
4.4ꢀ10À5
4.9ꢀ10À5
C
KM (mM)
0.18
0.054
640
280
5.5
0.19
0.046
1100
300
0.046
0.019
400
500
5.6
(HHSD)
kcat (minÀ1
kcat//kuncat
)
)
)
)
)
kcat//KM/k2
v
net/vuncat
11
C-C
KM (mM)
kcat (minÀ1
0.11
0.044
520
380
7.6
0.12
0.045
1000
470
0.017
0.022
450
1600
11
(HHSD-HHSD)
k
cat//kuncat
kcat//KM/k2
vnet/vuncat
16
iC-iC
(HHDS-HHSD)
KM (mM)
kcat (minÀ1
kcat//kuncat
kcat//KM/k2
vnet/vuncat
0.11
0.031
370
260
4.2
0.20
0.037
840
230
10.0
0.10
0.033
680
390
4.4
iA-C
(SSDH-HHSD)
KM (mM)
kcat (minÀ1
kcat//kuncat
kcat//KM/k2
vnet/vuncat
0.37
0.050
1200
180
8.0
iA-iC
(SSDH-HHSD)
KM (mM)
kcat (minÀ1
kcat//kuncat
kcat//KM/k2
vnet/vuncat
0.40
0.050
1100
160
9.0
Conditions and measurement method: 20 mM aqueous bis-Tris pH6.0, 5 mM dendrimer and 50–800 mM substrate at 258C.
The kinetic constants given are derived from the linear double-reciprocal plots of 1/vnet vs. 1/S, with r2 >0.96. The apparent
rate enhancement Vnet/Vuncat observed with S¼200 mM and 5 mM dendrimer. Vuncat is the hydrolysis rate of 9–11 in buffer alone,
and Vnet ¼Vapp –Vuncat, where Vapp is the apparent hydrolysis rate of 9–11 in the presence of dendrimer.
ure 2). Catalysis followed the Michaelis–Menten model, to that of the butyrate 10. Interestingly, the KM values
with substrate binding in the range of KM ¼10À4 M and with these dendrimers also dropped significantly be-
rate acceleration in the range of kcat/kuncat ¼103, which tween butyrate 10 and nonanoate 11. The longer aliphat-
is comparable to our previously reported catalytic pep- ic acid chain in this substrate induced a 4-fold lowering
tide dendrimers for similar ester hydrolysis reactions. of KM with C and a 7-fold lowering of KM with CC. By
Catalysis was proportional to dendrimer concentration contrast, the chain lengthening had no effect on either
up to 50 mM dendrimer, suggesting that dendrimers do kcat of KM with the diastereoisomeric dendrimer iCiC,
not aggregate in the buffer at pH6. Butyryl ester 10 which showed comparable kinetic parameter with all
was the best substrate in terms of rate acceleration kcat/ three substrates tested. The dependence of KM on the
kuncat, with the best value being kcat/kuncat ¼1200 in the aliphatic chain length in the case of C and CC suggests
case of dendrimer iAC. The catalytic rate constants kcat that these dendrimers achieve substrate binding partly
and KM were similar for acetate 9 and butyrate 10 in through hydrophobic interactions with the aliphatic
all cases, and the higher rate acceleration with the buty- chain, most likely via the hydrophobic core residues
rate was due to the fact that the uncatalyzed reaction of Phe and Leu. The effect, however, must depend on pre-
butyrate 10 was only half that of acetate 9.
cise molecular interactions since it is completely absent
In the case of dendrimers C and CC, the catalytic rate in the diastereoisomeric dendrimer iCiC, whose mono-
constant kcat decreased two-fold between the butyrate meric form iC is also less active than its stereoisomer C.
ester 10 and the nonanoyl ester 11, resulting in a corre-
Catalytic activity depends on the presence of two sur-
spondingly lower rate acceleration kcat/kuncat since the face histidine residues. The fact that several dendrimers
uncatalyzed reaction of the nonanoate 11 was similar are not active at all is evidence that the core histidine res-
Adv. Synth. Catal. 2004, 346, 1195–1204
asc.wiley-vch.de
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
1199

Anthony Clouet et al.
FULL PAPERS
idue at position A1 is not capable of inducing catalysis
alone. The pair of surface histidine residues at position
A3 probably catalyzes the reaction by nucleophilic catal-
ysis. Pyrene-1,3,6-trisulfonate esters 9–11 have been re-
ported to be hydrolyzed in aqueous conditions by self-
assembled peptide-based pores functionalized with
multiple histidine residues.[19] The hydrolysis of these es-
ters is also catalyzed by 4-methylimidazole in aqueous
buffer at pH6.0. The catalytic peptide dendrimers are
between 260-fold and 1600-fold more reactive that 4-
methylimidazole as measured by the ratio of the specif-
icity constant kcat/KM to the second order rate constant
for 4-methylimidazole catalysis k2. This rate accelera-
tion corresponds to between 60-fold and 400-fold rate
acceleration per histidine residue if one assumes that
each surface histidine residue acts independently.
Part of the reactivity enhancement observed for histi-
dine side-chains within the dendrimers over the reaction
of 4-methylimidazole in solution is explained by sub-
strate binding, which is also taken into account for the
calculation of the specificity constant kcat/KM. Since the
imidazole side chains are partly protonated under the
conditions of the assays, the histidine residues should
participate in recognition of the anionic pyrene-1,3,6-
trisulfonate group. A requirement for electrostatic bind-
ing can also explain the lack of catalysis with neutral and
cationic fluorogenic substrates despite the fact that
these are more reactive than the pyrene-trisulfonate es-
ters. Further evidence for an electrostatic component in
substrate binding was gained by investigating inhibition
of catalysis by 1,3,6,8-pyrene-tetrasulfonate 14
(Scheme 4, Figure 3). This tetra-anionic ligand compet-
itively inhibited dendrimer catalysis with Ki ¼60 mM for
C, Ki ¼20 mM for CC. This competitive inhibition is best
interpreted in terms of electrostatic interactions be-
tween the protonated histidine side chains and the sulfo-
nate groups. Inhibition is, however, too weak to reflect
transition state analogue type inhibition since the transi-
tion state dissociation constants for dendrimer catalysis
lie in the range of KTS ¼KM/(kcat/kuncat)¼0.4–0.04 mM,
which corresponds to a 100-fold to a 1000-fold stronger
binding of the transition state compared to tetrasulfo-
nate 14.
Figure 3. Dixon plot for competitive inhibition of dendrimer-
catalyzed hydrolysis of acetate ester 9 by 1,3,6,8-pyrene-tetra-
sulfonate 14. Measured in 20 mM bis-Tris pH6.0, 0.2 mM 9,
*
and dendrimer C ( ), and dendrimer CC (D). The competi-
tive inhibition constant is the x-coordinate of the intercept
between the fitted lines and the horizontal lines drawn at
y¼1/Vmax
.
Figure 4. Hydrolysis of 8-nonanoyloxypyrene-1,3,6-trisulfo-
nate 10 by catalytic peptide dendrimer CC involves electro-
static and hydrophobic substrate binding and nucleophilic
catalysis by histidine side-chains.
On the basis of the experiments above one can pro-
pose a catalytic mechanism for the most active dendrim- Conclusions
er CC (Figure 4). Substrate binding is mediated by elec-
trostatic interactions between the sulfonate groups and The orthogonally protected diamino acid branching unit
the protonated histidine side chains of the dendrimer, 1 was used to assemble peptide dendrimers with ahydro-
and further increased by a hydrophobic interaction be- phobic core and differentiated surface catalytic residues.
tween the aliphatic chain of the acyl group and the hy- The preparation of a small 27-member library allowed
drophobic core residues Phe and Leu in the case of non- us to observe esterolytic activity for the hydrolysis of
anoate 11. Hydrolysis proceeds by acyl group transfer to 8-acyloxypyrene-1,3,6-trisulfonate esters. Substrate rec-
one of the histidine side chains and hydrolysis to regen- ognition can be understood in terms of electrostatic in-
erate the catalyst. The 8-hydroxypyrene-1,3,6-trisulfo- teractions between protonated histidines and the nega-
nate leaving group 13 diffuses away, and could also tively charged sulfonate groups. In addition, the most ac-
show a weak product inhibition analogous to that ob- tive dendrimer CC showed enhanced substrate binding
served with the tetrasulfonate 14.
with long chain aliphatic acids, which may be explained
1200
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1195–1204

Esterolytic Peptide Dendrimers with a Hydrophobic Core
FULL PAPERS
which was followed by TLC, was complete in 2 h. After solvent
evaporation, the oily residue was chromatographed (hexane/
ethyl acetate 2/8) to give 4 as a white solid; yield: 3.97 g
(95%); Rf (ethyl acetate): 0.5; mp 498C; IR (neat): n¼3019 w,
1653 w, 1393 w, 1369 w, 1215 m, 1045 cmÀ1 w; 1HNMR
(CDCl3): d¼4.92 (s, 1H), 3.75 (m,1H), 3.59 (d, J¼5.15 Hz,
2H), 3.28 (broad s, 2H), 1.45 (s, 9H); 13C NMR (CDCl3): d¼
157.27, 80.07, 71.29, 63.68, 42.93, 28.37; HR-ESI-MS: calcd.
for C8H17NO4: 191.1158; found: 191.1055; [a]2D5: À10.0 (c
6.15 g mLÀ1 in CH2Cl2).
by hydrophobic interactions with the dendritic core. The
catalysis itself was triggered by the presence of at least
two surface histidine residues. Using a hydrophobic
core for binding together with surface catalytic residues
might provide a general strategy for the design of cata-
lytic peptide dendrimers.
The efficiency of the dendrimer synthesis is notewor-
thy. The synthetic yields of the purified dendrimers are
similar to those for linear peptides of similar length.
One of the key advantages of peptide dendrimers as en-
zyme models is their synthetic availability including a
potential for structural variations through variations of
the amino acid building blocks. The synthetic strategy
exposed here allows six individually variable positions
and defines 64 million monomers and 2·1015 dimers us-
ing the twenty proteinogenic amino acids. A combinato-
rial protocol should allow us to explore a much broader
structural diversity of peptide dendrimers and to test
their potential for selective catalysis.[9d]
(2S)-N-tert-Butoxycarbonyl-1,2-
epoxyethylaminopropane (5)
Methanesulfonyl chloride (46 mL, 0.577 mmol) was added to a
stirred solution of 4 (100 mg; 0.525 mmol) in pyridine
(0.645 ml) at 08C. After 10 min, the resulting solution was add-
ed over 5 min to a stirred solution of sodium hydroxide (64 mg;
1.6 mmol) in water (0.65 ml) plus dimethyl sulfoxide (0.43 mL)
at 08C. After an additional 10 min of stirring, the solution was
poured into ice-water and the mixture extracted with ether.
The ether phase was washed with water and brine, dried over
Na2SO4 and evaporated. The residue was chromatographed
(hexane/ethyl acetate 8/2) and the product 5 was obtained as
a white solid; yield: 67.2 mg (74%); Rf (ethyl acetate/hexane
4/6): 0.65; mp 76–778C; IR (neat): n¼3019 m, 2981 w,
2931 w, 1713 m, 1698 m, 1511 m, 1505 m, 1393 w, 1369 w,
1215 cmÀ1 s; 1HNMR (CDCl 3): d¼4.81 (s, 1H, NH), 3.50 (m,
1H), 3.15–3.25 (2dd, J¼6.25, 5.15, 6.2, 5.1 Hz, 1H), 3.08 (m,
1H), 2.77 (dd, J¼4.2, 4.2 Hz, 1H), 2.58 (dd, J¼4.8, 2.6 Hz,
1H), 1.42 (s, 9H); 13C NMR (CDCl3): d¼155.92, 79.58, 50.83,
45.00, 41.62, 28.30; HR-ESI-MS: calcd. for C8H15NO3:
173.1052; found: 173.0949; [a]2D5: À2.0 (c 7.05 g mlÀ1 in CH2
Cl2). Chiral GC analysis of epoxide 5[20] using a 25 m CST Rest-
ek, showed an enantiomeric ratio of 90 (S) to 10 (R) (80% ee).
Experimental Section
All reagents were either purchased from Aldrich or Fluka
Chemica (Switzerland). Amino acids and their derivatives
were purchased from Senn Chemicals or Novabiochem (Swit-
zerland). Rink amide resin was purchased from Novabiochem
(Switzerland). All solvents used were analytical grade. Peptide
syntheses were performed manually in a glass reactor. Analyt-
ical RP-HPLC was performed in a Waters (996 photo diode ar-
ray detector) chromatography system using a chromolith per-
formance RP-18e, 4.6ꢀ100 mm, flow rate 3 mL minÀ1 col-
umn. Compounds were detected by UV absorption at 214
and 234 nm. Preparative RP-HPLC was performed with
HPLC-grade acetonitrile and MilliQ deionized water in a Wa-
ters prepak cartridge 500 g (RP-C18 20 mm, 300 Š pore size)
installed on a Waters Prep LC4000 system from Millipore
(flow rate 100 mL minÀ1, gradient 1% minÀ1 CH3CN). Semi-
preparative RP-HPLC (flow rate, 4 mL/min; eluent A, water
and 0.1% TFA; eluent B, acetonitrile, water and TFA, 3/2/
0.1%); column, Vydac 218 TP (1.0 cmꢀ25 cm, 300 Š pore
size). Compounds were detected by UV absorption at
220 nm. Chromatography (flash) performed with Merck silica
gel 60 (0.040ꢁ0.063 mm), TLC with fluorescent F254 glass
plates. Fluorescence measurements were carried out with a
spectraMAX fluorescence detector. NMR spectra recorded
on Bruker AC-300 (1H, 300 MHz; 13C, 75 MHz, d in ppm), in-
frared spectroscopy with a Perkin-Elmer 1600 series FTIR [fre-
quencies (n) given in cmÀ1]. Optical rotations measured with a
Perkin-Elmer 241 digital polarimeter with a 1 dm cell. Melting
points were determined on a Büchi 510 apparatus and are not
corrected. MS was provided by Dr. Thomas Schneeberger
(University of Bern, Switzerland).
(2S)-N-tert-Butoxycarbonyl-2-hydroxy-1,3-
diaminopropane (6)
The epoxide 5 was mixed with a solution of 25% aqueous am-
monia (5.2 mL) and ethanol (2.1 mL) at room temperature.
The mixture was stirred for 90 min and the solvent evaporated.
The product 6 was obtained quantitatively as a white solid; mp
458C; IR (neat): n¼3022 s, 2981 w, 2931 w, 1712 m, 1698 m, 1511
1
m, 1505 m, 1393 w, 1369 w, 1227 m, 1205 cmÀ1 m; HNMR
(CDCl3): d¼5.28 (s, 1H, NH), 3.61 (m, 1H), 3.22 (m, 1H),
3.07 (m, 1H), 2.79 (NH2), 2.75 (m, 1H), 2.61 (m, 1H), 1.42 (s,
9H); 13C NMR (CDCl3): d¼156.52, 79.29, 70.61, 44.27, 43.91,
28.19; HR-ESI-MS: calcd. for C8H18N2O3Na: 213.1215; found:
213.1231; [a]2D5: þ10.0 (c 7.75 g mLÀ1 in CH2Cl2).
(2S)-3-N-Allyloxycarbonyl-N-tert-butoxycarbonyl-2-
hydroxy-1,3-diaminopropane (7)
(2R)-N-tert-Butoxycarbonylaminopropane-1,2-diol (4)
To a stirred solution of 6 (1.9 g; 10 mmol) in ethyl acetate
(10 mL) and saturated aqueous NaHCO3 (10 mL) was added
allyl chloroformate (1.70 mL, 16 mmol). The mixture was stir-
red during 4 h, then washed with water, saturated aqueous
NaHCO3, brine, dried over Na2SO4 and the solvents evaporat-
Di-tert-butyl dicarbonate (5.75 g, 26.34 mmol) in CH2Cl2
(7 ml), was added slowly to a stirred solution of (R)-3-amino-
propane-1,2-diol (3; 2 g, 21.95 mmol) in CH2Cl2-MeOH(1:5,
20 mL) and triethylamine (0.36 mL, 2.6 mmol). The reaction,
Adv. Synth. Catal. 2004, 346, 1195–1204
asc.wiley-vch.de
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
1201

Anthony Clouet et al.
FULL PAPERS
ed. The residue was chromatographed (hexane/ethyl acetate 6/
4) to give product 7 as a colorless oil; yield: 2.37 g (87%); Rf
(ethyl acetate): 0.73; IR (neat): n¼3018 s, 2980 w, 2935 w,
1720 m, 1716 m, 1698 m, 1524 m, 1515 m, 1393 w, 1368 w,
127.07, 125.09, 119,99, 117.86, 78.65, 67.01, 66.83, 65.91, 47.16,
41.10, 40.97; HR-LSI-MS: calcd. for C24H26N2O7 þHþ:
455.181827; found: 455.181090; [a]2D5: 0.0 (c 9.1 g mLÀ1 in
CH2Cl2).
1
1215 m, 1117 cmÀ1 w; HNMR (CDCl 3): d¼5.93 (m, 1H),
5.57 (broad s, 1H, NH), 5.30 (dd, J¼1.5, 17.1 Hz, 1H), 5.20
(overlap. dd and s, 2H), 4.56 (d, J¼5.5 Hz, 2H), 3.77 (m, 1H),
3.23 (m, 4H), 1.43 (s, 9H); 13C NMR (CDCl3): d¼157.07,
132.50, 117.59, 79.71, 70.43, 65.62, 43.60, 43.29, 28.14; HR-
ESI-MS: calcd. for C12H22N2O5: 274.1529; found: 274.1606;
[a]2D5: À3.0 (c 5.7 g mLÀ1 in CH2Cl2).
Synthesis of the Dendrimers by SPPS
Coupling of the Fmoc-protected amino acids: The resin was
washed and swelled inside the reactor with DCM (dichlorome-
thane) (2ꢀ5 mL) and DMF (1ꢀ5 mL). The Rink amide resin
(0.61 mmol/g) was acylated with 2.5 equivalents of N-Fmoc
amino acid in the presence of 2.5 equivalents of BOP [benzo-
triazol-1-yloxytris(dimethylamino)phosphonium hexafluoro-
phosphate] and 6 equivalents of DIEA (N,N’-diisopropylethyl-
amine) in DMF. After 30 min the resin was washed (3ꢀeach)
with DMF, DCM and MeOHand controlled with the TNBS
(trinitrobenzenesulfonic acid) test.
Cleavage of the Fmoc protecting group: The Fmoc protect-
ing group was removed with 5 mL of a solution of piperidine in
DMF (1:4) for 10 min. After filtration, the procedure was re-
peated and then washed (3ꢀeach) with DMF, DCM and
MeOH.
(2S)-3-N-Allyloxycarbonyl-N-tert-butoxycarbonyl-2-
(ethoxycarbonylmethoxy)-1,3-diaminopropane (8)
Ethyl bromoacetate (2.85 mL, 25.8 mmol) was added to a stir-
red solution of 7 (2.35 g, 8.6 mmol) in dry THF (2 mL) at room
temperature. Then sodium hydride (1.14 g, 4.5 equivs.) was
added during 1 h. After an additional 5 h the mixture was fil-
tered on Celite and washed with THFand the solvent evaporat-
ed. The residue was chromatographed (hexane/ethylacetate, 8/
2) and the product 8 was obtained as a colorless oil; yield: 2.04 g
(66%); Rf (ethyl acetate/hexane 1/1): 0.74; IR (neat): n¼2984
w, 2935 w, 1738 m, 1705 m, 1698 m, 1510 m, 1515 m, 1393 w,
Cleavage of the Alloc protecting group: The resin was treat-
ed with PhSiH3 (25 equivs.) in 5 mL of anhydrous DCM for 3
minutes under argon. After addition of Pd(PPh3)4 (0.2 equivs.)
the mixture was stirred for 20 min. The resin was washed with
DCM (10 mL), dioxane:H2O (9:1, 10 mL), DMF (10 mL) and
DCM (2ꢀ10 mL). The procedure was repeated.
1
1368 w, 1215 m, 1134 cmÀ1 m; HNMR (CDCl 3): d¼5.86–
5.82 (m, 1H plus 1H, NH), 5.35 (1H, NH), 5.25 (d, J¼1.5,
17.1 Hz, 1H), 5.15 (dd, J¼1.1, 10.3 Hz, 1H), 4.52 (d, J¼
4.5 Hz, 2H), 4.14 (m, 4H), 3.49–3.44 (m, 1H), 3.35–3.31 (m,
2H), 3.25–3.08 (m, 2H), 1.39 (s, 9H), 1.24 (t, 3H); 13C NMR
(CDCl3): d¼171.22, 156.72, 156.41, 132.90, 117.48, 79.14,
78.78, 67.02, 65.54, 61.17, 41.05, 40.55, 28.31, 14.09; HR-LSI-
MS: calcd. for C16H28N2O7 þHþ: 361.197477; found:
361.197450; [a]2D5: À5.0 (c 8.4 g mLÀ1 in CH2Cl2).
Capping of the N-terminus: At the end of the synthesis, the
resin was acetylated with a solution of acetic anhydride in
DCM (1:1) for 30 min. The resin was dried under vacuum
and stored at À208C.
TFA cleavage: The cleavage was carried out using TFA (tri-
fluoroacetic acid)/EDT (1,2-ethanedithiol)/H2O/TIS (triiso-
propylsilane) as a 94/2/2/2 solution for 3 h. The peptide was
precipitated with methyl tert-butyl ether then dissolved in a wa-
ter/acetonitrile mixture. All the dendrimers were purified by
preparative HPLC.
(3S)-4-(9-Fluorenylmethoxycarbonylamino)-3-
methyl(allyloxycarbonyl)aminoethoxyacetic Acid (1)
To a stirred solution of 8 (262 mg, 0.72 mmol) in THF (2 mL)
was added a solution of LiOH(53 mg, 2.19 mmol) in water
(2 mL). The mixture was stirred overnight. The THF was
evaporated and residue lyophilized. Then 2.5 mL of 3 N HCl
were added and the mixture was stirred at 608C during 3 h.
The solution pHwas adjusted to 5 with NaOHpellets and
then to pH8 with saturated aqueous NaHCO 3 (2.5 mL). A sol-
ution of 9-fluorenylmethyl chloroformate (Fmoc-Cl, 205 mg,
0.79 mmol) in dioxane (4.5 mL) was added at 08C and the mix-
ture was stirred overnight at room temperature, diluted with
water, and the aqueous phase was acidified to pH2. After an
extraction with ethyl acetate, the combined organic phases
were concentrated and the residue chromatographed (hex-
ane/ethyl acetate/acetic acid 3.8/6/0.2) to give 1 as white solid;
yield: 198 mg (60%); Rf (CHCl3/CH3OH/acetic acid 180/10/5):
0.56; mp 988C; IR (neat): n¼3019 w, 1734 m, 1718 m, 1700 m,
General procedure for homodimerization: The monomeric
dendrimer X (1 mg) was dissolved in water (25 mL) and a meth-
anolic solution of Aldrithiol was added (45 mM, 0.45 equivs.).
The pHof the solution was adjusted to 8 with an (NH 4)HCO3
buffer solution (20 mM). The solution was stirred for 30 mi-
nutes then acidified with one drop of TFA. The homodimer
XX was purified by semi-preparative RP-HPLC.
General procedure for heterodimerization: To a solution of
dendrimer X (1 mg/25 mL) in water was added a methanolic
solution of Aldrithiol (45.4 mM, 5 equivs.). The thiol-activa-
tion was followed by HPLC and, after evaporation of metha-
nol, the Aldrithiol excess was removed by extraction (3ꢀ)
with DCM. To this activated dendrimer solution was added a
solution of dendrimer Y (1 mg/25 mL, 1 equiv.) in water and
the pHof the solution adjusted to 8 with an (NH 4)HCO3 buffer
solution (20 mM). The solution was stirred during 30 minutes
then acidified with one drop of TFA. The heterodimer XY
was purified by semi-preparative RP-HPLC.
1
1517 m, 1477 m, 1215 s, 1135 cmÀ1 w; HNMR (CDCl 3): d¼
7.77 (d, J¼7.3 Hz, 2H), 7.59 (d, J¼7.32 Hz, 2H), 7.37 (dd,
J¼8.4, 14.3 Hz, 2H), 7.27 (dd, J¼8.4, 14.25 Hz, 2H), 5.91–
5.82 (m, 1H), 5.74 (1H, NH), 5.30–5.21 (2 overlapping dd,
2H), 4.57 (d, J¼5.8 Hz, 2H), 4.39 (d, J¼7.0 Hz, 2H), 4.22 (m,
3H), 3.54 (m, 1H), 3.36–3.26 (m, 4H); 13C NMR (CDCl3):
d¼177.01, 157.26, 157.15, 143.78, 141.28, 132.62, 127.72,
Yields, purification method and characterization of all den-
drimers are described in the Supporting Information.
1202
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1195–1204

Esterolytic Peptide Dendrimers with a Hydrophobic Core
FULL PAPERS
Kinetic Measurements
References and Notes
The kinetic measurements were carried out by using a SPEC-
TRAMax fluorescence detector with preset values of the exci-
tation and emission wavelengths corresponding to the meas-
ured pyrene substrate (lex ¼460 nm, lem ¼530 nm) at 25.28C.
Assays were followed in individual wells of round-bottom pol-
ystyrene 96-well plates (Costar). Kinetic experiments were fol-
lowed for 2 h. The dendrimers were stored at À208C in 1 mM
stock solution in B (acetonitrile/water: 1/1). Dendrimer stock
solutions were freshly diluted to 0.05 mM solution in 20 mM
aqueous bis-Tris pH6.0. The bis-Tris buffer, pH6.0 was pre-
pared using MilliQ deionized water. Fluorescence data were
converted to product concentration by means of a calibration
curve. Initial reaction rates were calculated from the steepest
part observed during the first 2000 sec of each curve. In a typ-
ical experiment, 20 mL of aqueous bis-Tris pH6.0 (20 mM)
were first added in a well, then 2.5 mL of a dendrimer solution
(0.05 mM in aqueous bis-Tris pH6.0, final concentration in the
well: 5 mM), and last 2.5 mL of substrate solution (2 mM in B,
final concentration in the well: 200 mM). The rate observed un-
der these conditions is the apparent rate Vapp. Vuncat is the rate
observed with 22.5 mL aqueous bis-Tris pH6.0 (20 mM) and
2.5 mL of substrate solution (2 mM in B, final concentration
in the well: 200 mM). The observed rate enhancement Vnet/Vuncat
is defined as (Vapp/Vuncat)–1.
[1] a) L. Stryer, Biochemistry, New York, WHFreeman and
Company, 1999; b) A. Fersht, Structure and Mechanism
in Protein Science, New York, WHFreeman and Compa-
ny, 1999.
[2] A. J. Kirby, Angew. Chem. 1996, 108, 770–790; Angew.
Chem. Int. Ed. 1996, 35, 707–724.
[3] a) A. Tramontano, K. D. Janda, R. A. Lerner, Science
1986, 234, 1566–1570; b) S. J. Pollack, J. W. Jacobs, P. G.
Schultz, Science 1986, 234, 1570–1573; c) R. A. Lerner,
S. J. Benkovic, P. G. Schultz, Science 1991, 252, 659–
667; d) P. G. Schultz, R. A. Lerner, Acc. Chem. Res.
1993, 26, 391–395; e) P. G. Schultz, R. A. Lerner, Science
1995, 269, 1835–1842; f) G. MacBeath, D. Hilvert, Chem.
Biol. 1996, 3, 433–445; g) N. R. Thomas, Nat. Prod. Rep.
1996, 13, 479–511; h) J. Hasserodt, Synlett 1999, 12,
2007–2022; i) J.-L. Reymond, Top. Curr. Chem. 1999,
200, 59–93; j) J. D. Stevenson, N. R. Thomas, Nat.
Prod. Rep. 2000, 17, 535–577; k) P. G. Schultz, J. Yin,
R. A. Lerner, Angew. Chem. 2002, 114, 4607–4618;
Angew. Chem. Int. Ed. 2002, 41, 4427–4437.
´
[4] a) S. Julia, J. Masana, J. C. Vega, Angew. Chem. 1980, 19,
968–969; Angew. Chem. Int. Ed. Engl. 1980, 19, 929–
´
930; b) S. Julia, J. Guixer, J. Masana, J. Rocas, S. Colonna,
Michaelis–Menten parameters were obtained from the line-
ar double reciprocal plot of 1/Vnet vs. 1/[S] measured similarly
with (final concentrations) 5 mM dendrimer (Vapp) or no den-
drimer (Vuncat) and 50, 75, 100, 150, 200, 300, 400, 500, 600,
700, 800 mM substrate, 20 mM bis-Tris, 258C. Kinetic param-
eters such as maximum velocity (Vmax) and Michaelis constant
(KM) were determined by the least squares method from a
Lineweaver–Burk plot between 1/v0 (reciprocal value of initial
velocity) and 1/[S]. The reaction activity reflected in kcat for the
R. Annuziata, H. Molinari, J. Chem. Soc. Perkin Trans. 1
1982, 1317–1324; c) P. A. Bentley, R. W. Flood, S. M.
Roberts, J. Skidmore, C. B. Smith, J. A. Smith, Chem.
Commun. 2001, 1616- 1617.
[5] a) L. K. Baltzer, S. Broo, H. Nilsson, J. Nilsson, Bioorg.
Med. Chem. 1999, 7, 83–91; b) J. Nilsson, L. Baltzer,
Chem. Eur. J. 2000, 6, 2214–2220; c) S. Kerstin, H. Nils-
son, J. Nilsson, L. Baltzer, J. Am. Chem. Soc. 1998, 120,
10287–10295.
hydrolysis (deacylation of substrates) was evaluated by kcat
¼
Vmax/[D]0, where [D]0 indicates the initial concentration of den-
drimers. The reaction rate with 4-methylimidazole (4-MeIm)
was obtained under the same conditions with 50, 75, 100, 150,
200, 300, 400, 500, 600, 700, 800 mM 4-MeIm and 200 mM sub-
strate. The second order rate constants k2 were calculated
from linear regression of the experimentally measured pseu-
do-first order rate constants k’ as a function of N-methylimida-
zole concentrations [M] where [S] indicates the initial substrate
concentration: Vnet ¼k’[S], k’¼k2[M]. The inhibition constant
KI was obtained from the linear reciprocal plot of 1/Vnet vs.
[I] measured similarly with (final concentrations) 5 mM den-
drimer and 200 mM of substrate (Vapp) and 0, 10, 20, 40, 60,
80, 100, 200, 400, 600, 800, 1000 mM inhibitor, 20 mM bis-Tris,
258C. Ki was calculated from a Dixon plot of inhibition data
and gave reproducible values withꢁ10% error.
[6] K. W. Hahn, W. A. Klis, J. M. Stewart, Science 1990, 248,
1544–1547; b) K. Johnsson, R. K. Allemann, H. Widmer,
S. A. Benner, Nature 1993, 365, 530–532.
[7] a) P. Krattiger, C. McCarthy, A. Pfaltz, H. Wennemers,
Angew. Chem. 2003, 115, 1763–1766; Angew. Chem.
Int. Ed. 2003, 42, 1722–1724; b) H. Wennemers, M. C.
Nold, M. M. Conza, K. J. Kulicke, M. Neuburger,
Chem. Eur. J., 2003, 9, 442–448.
[8] For example, a) E. R. Jarvo, S. J. Miller, Tetrahedron
2002, 58, 2481–2495; b) B. List, Tetrahedron 2002, 58,
5573–5590; c) G. C. Fu, Acc. Chem. Res. 2000, 33, 412–
420; d) M. H. Fonseca, B. List, Curr. Opin. Chem. Biol.
2004, 8, 319–326.
[9] a) A. Esposito, E. Delort, D. Lagnoux, F. Djojo, J. L.
Reymond, Angew. Chem. 2003, 115, 1419–1421; Angew.
Chem. Int. Ed. 2003, 42, 1381–1383; b) D. Lagnoux, E.
Delort, C. Douat-Casassus, A. Esposito, J. L. Reymond,
Chem. Eur. J. 2004, 10, 1215–1226; c) C. Douat-Casassus,
T. Darbre, J. L. Reymond, J. Am. Chem. Soc. 2004, 126,
7817–7826; d) A. Clouet, T. Darbre, J.-L. Reymond, An-
gew. Chem. Int. Ed. 2004, in press.
Acknowledgements
This work was financially supported by the University of Bern
and the Swiss National Science Foundation.
[10] a) A. W. Bosman, H. M. Janssen, E. W. Meijer, Chem.
Rev. 1999, 99, 1665–1688; b) D. K. Smith, Tetrahedron
´
2003, 59, 3797–3798; c) S. Hecht, J. M. J. Frechet, Angew.
Chem. 2001, 113, 76–94; Angew. Chem. Int. Ed. 2001, 40,
Adv. Synth. Catal. 2004, 346, 1195–1204
asc.wiley-vch.de
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
1203

Anthony Clouet et al.
FULL PAPERS
74–91; d) G. R. Newkome, C. N. Moorefield, F. Vögtle,
Dendritic Molecules: Concepts, Synthesis, Perspectives,
VCH, Weinheim, 1996.
[14] J. C. Martin, D. F. Smee, J. P. H. Verheyden, J. Org.
Chem. 1985, 50, 755–759.
[15] a) P. Lloyd-Williams, F. Albericio, E. Giralt, Chemical
Approaches to the Synthesis of Peptides and Proteins,
CRC Press, Boca Raton, New York, 1997; b) C. Chan,
P. D. White, Fmoc Solid Phase Peptide Synthesis: a prac-
tical approach, Oxford Press, New York, 2000.
[16] L. A. Carpino, A. El-Faham, J. Am. Chem. Soc. 1995,
117, 5401–5402.
[17] a) K. Akaji, K. Fujino, T. Tatsumi, Y. Kiso, Tetrahedron
Lett. 1992, 33, 1073–1076; b) M. Ruiz-Gayo, M. Royo,
I. Fernandez, F. Albericio, E. Giralt, M. Pons, J. Org.
Chem. 1993, 58, 6319–6328.
[11] a) D. A. Tomalia, P. R. Dvornic, Nature 1994, 372, 617–
618; b) J. W. J. Knapen, A. W. Van der Made, J. C. De
Wilde, P. W. N. M. van Leeuwen, P. Wijkens, D. M.
Grove, G. van Koten, Nature 1994, 372, 659–662; c) L.
Balogh, D. A. Tomalia, J. Am. Chem. Soc. 1998, 120,
7355–7356; d) M. Zhao, R. M. Crooks, Angew. Chem.
1999, 111, 375–377; Angew. Chem. Int. Ed. 1999, 38,
364–366; e) B. Bhyrappa, G. Vaijayanthimala, J. Am.
Chem. Soc. 1999, 121, 262–263; f) S. Nlate, J. Ruiz, V.
Sartor, R. Navarro, J. C. Blais, D. Astruc, Chem. Eur. J.
2000, 6, 2544–2553; g) P. Weyermann, F. Diederich,
Helv. Chim. Acta 2002, 85, 599–617; h) H. Sellner, P. B.
Rheiner, D. Seebach, Helv. Chim. Acta 2002, 85, 352–
387; i) L. J. Twyman, A. S. H. King, I. K. Martin, Chem.
Soc. Rev. 2002, 31, 69–82.
[18] a) N. Bensel, M. T. Reymond, J.-L. Reymond, Chem.
Eur. J., 2001, 7, 4604; b) E. Leroy, N. Bensel, J.-L. Rey-
mond, Bioorg. Med. Chem. Lett. 2003, 13, 2105–2108.
´
[19] a) N. Sakai, N. Sorde, S. Matile, J. Am. Chem. Soc. 2003,
125, 7776–7777; b) B. Baumeister, N. Sakai, S. Matile,
Org. Lett. 2001, 3, 4229–4232.
[12] a) K. Sivanandan, D. Vutukuri, S. Thayumanavan, Org.
Lett. 2002, 4, 3751–3753; b) W. Zhang, D. T. Nowlan,
III, L. M. Thomson, W. M. Lackowski, E. E. Simaneck,
J. Am. Chem. Soc. 2001, 123, 8914–8922 and references
[20] Obtained previously by kinetic resolution catalyzed by
chiral (salen)CoIII complexes: S. E. Schaus, B. D.
Brandes, J. F. Larrow, M. Tokunaga, K. B. Hansen,
A. E. Gould, M. E. Furrow, E. N. Jacobsen, J. Am.
Chem. Soc. 2002, 124, 1307–1315.
´
therein; c) S. M. Grayson, J. M. J. Frechet, J. Am.
Chem. Soc. 2000, 122, 10335–10344.
´
[13] a) N. Thieriet, J. Alsina, E. Giralt, F. Guibe, F. Albericio,
Tetrahedron Lett. 1997, 38, 7275–7278; b) E. J. Corey,
J. W. Suggs, J. Org. Chem. 1973, 38, 3223–3224.
1204
ꢀ 2004 WILEY-VCHVerlag GmbH& Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1195–1204