Solid-supported synthesis of highly functionalized tripodal peptides with flexible but preorganized geometry: Towards potential serine protease mimics

Article abstract of DOI:10.1002/ejoc.200600145

Tripodal scaffold 1 has been used in the synthesis of a representative member of a library of serine protease mimics, possessing three independent functionalized peptide chains on a central core. Each peptide chain contains one residue of the classical ca

Full text of DOI:10.1002/ejoc.200600145

FULL PAPER
DOI: 10.1002/ejoc.200600145
Solid-Supported Synthesis of Highly Functionalized Tripodal Peptides with
Flexible but Preorganized Geometry: Towards Potential Serine Protease
Mimics
An Gea,[a] Nadia Farcy,[a] Núria Roqué i Rossell,[a] José C. Martins,[a] Pierre J. De Clercq,[a]
and Annemieke Madder*[a]
Keywords: Serine protease / Enzyme models / Tripodal scaffold / Protecting groups / Orthogonal deprotection
Tripodal scaffold 1 has been used in the synthesis of a repre-
sentative member of a library of serine protease mimics, pos-
sessing three independent functionalized peptide chains on
a central core. Each peptide chain contains one residue of
the classical catalytic triad (serine, histidine and aspartate)
found in the active site of the serine protease α-chymotryp-
sin. A particular feature of the novel tripodal design is its
essentially flexible yet preorganized structure as deduced
from molecular modeling studies. The choice of suitable scaf-
fold and side-chain protecting groups was intensively
studied and optimized to ensure complete orthogonality. In-
clusion of a photolabile linker enabled the release of interme-
diates and final structures from the solid-support polymer al-
lowing positive evaluation of the present strategy for the syn-
thesis of future tripodal libraries of serine protease mimics.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
scaffolds described, which are mostly based on a rigid struc-
ture,[8] the present design concerns the presentation of the
different functionalities necessary for the generation of in-
dependent strands on an essentially flexible core unit. How-
Introduction
Mimicking the efficiency and the specificity of enzymes
has already been the subject of many investigations.[1] The
catalytic power of the serine protease α-chymotrypsin in the
cleavage of amide bonds is a typical example.[2] However,
the development of efficient catalysts for this process still
remains a challenge in organic chemistry. Past efforts to de-
velop α-chymotrypsin models relied on the design of a
unique non-peptidic[3] or peptidic[4] molecule possessing a
suitable geometric arrangement of functionalities for cataly-
sis. Several years ago, combinatorial chemistry was intro-
duced as a method to speed up the search for catalytic ac-
tivity.[5] Therefore, the traditional approach of carefully
designing one target molecule was replaced by the design
and synthesis of libraries of molecules with the hope of in-
creasing the chance of finding a catalyst.
To explore this field, a long-term program was set-up in
our laboratory to develop a potentially catalytic hydrolytic
system in which three independent peptide chains are con-
structed, each containing one residue of the catalytic triad,
i.e. serine, histidine and aspartate. Initial efforts focussed on
the construction of dipodal model systems based on a cho-
lic acid template.[6] These were followed by the synthesis of
tripodal construct 1 as a scaffold.[7] In contrast to the other
Figure 1. Result of molecular modeling studies on orthogonally
protected tripodal scaffold 1 in H2O (frontal view): a) minimum
energy conformation, b) conformation according to diamond lat-
tice structure 1 with benzylic chains in “axial” positions and c)
conformation according to diamond lattice structure 1 with ben-
zylic chains in “equatorial” positions. Top views are included in the
supporting information.
[a] Department of Organic Chemistry, Ghent University,
Krijgslaan 281, S4; 9000 Gent, Belgium
Fax: +32-264-49-98
E-mail: annemieke.madder@ugent.be
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4135

A. Madder et al.
FULL PAPER
ever, despite this flexibility, molecular modeling studies
have shown that scaffold 1 preferentially adopts a fairly pre-
organized structure in which the three benzylic chains are
oriented in parallel.[9] As illustrated in Figure 1, the global
minimum (a) at 0 kJ/mol illustrates a conformation where
the three benzylic chains are perfectly parallel. Other con-
formations presenting the three benzylic ethers in the same
direction (such as b) are also populated. The corresponding
“all-equatorial” conformation (c) is considerably less stable
as evidenced by its higher relative energy value. Both in
water and in vacuo the same trend is observed (see Support-
ing Information). This important feature is expected to
favor the proximity and interactions between the catalytic
residues present in each appended peptide strand. However,
the usefullness of this scaffold for the generation of multi-
functional serine protease mimic libraries remains to be
proven. In this work we report on our current efforts aimed
at the use of the flexible yet preorganized scaffold 1 in the
generation of tripodal libraries with potential serine prote-
ase like activity. The validity of the approach is demon-
strated by the synthesis and detailed characterisation of one
possible library member.
Results and Discussion
The synthesis of tripodal scaffold 1, in its racemic form,
started from pentaerythritol by consecutive Williamson
ether syntheses (Scheme 1). The first benzylic chain was in-
troduced by a three-step sequence involving, in first in-
stance, the temporary protection of one of the hydroxy
functions as a tert-butyldiphenylsilyl ether. Subsequently,
classical ether formation with building block 6 followed by
deprotection of the silyl ether led to the triol 2. The other
two benzylic chains were introduced by similar procedures
leading to the alcohol 3. Subsequently, the remaining hy-
droxy function was treated with a large excess of succinic
anhydride to afford scaffold 1 in 90% yield. The succinic
moiety serves as a spacer for the attachment to the solid-
phase. The benzylic bromides 6, 7 and 8 were readily pre-
pared from the commercially available 4-(methyl)phenylace-
tic acid by bromination followed by the reduction of the
carboxylic acid function with a borane–methyl sulfide com-
plex to form the alcohol 5.
Scheme 1. Convergent synthesis of tripodal scaffold 1. Reagents:
(a) TBDPSCl, imidazole, DMF, 68%; (b) 6, NaH, Bu4NI, THF,
43%; (c) TBAF, THF, 83%; (d) 7, NaH, Bu4NI, THF, 30%, (e) 8,
NaH, Bu4NI, THF, 49%; (f) succinic anhydride, DMAP, CH2Cl2,
90%; (g) Br2, AIBN, CCl4, 62%; (h) BH3.SMe2, THF, 86%; (i)
PPTS, dihydropyran, CH2Cl2, 88%; (j) trichloroacetimidate 9,
PPTS, CH2Cl2, 69% and (k) TBDPSCl, DIPEA, DMF, 71%.
The introduction of the three different protective groups
was carried out according to standard procedures.[10] It was
further shown[7] that after immobilisation on TentaGel-
NH2 via the use of a photocleavable linker,[11] selective re- der to ensure the absence of error sequences and missing
moval of the protecting groups was possible. In this way members, especially in view of the high degree of function-
scaffold 1 is ideally suited for the construction of libaries of alisation of the envisaged library. Therefore a representative
serine protease mimics containing three differently func- tripodal member was synthesized in order to evaluate dif-
tionalised parallel peptide chains.
ferent parameters. The stability of the benzylic ethers pres-
Condensation of the polymer support TentaGel-NH2 ent in scaffold 1 as well as the stability of the ester linkages
with the carboxylic acid 10 was followed by Fmoc deprotec- to the peptide chains were examined. Moreover, the orthog-
tion, after which tripod 1 was attached to the free amino onality and stability of all protecting groups present, either
function affording the polymer supported scaffold 11 on the scaffold or on the side-chain of the catalytic residues
(Scheme 2).
(serine, histidine and aspartic acid), had to be verified. In
Prior to the generation of large libraries, the synthetic addition, these protecting groups should be easily and
processes need to be carefully checked and optimized in or- smoothly removable at the end of the synthesis. In view of
4136
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146

Potential Serine Protease Mimics
FULL PAPER
Subsequently, a first peptide strand Scaffold-Gly-Ser-
(OTrt)-Phe-NHAc (CǞN) was generated using a classical
peptide synthesis protocol. First attempts to obtain a quan-
titative coupling of Fmoc--glycine on the free hydroxy
function, used the reactive reagent FmocGlyCl in a large
excess. Since this reagent needed to be freshly prepared be-
fore each coupling, the method was replaced by the easy
and efficient use of N,NЈ-diisopropylcarbodiimide/1-hy-
droxy-1H-benzotriazole (DIC/HOBt) in presence of an ex-
cess of diisopropylethylamine and a catalytic amount of 4-
(dimethylamino)pyridine.[12] The efficiency of this coupling
reaction was monitored using UV/Vis spectroscopy.[13] For
the remaining peptide couplings, the N-methylmorpholine/
(benzotriazole-1-yloxy)tripyrrolidinophosphonium hexaflu-
orophosphate (NMM/PyBOP) strategy was initially
tested.[14] The success of each coupling reaction was moni-
tored by two colorimetric tests, TNBS[15] and NF31[16] and
by exposure of a sample of the resin (1 mg) to UV light at
365 nm for 3 h in acetonitrile followed by ES-MS analysis
(see Exp. Sect. and Supporting Information). After final
Fmoc deprotection, the free amine was acylated using a
large excess of acetylimidazole. The structure of the mono-
podal member was confirmed after submission of a sample
of resin 12 to photolytic cleavage. The structure of the re-
1
sulting compound 13 was confirmed by H NMR and ES-
MS. Accordingly, the second tripeptide was built. A slight
excess of dichlorodicyanoquinone at 0 °C for 1 h was used
to remove the 3,4-dimethoxyphenylmethyl (DMPM) pro-
tecting group. In this step, the reaction time and the tem-
perature were particularly crucial. Cleavage of the benzyl
Scheme 2. Synthesis of polymer bound scaffold 11. Reagents: (a)
TentaGel-NH2, DIC, HOBt, DMF; (b) 20% pip./DMF and (c) 1,
EDC, DMAP, NMP.
the ester attachment of the peptide chains, special attention ethers present in the tripodal scaffold was observed when
needs to be made during deprotection to ensure no trans- longer reaction times and/or higher temperatures were
esterification takes place.
applied. The functionalised sequence Scaffold-Gly-
The synthesis of tripodal member 21 is outlined in His(NMtt)-Ala-NHAc (CǞN) was generated from the free
Scheme 3 (see Supporting Information for ES-MS spectra hydroxy function using similar coupling conditions as be-
of all intermediates and final compounds).
fore.
After removal of the side-chain trityl ethers in compound
Following subsequent deprotection of the tripodal poly-
mer-supported scaffold 11, three different tripeptides were 14 with chloroacetic acid, photolytic cleavage in acetonitrile
constructed on the liberated hydroxy functions. On each of a sample (1 mg) of the intermediate followed by ES-MS
peptidic chain, one catalytic residue was introduced. The analysis revealed a mixture of the desired compound and a
Fmoc/tBu strategy was applied, thus Nα-Fmoc protected side-product with a lower molecular weight. Surprisingly,
amino acids containing an acid-labile side-chain protecting characterisation of this side-product showed the cleavage of
group were selected. This implies that the tetrahydropyranyl the first peptide strand that was introduced. This side-reac-
(THP) group, which is cleaved under acidic conditions, tion was easily avoided by replacing the NMM/PyBOP
needs to be removed first. In a previous communication, methodology by the DIC/HOBt coupling strategy.[17]
the selective deprotection of the THP acetal group was car-
Using the new coupling strategy, the dipodal peptide 15
ried out with a solution of AcOH/CH2Cl2/H2O in a 80:15:5 was obtained and its structure confirmed after photolysis
1
ratio at 60 °C and monitored using gel phase 13C NMR of resin 14 and characterisation by H NMR spectroscopy.
spectroscopy.[7] Nevertheless, in the synthesis of member 21, Further analysis of 14 was performed by ES-MS after re-
as we chose to follow and verify reactions by a cleavage/MS moval of the side-chain protecting groups on serine and his-
analysis protocol, we found that, using this more sensitive tidine with a solution of chloroacetic acid in dichlorometh-
and accurate approach, the earlier reaction conditions de- ane. Although a solution of 5% trifluoroacetic acid[18] is
scribed cause an esterification side-reaction leading, to a usually applied for these deprotections, we chose this milder
small extent, to concurrent acetylation of the liberated hy- alternative since the tripodal scaffold is sensitive under
droxy group (which was not observed using the less sensi- harsh acid conditions.
tive gel phase 13C NMR). As an alternative, pyridinium
para-toluenesulfonate removed the THP protecting group to be removed to allow the introduction of the third peptide
Finally, the tert-butyldiphenylsilyl (TBDPS) ether needed
without problems.
chain including aspartic acid. Several known methods to
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4137

A. Madder et al.
FULL PAPER
Scheme 3. Synthesis of tripodal member 16.
deprotect the silyl ether were screened. The standard pro- tion but when the combination of TBAF/AcOH was used
cedure using tetrabutylammonium fluoride (TBAF) can to remove the silyl ether in 14, partial deprotection of the
cause problems during subsequent photolytic cleavage due acid-labile side-chain protecting groups (trityl and meth-
to the destruction of the o-nitrobenzyl unit present in the yltrityl) occurred. Similarly, the widely used conditions in-
photolabile linker 10, a problem that was observed earlier volving hydrogen fluoride/pyridine were apparently too
by Nestler.[19] The addition of a stoichiometric amount of acidic since the trityl and methyltrityl were removed.[20] As
acetic acid to the reaction mixture suppresses this side-reac- an alternative, the use of ammonium fluoride was tested.
4138
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146

Potential Serine Protease Mimics
FULL PAPER
Unfortunately, this led to partial hydrolysis of the ester link-
The final deprotection of the side-chain protecting
ages present in compound 14 as side-reaction.[21] Finally, groups of serine, histidine and aspartic acid was then tack-
the use of the less common reagent TASF [tris(dimethylam- led. The trityl protection on the serine and the methyltrityl
ino)sulfonium (trimethylsilyl)difluoride] was successful in protection on the histidine were smoothly removed using
removing the TBDPS group and led to a clean deprotection a solution of chloroacetic acid in dichloromethane to give
after 8h of reaction.[22] It should be noted that the quality compound 18. Following cleavage from the support, inter-
of the TASF reagent used is important for the success of mediate 19 was characterized by ES-MS (see Supporting
this reaction. Older samples of reagent gave rise to incom- Information, spectrum 11). However, during the aspartic
plete deprotections. Subsequently, a last tripeptide Scaffold- acid deprotection difficulties were encountered. The usual
Gly-Asp(COO(CH2)2SiMe3)-Val-NHAc (CǞN) was built method described in the literature to remove the TMSE
using the same DIC/HOBt protocol. The most frequently group involves the use of TBAF in DMF.[24] No removal of
used side-chain protected aspartic acid derivatives include the TMSE group was observed upon use of TBAF/AcOH.
benzyl, allyl, adamantyl or tert-Bu esters. These protecting As for the deprotection using TASF, concurrent aspartim-
groups are removed under hydrogenolysis, palladium treat- ide formation took place, as evidenced by the observation
ment or in the presence of a strong acid.[23] All these depro- of the dehydrated product in the ES-MS spectrum. Asparti-
tection conditions are incompatible with the benzyl ethers mide formation is known to be a problem during the syn-
in scaffold 1. Therefore, in order to assure full orthogonal- thesis and deprotection of aspartic acid containing pep-
ity, a trimethylsilylethyl ester (TMSE) protection was cho- tides[26] and can usually be suppressed by using sterically
sen for the aspartic acid side-chain. The required Fmoc- hindered protections on the aspartyl side chain such as tert-
protected Asp(OTMSE) was prepared according to litera- butyl or adamantyl esters.[27] These are to be removed un-
ture procedures and was successfully incorporated in the der strongly acidic conditions which, as earlier described,
last peptide strand.[24] Despite the well-known occurrence are not compatible with the structure of our tripodal scaf-
of aspartimide formation during synthesis of Asp-Gly se- fold. Even when the latter is not an issue, difficulties have
quences, the desired tripodal member was synthesized with- been encountered during the deprotection of tert-butyl or
out problems.
adamantyl protected aspartic acid residues and additional
This is supported by the structure analysis by 2D NMR amide backbone protection seems to be necessary to com-
assignment techniques at 500 MHz of the fully protected pletely exclude aspartimide formation.[28] However, in ear-
tripodal member 17, obtained after photolytic cleavage of lier reports of the synthesis of Asp containing combinato-
a sample of resin 16 (DQF-COSY, TOCSY and ROESY rial peptide libraries this problem was not mentioned,[29]
spectra are included in the Supporting Information). Com- although it is clear from the current discussion that the
bined analysis of DQF-COSY and TOCSY allowed the complete absence of aspartimide formation can only be se-
straightforward identification of the single Ala and Val resi- cured via rigourous quality control at all steps of the syn-
dues, as well as the three Gly residues from their typical thesis. Also in our case, the desired product can be synthe-
connectivity patterns.[25] For assignment of the Asp, His, sized and fully characterised in the protected form (vide
Phe and Ser residues, the amide to aliphatic area of the supra no aspartimide nor β-aspartyl residues were detected
ROESY spectrum was analyzed for the presence of strong during NMR analysis) and problems only arise during the
and unambiguous αHi-NHj and βHi-NHj NOE contacts final deprotection step.
connecting pairs of amino acid spin systems i and j iden-
For this reason we resynthesized test member 22a revers-
tified previously. This allows the assignment and confirm- ing the order in the third strand to give an Asp-Val-Gly
ation of the identity of the Val-Asp(COO(CH2)2SiMe3)-Gly sequence where the more sterically hindered valine residue
and Ala-His-Gly[2] tripeptides. For the remaining tripeptide should provide protection against aspartimide formation
the N-terminal residue was identified from a NOE of its (Scheme 4).
NH to the preceding N-acetyl methyl resonance and estab-
The trityl and methyltrityl protections on Ser and His
lished as a Phe, from the observation of the 75 ms TOCSY were smoothly removed with the usual chloroacetic acid
spectrum of weak but clear correlations indicative of long- solution. Nevertheless, using the fluoride-based reagent
range coupling between resonances in the aromatic region TASF for TMSE removal from Asp, dehydration products
and both βH resonances. The middle residue features the resulting from aspartimide formation during TMSE depro-
most downfield βHi resonances, at a position consistent tection, could still be observed via ES-MS (see Supporting
with a serine-type side-chain, thereby establishing the third Information, spectrum 14). We therefore decided to system-
tripeptide as Phe-Ser(OTrt)-Gly (the detailed analysis pro- atically explore a variety of deprotecting reagents taking
cedure is outlined in the Supporting Information). Notwith- both the lability of the TMSE group and the stability of the
standing the fact that 17 still features complete functional scaffold into account. An overview of the conditions tested
group protection as well as the complete scaffold and taking is given in Table 1.
sequential incorporation of each tripeptide onto the
It is clear from the presented data that no satisfying con-
scaffold into account, our NMR analyses provides con- ditions could be found for the TMSE deprotection. Apply-
clusive evidence for the incorporation of the various ing HF/pyridine, after trityl and methyltrityl removal,
residues at the desired positions in the polymer bound showed that deprotection of 22a under these conditions
precursor 16.
proceeds by concurrent ester cleavage. Use of boron trifluo-
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4139

A. Madder et al.
FULL PAPER
solution of chloroacetic acid in DCM/H2O in a 3:1 ratio.
In this way, efficient removal of all side-chain protections
was achieved as evidenced by the ES-MS spectrum of 23
where the corresponding aspartimide (M-18) could not be
detected (see Supporting Information).
From the study presented here, it is clear that for the
production of highly-funtionalised tripodal peptide libraries
envisaged, aspartimide formation is a serious concern.
Therefore, the synthesis strategy should be carefully de-
signed in order to avoid problems of aspartimide formation
and more specifically avoid sequences known to be particu-
larly prone to cyclization such as Asp-Gly, Asp-Ala, Asp-
Ser and Asp-Asn.[26] Furthermore, although the Asp-
OTMSE or Asp-O-allyl protective groups in principle allow
orthogonal methods to be used for their deprotection, care
should be taken since in the current design, using these pro-
tections, aspartimide formation could not be avoided. The
Asp(O-2-PhiPr) protection has proven to be the ideal solu-
tion for the current orthogonality problem and moreover
provides complete protection against unwanted aspartimide
formation. The synthesis of libraries of potential serine pro-
tease models based on the strategy developed and described
here is currently under study.
Scheme 4. Structure of the new Asp-Val-Gly member.
ride etherate (BF3·OEt2) gave rise to a considerable amount
of side products resulting from concurrent cleavage of the
various esters and benzyl ethers present in the tripodal pep-
tide.
Conclusion
In conclusion, a new tripodal scaffold has been used in
the construction of a single potential member of a library
of serine protease mimics, containing the catalytic residues
serine, histidine and aspartic acid. In this study, suitable
protecting groups were selected and their compatibility was
demonstrated. Particular attention has been paid to the
analysis and characterisation of the compounds throughout
the synthesis. It has been shown that the design of the syn-
thesis strategy deserves special attention when aspartic acid
residues are to be included. Only through extensive quality
control, e.g. by using ES-MS at every step of the synthesis,
the integrity of the final products can be assured. The devel-
opment of highly functionalised tripodal peptide libraries
can now be envisaged and will be reported upon in due
course.
In view of the difficulties encountered during the final
deprotection of the TMSE ether, another side chain protect-
ing group for the aspartic residue had to be selected. Depro-
tection studies on a representative test sequence showed the
Asp(O-allyl) protection to cause even more problems re-
lated to aspartimide formation, occurring here during
Fmoc removal and capping of the sequence. The 2-phenyl-
isopropyl protective group is reported to be removed with
1% TFA,[30] conditions which are still too harsh for the
described tripodal scaffold, as evidenced from preliminary
tests. However, during test reactions on the protected amino
acid building block, it was shown that deprotection could
be readily achieved by treatment with chloroacetic acid,
conditions previously used for the removal of the trityl and
methyltrityl protections on serine and histidine, respectively.
Thus, member 22b (Scheme 4) was synthesized using Fmoc-
Asp(O-2-PhiPr)-OH in the final coupling step. After Fmoc
Experimental Section
General Remarks: DMF and MeOH were HPLC grade and were
removal and capping, the resin was treated with a 52wt.-% provided by Riedel-De-Haën. The coupling reactions were per-
Table 1. Overview of deprotection tests for TMSE removal.
Reagent
Scaffold Stability
TMSE Deprotection
TBAF
TBAF/HOAc
TASF
HF/pyridine
BF3·OEt2
Ph.-L. not stable
n.t.[a]
+
SM[b]
+
aspartimide formation
fast deprotection
complete ester cleavage
partial ester and benzyl ether cleavage depending on deprotection observed without aspartimide forma-
concentration and reaction time
tion; completion dependent on concentration, temp.
and reaction time
SM
ClCH2COOH 52wt.-%,
DCM/H2O, 3:1
+
[a] n.t. = not tested. [b] SM = only starting material present.
4140
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146

Potential Serine Protease Mimics
FULL PAPER
formed in dry DMF purchased from Biosolve. Dichloromethane
(CH2Cl2) was distilled from calcium hydride and 1,4-dioxane was
distilled from NaBH4 under nitrogen. TentaGel-S-NH2 resin,
Fmoc-amino acids and HOBt were purchased from Novabiochem.
Diisopropylcarbodiimide (DIC) and 2,4,6-trinitrobenzenesulfonic
acid (TNBS) were obtained from Fluka. All reagents and solvents
were used as received without further purification. NF31 was syn-
thesized as described previously.[16]
132.5 (C), 135.6 (CH) ppm. MS: m/z (%) = 233 (10), 199 (100), 181
(19), 139 (76), 105 (25), 91 (91), 57 (51).
2-[(tert-Butyldiphenylsiloxy)methyl]-2-{4-(2-tetrahydro-2H-pyran-2-
yloxy)ethyl]benzyloxymethyl}propane-1,3-diol: See procedure (b) in
Scheme 1. Sodium hydride (3.8 g, 128 mmol) (80% dispersion in
mineral oil) in THF (250 mL) was added to a solution of mono-
TBDPS-protected pentaerythritol (64.0 g, 171 mmol) in dry THF
(100 mL). The mixture was stirred for 30 min at 40 °C. After cool-
ing to 10 °C, benzyl bromide 6 (38.5 g, 128 mmol) in THF (60 mL)
was added dropwise over a period of 20 min followed by the ad-
dition of tetrabutylammonium iodide (0.2 g, 54 mmol). The mix-
ture was then refluxed overnight. The THF was evaporated after
addition of an aqueous solution of ammonium chloride (200 mL).
The aqueous phase was extracted with EtOAc. The organic phase
was washed with water and then dried with anhydrous magnesium
sulfate. After filtration and evaporation of the solvent under re-
duced pressure, the crude compound was purified by column
chromatography (toluene/EtOAc, 8:2 to pure EtOAc) affording
34 g (48 %) of a yellow oil. Rf = 0.57 (toluene/EtOAc, 1:1). IR
1
1D H NMR and regular DEPT 13C NMR spectra were recorded
at 200 MHz and 50.3 MHz, respectively on a Varian Gemini-200
spectrometer, or at 500 MHz and 125.77 MHz, respectively on a
Bruker Avance DRX-500 MHz spectrometer. NMR structure
analysis of the fully protected tripodal member 17 was performed
on 3 mg of the compound dissolved in [D6]DMSO. A 2D DQF-
COSY spectrum and 2D TOCSY spectrum with a 75 ms MLEV17
spin-lock was recorded using standard sequences available from the
Bruker pulse program library. The off-resonance 2D ROESY[31]
spectrum was recorded with 300 ms mixing time to obtain suffic-
iently intense NOE cross-peaks for analysis. For all 2D NMR ex-
periments, 512 t1 increments with 4 K data points each were re-
corded with typically 16 to 64 scans. Following apodisation by a
squared cosine bell along both dimensions and zero-filling along
F1 to 2 K datapoints a double fourier transformation was executed,
followed by a 3rd order polynomial baseline correction along both
dimensions.
(neat): ν = 3438 (b), 2941 (s), 2864 (s), 1514 (m), 1467 (m), 1110
˜
(s), 1029 (s) cm–1. 1H NMR (500 MHz, CDCl3, 25 °C): δ = 1.05 (s,
9 H), 1.50–1.61 (m, 4 H), 1.70 (m, 1 H), 1.81 (m, 1 H), 2.45 (m, 2
H), 2.91 (t, 2 H, J = 7.2 Hz), 3.46 (m, 1 H), 3.55 (s, 2 H), 3.61 (m,
1 H), 3.70 (s, 6 H), 3.76 (m, 1 H), 3.94 (m, 1 H), 4.45 (s, 2 H), 4.60
(m, 1 H), 7.18 (2 H, AB-d, J = 7.9 Hz), 7.21 (2 H, AB-d, J =
7.9 Hz), 7.36–7.45 (m, 6 H), 7.64 (d, 4 H, J = 6.9 Hz) ppm. 13C
NMR (50 MHz, CDCl3, 25 °C): δ = 19.2, 19.5, 25.4, 26.8, 30.6,
36.0, 45.6, 62.9, 64.8, 64.9, 68.2, 71.6, 73.5, 98.7, 127.6, 127.8,
129.1, 129.9, 132.8, 136.5, 138.7 ppm. MS (ES): m/z 615 [M+Na+].
C35H48O6Si (592.32): calcd. C 70.91, H 8.15; found C 70.91, H
8.19.
Mass-spectra were recorded using a Finnigan MAT LCQ mass
spectrometer equipped with an ESI source (Thermofinnigan, San
José, CA). All measurements were carried out in the positive mode.
Deprotection of Nα-Fmoc was performed using 20% piperidine in
DMF (2×20 min). Washing volumes were about 2 times the vol-
ume used for the reactions. Each coupling and deprotection step
was monitored using the TNBS and NF31 tests. The coupling of
the free hydroxy function was monitored by Fmoc quantitation
using UV/Vis spectroscopy on a Varian Cary 3E spectrophotome-
ter at 300 nm. Resin photolyses were carried out at a distance of
1 cm using a 450-W Hg arc lamp, medium pressure, set at 365 nm.
For small quantities of resin (1 mg), photolysis was performed at a
distance of 1 cm using a 4 W bioblock Scientific compact UV lamp
set at 365 nm (relative intensity at 15 cm of 340 µW/cm2). All resin
bound compounds were analysed by ES-MS after photolytic cleav-
age (see Supporting Information for spectra of all intermediate and
final products).
2-Hydroxymethyl-2-{4-[2-(tetrahydro-2H-pyran-2-yloxy)-
ethyl]benzyloxymethyl}propan-1,3-diol (2): See procedure (c) in
Scheme 1. Tetrabutylammonium fluoride (70 mL, 1  in THF) was
added to a solution of diol (36.0 g, 61 mmol) in THF (250 mL).
The mixture was stirred at room temperature for 16 h. The solvent
was then evaporated under reduced pressure and the oil was puri-
fied by column chromatography (toluene/EtOAc, 9:1 to pure
EtOAc) affording 18 g of the desired triol (84 %). Rf = 0.31
(CH Cl /MeOH, 93:7). IR (neat): ν = 3394 (b), 2933 (s), 2861 (s),
˜
2
2
1651 (m), 1507 (m), 1456 (m), 1359 (s), 1318 (m), 1256 (m), 1200
(w), 1133 (s), 1113 (s), 1072 (s), 1026 (s) cm–1. 1H NMR (500 MHz,
CDCl3, 25 °C): δ = 1.47–1.58 (m, 4 H); 1.70 (m, 1 H), 1.80 (m, 1
H), 2.42 (t, 3 H, J = 5.5 Hz), 2.91 (t, 2 H, J = 7.2 Hz), 3.46 (m, 1
H), 3.50 (s, 2 H), 3.61 (dt, 1 H, J = 9.7, 7.2 Hz), 3.72 (d, 6 H, J =
5.3 Hz), 3.77 (m, 1 H), 3.94 (dt, 1 H, J = 9.7, 7.2 Hz), 4.48 (s, 2
H), 4.59 (t, 1 H, J = 3.6 Hz), 7.22 (2 H, AB-d, J = 8.5 Hz), 7.23 (2
H, AB-d, J = 8.5 Hz) ppm. 13C NMR (50 MHz, CDCl3, 25 °C): δ
= 19.5, 25.4, 30.6, 36.0, 44.9, 62.3, 64.5, 68.2, 72.2, 73.6, 98.7,
127.7, 129.2, 135.4, 138.9 ppm. MS (ES): m/z 377 [M + Na+].
C19H30O6 (354.20): C 64.39, H 8.53; found C 64.41, H 8.47.
Synthesis of Triol 2
2-[(tert-Butyldiphenylsiloxy)methyl]-2-(hydroxymethyl)propane-1,3-
diol: See procedure (a) in Scheme 1. To a solution of pentaerythritol
(25 g, 0.184 mol) and imidazole (13.8 g, 0.202 mol) in dimethyl-
formamide (1.2 L), tert-butyldiphenylsilyl chloride (26.8 g,
0.098 mol) was added dropwise. The mixture was stirred for 60 h
at room temperature. After addition of water, the solution was ex-
tracted with ethyl acetate. The organic phase was washed with
water, dried with anhydrous magnesium sulfate, filtered and con-
centrated. The remaining dimethylformamide was removed by kug-
elrohr distillation under reduced pressure. The oil was purified by
HPLC (ethyl acetate/isooctane, 1:1) to provide the mono-protected
triol 2 (25 g, 68%). M. p. 50 °C. Rf = 0.25 (ethyl acetate/isooctane,
Synthesis of Alcohol 3
2-{4-[2-(3,4-Dimethoxybenzyloxy)ethyl]benzyloxymethyl}-2-{4-[2-
(tetrahydro-2H-pyran-2-yloxy)ethyl]benzyloxymethyl}propan-1,3-
diol: See procedure (d) in Scheme 1. Sodium hydride (1.7 g,
56 mmol) (80% dispersion in mineral oil) in dry THF (140 mL)
was added to a solution of triol 2 (20 g, 56 mmol) in dry THF
(40 mL). The mixture was stirred for 30 min at 40 °C. After cooling
to 10 °C, benzyl bromide 7 (20.4 g, 56 mmol) in dry THF (100 mL)
1:1). IR (KBr): (ν) = 3386, 2930, 2857, 1472, 1428, 1390, 1361,
˜
1
1113, 1036, 824, 740, 702, 504 cm–1. H NMR (500 MHz, CDCl3,
25 °C): δ = 1.07 (s, 9 H), 2.75 (s, 3 H), 3.63 (m, 2 H), 3.71 (m, 6
H), 7.41 (m, 6 H), 7.64 (dd, 4 H, J = 8.0, 1.3 Hz) ppm. 13C NMR/ was added dropwise over a period of 20 min followed by the ad-
DEPT (50 MHz, CDCl3, 25 °C): δ = 19.1 (CH2), 26.9 (CH3), 45.6 dition of tetrabutylammonium iodide (0.2 g, 54 mmol). The solu-
(C), 64.3 (CH2), 64.7 (CH2), 65.3 (CH2), 127.9 (CH), 130.0 (CH),
tion was then refluxed overnight. After addition of a solution of
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4141

A. Madder et al.
FULL PAPER
ammonium chloride, the THF was evaporated. The aqueous phase
was extracted with EtOAc. The organic phase was washed with
water and dried with anhydrous magnesium sulfate. After filtration
1712 (w), 1656 (w), 1630 (w), 1512 (w), 1456 (w), 1425 (w), 1359
(w), 1256 (w), 1153 (w), 1092 (w), 10125 (w) cm–1. 1H NMR
(500 MHz, CDCl3, 25 °C): δ = 1.02 (s, 9 H), 1.55 (m, 4 H), 1.70
and evaporation, the crude mixture was purified by column (m, 1 H), 1.80 (m, 1 H), 2.45 (2 H, AB-t, J = 5.3 Hz), 2.48 (2 H,
chromatography (toluene/EtOAc, 9:1 to pure EtOAc) to give 14 g AB-t, J = 5.3 Hz), 2.83 (t, 2 H, J = 7.0 Hz), 2.89 (t, 2 H, J =
of an oil (37%). R = 0.26 (toluene/EtOAc, 1:1). IR (neat): ν = 7.1 Hz), 2.90 (t, 2 H, J = 6.8 Hz), 3.47 (s, 2 H), 3.49 (s, 4 H), 3.60
˜
f
3470 (b), 2939 (s), 2865 (s), 1606 (m), 1594 (m), 1556 (s), 1464 (s),
(m, 1 H), 3.66 (m, 1 H), 3.66 (t, 2 H, J = 7.0 Hz), 3.79 (m, 1 H),
1420 (m), 1261 (s), 1030 (s) cm–1. 1H NMR (500 MHz, CDCl3, 3.82 (t, 2 H, J = 7.0 Hz), 3.84 (s, 3 H), 3.87 (s, 3 H), 3.92 (m, 1 H),
25 °C): δ = 1.48–1.59 (m, 4 H), 1.70 (m, 1 H), 1.79 (m, 1 H), 2.63
(m, 2 H), 2.90 (t, 2 H, J = 6.9 Hz), 2.92 (t, 2 H, J = 6.9 Hz), 3.46
(m, 1 H), 3.54 (s, 2 H), 3.55 (s, 2 H), 3.60 (m, 1 H), 3.66 (m, 6 H),
3.75 (m, 1 H), 3.85 (s, 3 H), 3.87 (s, 3 H), 3.94 (m, 1 H), 4.46 (s, 6
H), 4.59 (t, 1 H, J = 3.4 Hz), 6.82 (m, 3 H), 7.21 (m, 8 H) ppm.
13C NMR (50 MHz, CDCl3, 25 °C): δ = 19.5, 25.4, 30.6, 36.0, 44.8,
4.22 (s, 2 H), 4.41 (s, 3 H), 4.42 (s, 3 H), 4.47 (s, 2 H), 4.61 (t, 1 H,
J = 3.3 Hz), 6.82 (m, 3 H), 7.09 (d, 2 H, J = 7.9 Hz), 7.18 (m, 10
H), 7.33–7.41 (m, 6 H), 7.60 (dd, 4 H, J = 7.8, 1.2 Hz) ppm. 13C
NMR (50 MHz, CDCl3, 25 °C): δ = 19.1, 19.4, 26.3, 26.7, 28.2,
28.6, 29.0, 30.6, 35.9, 38.9, 44.5, 55.6, 55.8, 62.1, 64.0, 65.0, 68.2,
68.9, 70.8, 72.7, 73.1, 98.7, 110.8, 110.9, 120.1, 127.4, 127.4, 128.7,
55.8, 55.9, 62.2, 65.0, 68.2, 70.8, 72.0, 72.9, 73.6, 98.7, 110.8, 110.9, 128.8, 129.0, 129.5, 130.8, 133.6, 135.4, 136.3, 138.0, 138.2, 148.4,
120.1, 127.5, 127.6, 128.9, 129.0, 129.1, 130.8, 135.6, 135.7, 138.6,
138.8, 148.7, 148.9 ppm. MS (ES): calcd. for C37H50O9Na: 661,
found m/z: 661 [M+Na+].
148.8, 171.7, 176.6 ppm. MS (ES): m/z 1128 (M+NH4+).
C66H82O13Si (1110.6): calcd. C 71.36, H 7.44; found C 70.72, H
7.63.
3-{4-[2-(tert-Butyldiphenylsilyloxy)ethyl]benzyloxy}-2-{4-[2-(3,4-di-
methoxybenzyloxy)ethyl]-benzyloxymethyl}-2-{4-[2-[tetrahydro-2H-
pyran-2-yloxy)ethyl]benzyloxymethyl}propan-1-ol (3): See procedure
(e) in Scheme 1. Sodium hydride (1.1 g, 36.5 mmol) (80% disper-
sion in mineral oil) in dry THF (140 mL) was added to a solution
of diol (23.3 g, 36.5 mmol) in dry THF (40 mL). The mixture was
stirred for 30 min at 40 °C. After cooling to 10 °C, benzyl bromide
7 (16.56 g, 36.5 mmol) in dry THF (60 mL) was added dropwise
over a period of 20 min followed by the addition of tetrabutylam-
monium iodide (0.2 g, 54.0 mmol). The mixture was then refluxed
overnight. The THF was then evaporated after addition of a solu-
tion of ammonium chloride (200 mL). The aqueous phase was ex-
tracted with EtOAc. The organic phase was washed with water and
then dried with anhydrous magnesium sulfate. After filtration and
evaporation the crude compound was purified by column
chromatography (toluene/EtOAc, 1:1) to afford 11.8 g of a yellow
Synthesis of Building Block 6, 7 and 8
Bromination of Commercially Available 2-(4-Methylphenyl)acetic
Acid. Synthesis of Acid 4: See procedure (g) in Scheme 1. To a solu-
tion of 2-(4-methylphenyl)acetic acid (25 g, 0.17 mol) in carbon tet-
rachloride (300 mL), bromine (8.7 mL, 0.17 mol) and 2,2Ј-azobis(i-
sobutyronitrile) (catalytic amount) were added. The solution was
irradiated using a sodium lamp (400 W) and was heated to 40 °C.
After 5 min, white crystals precipitated. The mixture was cooled
to 0 °C and the crystals were filtered off and washed with carbon
tetrachloride. The residue was recrystallized from ethyl acetate to
provide the product 4 as white crystals (24 g, 62%). M. p. 180 °C.
R = 0.20 (toluene/ethyl acetate, 8:2). IR (KBr): ν = 3027, 2972,
˜
f
1699, 1516, 1408, 1343, 1244, 1184, 1093, 894, 838, 780, 675,
1
602 cm–1. H NMR (500 MHz, CD3OD, 25 °C): δ = 3.60 (s, 2 H),
4.50 (s, 2 H), 7.26 (m, 2 H), 7.36 (m, 2 H) ppm. 13C NMR/DEPT
(50 MHz, CDCl3 +a few drops of [D6]DMSO, 25 °C): δ = 33.0
(CH2), 40.6 (CH2), 128.8 (CH), 129.4 (CH), 134.6 (C), 135.9 (C),
173.0 (C) ppm. MS: m/z (%) = 230 (M+·Br81), 228 (M+·Br79), 150
(13), 150 (19), 149 (100), 131 (5), 104 (42), 77 (26), 63 (9), 45 (20).
oil (32%). R = 0.16 (isooctane/EtOAc, 65:35). IR (neat): ν = 3518
˜
f
(b), 2934 (s), 1862 (s), 1513 (s), 1482 (s), 1359 (m), 1261 (s), 1108
(s), 1087 (s) cm–1. 1H NMR (500 MHz, CDCl3, 25 °C): δ = 1.02 (s,
9 H), 1.54–1.61 (m, 4 H), 1.70 (m, 1 H), 1.81 (m, 1 H), 2.84 (t, 2
H, J = 6.8 Hz), 2.86 (t, 1 H, J = 5.7 Hz), 2.90 (t, 2 H, J = 7.0 Hz),
2.91 (t, 2 H, J = 7.0 Hz), 3.46 (m, 1 H), 3.54 (s, 4 H), 3.55 (s, 2 H),
3.60 (m, 1 H), 3.66 (t, 2 H, J = 7.1 Hz), 3.75 (d, 2 H, J = 6.3 Hz),
3.77 (m, 1 H), 3.83 (t, 2 H, J = 7.0 Hz), 3.85 (s, 3 H), 3.87 (s, 3 H),
3.94 (m, 1 H), 4.44 (s, 6 H), 4.46 (s, 2 H), 4.59 (t, 1 H, J = 3.5 Hz),
6.83 (m, 3 H), 7.10 (2 H, AB-d, J = 8.0 Hz), 7.17 (2 H, AB-d, J =
8.0 Hz), 7.19 (m, 8 H), 7.33–7.42 (m, 6 H), 7.59 (d, 4 H, J = 6.8 Hz)
ppm. 13C NMR (50 MHz, CDCl3, 25 °C): δ = 19.2, 19.5, 25.4, 26.8,
30.7, 36.1, 38.4, 45.0, 55.8, 55.9, 62.2, 65.1, 66.2, 68.2, 70.9, 72.9,
73.4, 98.7, 110.8, 111.0, 120.1, 127.4, 127.5, 127.6, 128.9, 129.0,
129.2, 129.6, 129.9, 130.8, 133.7, 135.6, 136.0, 136.1, 136.2, 138.3,
138.5, 146.5, 148.9 ppm. MS (ES): m/z 1028 (M+NH4+).
C62H78O10Si (1010.5): calcd. C 73.63 H 7.78; found C 73.53, H
8.07.
Reduction of Carboxylic Acid 4 into Alcohol 5: See procedure (h) in
Scheme 1. To a cooled (0 °C) solution of carboxylic acid 4 (20.0 g,
0.087 mol) in dry tetrahydrofuran (120 mL) borane–dimethyl sul-
fide complex (11.6 mL, 0.13 mol) was added dropwise. The solution
was stirred at 0 °C for 2 h and at room temperature for 2 h. Water
was added slowly and tetrahydrofuran was removed under reduced
pressure. The aqueous phase was extracted with ethyl acetate. The
combined organic phases were dried with anhydrous magnesium
sulfate, filtered and concentrated. The residue was chromato-
graphed on silica gel (ethyl acetate/isooctane, 4:6) to afford alcohol
5 (16 g, 86%). M. p. 76 °C. Rf = 0.28 (isooctane/ethyl acetate, 6:4).
IR (KBr): ν = 3395, 3333, 2950, 1513, 1477, 1418, 1228, 1201, 1100,
˜
1044, 836, 766 cm–1. 1H NMR (500 MHz, CDCl3, 25 °C): δ = 2.87
(t, 2 H, J = 6.5 Hz), 3.86 (t, 2 H, J = 6.5 Hz), 4.49 (s, 2 H), 7.22
(m, 2 H), 7.35 (m, 2 H) ppm. 13C NMR/DEPT (50 MHz, CDCl3,
25 °C): δ = 33.4 (CH2), 38.8 (CH2), 63.4 (CH2), 129.2 (CH), 129.4
(CH), 135.9 (C), 138.9 (C) ppm. MS: m/z (%) = 216 (M+·Br81), 214
(M+·Br79), 135 (100), 105 (63), 104 (54), 91 (10), 77 (22), 63 (10),
51 (13).
Synthesis of Tripodal Scaffold Acid 1: See procedure (f) in
Scheme 1. To a solution of alcohol 3 (0.5 g, 0.50 mmol) in CH2Cl2
(35 mL), succinic anhydride (0.513 g, 4.95 mmol) and 4-(dimeth-
ylamino)pyridine (DMAP) (0.73 g, 5.93 mmol) were added. The
solution was stirred at room temperature for 4 h. The mixture was
extracted with an saturated aqueous solution of ammonium chlo-
ride. The organic phase was dried with anhydrous magnesium sul-
fate, filtered and evaporated under reduced pressure. The resulting
oil was then purified by column chromatography (isooctane/
Synthesis of Building Block 6: See procedure (i) in Scheme 1. Dihy-
dropyran (5.10 mL, 0.056 mol) and pyridinium p-toluenesulfonate
(1.25 g, 0.005 mol) were added to a solution of alcohol 5 (10 g,
0.047 mol) in dry dichloromethane (40 mL). The solution was
EtOAc, 6:4) to provide 0.5 g of acid 1 (90%). Rf = 0.34 (EtOAc/ stirred overnight at room temperature. The mixture was extracted
isooctane, 4:6). IR (neat): ν = 3400 (s), 2922 (s), 2853 (s), 1738 (w), with a saturated aqueous solution of sodium hydrogen carbonate
˜
4142
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146

Potential Serine Protease Mimics
FULL PAPER
and then with brine. The organic phase was dried with anhydrous
magnesium sulfate, filtered and evaporated under reduced pressure.
The residue was purified by column chromatography (isooctane/
ethyl acetate: 9:1) providing building block 6 (12.4 g, 88%). Rf =
tane, 1:9) to give building block 8 (15 g, 71%). Rf = 0.58 (diethyl
ether/isooctane, 1:9). IR (KBr): ν = 3047, 2930, 2857, 1742, 1427,
˜
1266, 1111, 823, 702, 505 cm–1. 1H NMR (500 MHz, CDCl3,
25 °C): δ = 1.04 (s, 9 H), 2.87 (t, 2 H, J = 6.8 Hz), 3.85 (t, 2 H, J
= 6.8 Hz), 4.59 (s, 2 H), 7.16 (2 H, AB-d, JAB = 8.0 Hz), 7.30 (2
H, AB-d, J = 8.0 Hz), 7.37 (m, 4 H), 7.42 (m, 2 H), 7.60 (dd, 4 H,
0.69 (toluene/ethyl acetate, 8:2). IR (KBr): ν = 2941, 2868, 1514,
˜
1439, 1352, 1229, 1200, 1120, 1077, 1031, 972, 907, 869, 814,
1
606 cm–1. H NMR (500 MHz, CDCl3, 25 °C): δ = 1.54 (m, 4 H), J = 6.6, 1.4 Hz) ppm. 13C NMR/DEPT (50 MHz, CDCl3, 25 °C):
1.69 (m, 1 H), 1.80 (m, 1 H), 2.91 (t, 2 H, J = 7.1 Hz), 3.45 (m, 1
δ = 19.1 (C), 26.8 (CH2), 38.9 (CH2), 46.9 (CH2), 64.9 (CH2), 127.6
H), 3.62 (dt, 1 H, J = 9.7, 7.2 Hz), 3.73 (m, 1 H), 3.94 (dt, 1 H, J (CH), 128.5 (CH), 129.6 (CH), 133.7 (C), 135.3 (C), 135.5 (CH),
= 9.7, 7.2 Hz), 4.48 (s, 2 H), 4.59 (m, 1 H), 7.22 (2 H, AB-d, JAB
= 8.0 Hz), 7.31 (2 H, AB-d, JAB = 8.0 Hz) ppm. 13C NMR/DEPT
(50 MHz, CDCl3, 25 °C): δ = 19.4 (CH2), 25.3 (CH2), 30.5 (CH2),
33.6 (CH2), 35.9 (CH2), 62.1 (CH2), 67.9 (CH2), 98.6 (CH), 128.9
(CH), 129.3 (CH), 135.5 (C), 139.5 (C) ppm. MS: m/z (%) = 299
(M+·Br81), 297 (M+·Br79), 117 (80), 85 (100), 67 (71), 41 (30).
139.6 (C) ppm. MS: m/z (%) = 351 (57), 247 (20), 217 (100), 181
(42), 117 (64), 91 (54), 57 (44).
Synthesis of Monopodal Member 13
Coupling of the Photocleavable Linker 10 onto TentaGel-NH2: Di-
isopropylcarbodiimide (DIC) (27 µL, 0.174 mmol) and 1-hydroxy-
benzotriazole (HOBt) (27 mg, 0.200 mmol) were added to a solu-
tion of linker 10 (90.5 mg, 0.174 mmol) in dry DMF (3 mL). The
mixture was stirred for 30 min at room temperature. After adding
the solution to the resin (300 mg, 0.21 mmol/g, 0.063 mmol), the
suspension was shaken overnight at room temperature. The resin
was washed with DMF (3×5 mL), MeOH (3×5 mL) and CH2Cl2
(3×5 mL). To monitor the completeness of the coupling reaction,
two colorimetric tests (TNBS and NF31) were performed and led
to colourless beads.
Synthesis of Building Block 7
Synthesis of Trichloroacetimidate 9: A solution of 3,4-dimethoxy-
benzyl alcohol (25 g, 0.149 mol) in tetrahydrofuran (30 mL) was
slowly added to a suspension of sodium hydride (60% dispersion
in oil) (1.19 g, 0.030 mol) in tetrahydrofuran (30 mL). The solution
was then cooled to 0 °C and trichloroacetonitrile (17.6 mL,
0.175 mol) was added dropwise. The mixture was stirred for 1 h
at 0 °C and 2 h at room temperature. Pentane (50 mL) containing
methanol (1 mL) was added followed by activated carbon. The mix-
ture was stirred for 1 h before being filtered through celite. The
celite was then washed with pentane. The organic phase was con-
centrated under reduced pressure affording the trichloroacetimid-
ate, which was used without further purification (45 g, 97%). Rf =
TNBS: To a few beads, 3 drops of a solution of 10% diisopropyl-
ethylamine (DIPEA) in DMF and 3 drops of trinitrobenzenesul-
fonic acid (TNBS) are added. The appearance of an orange colour
shows the presence of remaining uncoupled amino functions
whereas colourless beads indicate completeness of the coupling.
0.59 (toluene/ethyl acetate, 8:2). IR (KBr): ν = 3330, 2935, 1662,
˜
NF 31: A few beads are suspended in NF31 (100 µL, 0.002 )
in acetonitrile. After heating for 10 min at 70 °C, the beads are
thoroughly washed with DMF, MeOH and DCM. Beads contain-
ing free amino functions appear red, while completely coupled be-
ads remain colourless.
1593, 1518, 1463, 1420, 1267, 1240, 1160, 1079, 1028, 828, 796,
648 cm–1. 1H NMR (500 MHz, CDCl3, 25 °C): δ = 3.88 (s, 6 H),
5.28 (s, 2 H), 6.86 (d, 1 H, J = 8.1 Hz), 6.98 (m, 2 H), 8.37 (s, 1 H)
ppm. 13C NMR/DEPT (50 MHz, CDCl3, 25 °C): δ = 55.8 (CH3),
70.7 (CH2), 91.4 (C), 110.8 (CH), 111.1 (CH), 120.6 (CH), 127.8
(C), 148.9 (C), 149.0 (C), 162.4 (C) ppm.
General Procedure for the Deprotection of the Fmoc Group: The
resin was treated with a solution of 20% piperidine in DMF
(2×20 min). The resin was then washed with DMF (3×5 mL),
MeOH (3×5 mL) and CH2Cl2 (3×5 mL). Both colorimetric tests
were carried out to confirm deprotection.
Synthesis of Building Block 7: To a solution of alcohol 5 (6.5 g,
0.030 mol) and pyridinium p-toluenesulfonate (3.8 g, 0.015 mol) in
dichloromethane (48 mL) was added trichloroacetimidate 9 (18.8 g,
0.06 mol). The solution was stirred overnight at room temperature.
After addition of a saturated aqueous solution of sodium hydro-
gencarbonate, the mixture was extracted with dichloromethane,
dried with magnesium sulfate and filtered. The residue was purified
by column chromatography (20% ethyl acetate in toluene/pentane,
1:1) to afford building block 6 (7.6 g, 69%). Rf = 0.61 (toluene/
Synthesis of Construct 11: 1-[3-(Dimethylamino)propyl]-3-ethylcar-
bodiimide hydrochloride (EDC) (67 mg, 0.348 mmol) and 4-(di-
methylamino)pyridine (DMAP) (2 mg, 0.016 mmol) were added to
a solution of scaffold 1 (140 mg, 0.126 mmol) in CH2Cl2 (3 mL).
After activation of the acid for 30 min at room temperature, the
solution was added to the resin (300 mg, 0.060 mmol, 0.20 mmol/
g). The suspension was shaken overnight at room temperature. The
resin was washed with DMF (3×5 mL), MeOH (3×5 mL) and
CH2Cl2 (3×5 mL). Both colorimetric tests were carried out and
showed colourless beads. Gel-phase 13C NMR (50 MHz, 100 mg
resin suspended in CD2Cl2, 25 °C): δ = 20.3, 22.9, 26.3, 28.0, 30.4,
31.7, 31.8, 32.1, 33.6, 37.3, 40.3, 40.6, 45.8, 47.5, 63.3, 64.6, 65.0,
66.5, 71.5 (polyethylene glycol peak), 73.9, 74.5, 99.9, 110.6, 112.4,
112.5, 121.3, 128.9, 129.0, 129.3, 129.4, 130.2, 130.5, 130.8, 131.0,
131.1, 132.5, 135.0, 136.8, 137.0, 137.8, 139.8, 139.9, 141.6, 148.1,
148.2, 150.4, 155.2, 172.2, 173.3, 174.1 ppm.
ethyl acetate, 8:2). IR (KBr): ν = 2935, 2857, 1593, 1516, 1464,
˜
1419, 1264, 1237, 1157, 1094, 1028, 809, 607 cm–1. 1H NMR
(500 MHz, CDCl3): δ = 2.91 (2 H, J = 6.9 Hz), 3.66 (t, 2 H, J =
6.9 Hz), 3.84 (s, 3 H), 3.87 (s, 3 H), 4.45 (s, 2 H), 4.48 (s, 2 H), 6.82
(m, 3 H), 7.20 (2 H, AB-d, JAB = 8.0 Hz), 7.31 (2 H, AB-d, JAB
=
8.0 Hz) ppm. 13C NMR/DEPT (50 MHz, CDCl3): δ = 33.5 (CH2),
35.9 (CH2), 55.7 (CH3), 55.8 (CH3), 70.5 (CH2), 72.8 (CH2), 110.7
(CH), 120.0 (CH), 128.9 (CH), 129.3 (CH), 130.7 (C), 135.6 (C),
139.5 (C), 148.4 (C), 148.9 (C) ppm. MS: m/z (%) = 366 (M+·Br81),
364 (M+·Br79), 151 (100), 104 (18), 77 (12), 65 (8).
Synthesis of Building Block 8: tert-Butyldiphenylsilyl chloride
(25.6 g, 0.093 mol) and diisopropylethylamine (28.4 mL, 0.163 mol) Removal of the THP-Protecting Group: Pyridinium p-toluenesulfo-
were added to a solution of alcohol 5 (10 g, 0.047 mol) in dimethyl- nate (0.066 g, 0.265 mmol) was added to a suspension of resin 11
formamide (50 mL). The mixture was stirred at room temperature (0.300 g, 0.049 mmol, 0.163 mmol/g) in dry ethanol (3.2 mL). The
for 4 h. The mixture was added to ice and extracted with diethyl
ether. The organic phase was dried with anhydrous magnesium sul-
fate, filtered and concentrated under reduced pressure. The crude
oil was purified by column chromatography (diethyl ether/isooc-
mixture was heated overnight at 55 °C. After filtration, the resin
was washed with ethanol (3×5 mL), dimethylformamide
(3×5 mL), methanol (3×5 mL) and dichloromethane (3×5 mL).
A sample of resulting resin (2 mg, 0.33 µmol, 0.163 mmol/g) was
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4143

A. Madder et al.
FULL PAPER
suspended in 100 µL acetonitrile and was irradiated with UV light
(365 nm) for 3 h. The solution was analysed by ES-MS. ES-MS:
m/z = 1048 [M+Na]+ (see also Supporting Information, spectrum
1).
71.1, 70.2, 67.1, 66.8, 65.0, 57.9, 57.4, 47.7, 46.2, 40.9, 40.5, 37.6,
36.2, 34.0, 32.0, 28.3, 26.6, 24.4, 23.1, 20.6 ppm.
Characterisation of 13: A sample of resin 12 (34.2 mg, 0.005 mmol,
0.152 mmol/g) in suspension in 1,4-dioxane (1 mL) containing 1%
of dimethyl sulfoxide was irradiated with UV light for 4 times 2 h.
After each cleavage, the resin was filtered and washed with 1,4-
dioxane. After evaporation of 1,4-dioxane and removal of the di-
methyl sulfoxide in a kugelrohr, the sample was purified by column
chromatography (ethyl acetate) to provide 3.7 mg of the cleaved
Coupling of Fmoc-L-glycine: To a suspension of resin (0.30 g,
0.050 mmol, 0.165 mmol/g) in dry dimethylformamide (3 mL) were
added FmocGlyOH (125 mg, 0.42 mmol), 1-hydroxybenzotriazole
(56.7 mg,
0.42 mmol),
4-(dimethylamino)pyridine
(2.6 mg,
0.021 mmol), diisopropylcarbodiimide (67 µL, 0.42 mmol) and di-
isopropylethylamine (73 µL, 0.42 mmol). The mixture was shaken
overnight at room temperature. After filtration, the resin was
washed with dimethylformamide (3×5 mL), methanol (3×5 mL)
and dichloromethane (3×5 mL). The yield of the coupling was de-
termined by UV/Vis spectroscopy monitoring at 300 nm (102%)
(see also Supporting Information, ES-MS spectrum 2 of a represen-
tative sample upon cleavage). The Fmoc-protecting group was re-
moved according to the general procedure.
1
product 13 (45%). Rf = 0.53 (ethyl acetate). H NMR (500 MHz,
CD2Cl2, 25 °C): δ = 1.01 (s, 9 H), 1.88 (s, 3 H), 2.35 (2 H, AB-t, J
= 7.0 Hz), 2.50 (2 H, AB-t, J = 7.0 Hz), 2.84 (t, 2 H, J = 6.7 Hz),
2.87 (t, 2 H, J = 6.7 Hz), 2.93 (t, 2 H, J = 7.0 Hz), 2.93 (dd, 1 H,
J = 7.5, 14.0 Hz), 3.06 (dd, 1 H, J = 6.8, 14.0 Hz), 3.18 (dd, 1 H,
J = 5.1, 9.0 Hz), 3.49 (m, 6 H), 3.52 (m, 1 H), 3.65 (t, 2 H, J =
7.0 Hz), 3.78 (s, 3 H), 3.80 (s, 3 H), 3.83 (t, 2 H, J = 7.0 Hz), 3.84
(m, 1 H), 4.03 (1 H, AB-d, J = 7.78 Hz), 4.16 (s, 2 H), 4.32 (t, 2
H, J = 7.0 Hz), 4.41 (m, 8 H), 4.52 (dd, 1 H, J = 4.9 Hz), 4.58 (dd,
1 H, J = 7.2 Hz), 6.81 (m, 3 H), 7.27 (m, 38 H), 7.58 (dd, 4 H, J
= 1.4, 7.9 Hz) ppm. ES-MS: m/z = 1624 [M+Na]+, 1639 [M+K]+
(see Supporting Information, spectrum 5).
Coupling of FmocSer(OTrt)OH: FmocSer(OTrt)OH (0.144 g,
0.252 mmol) was activated with diisopropylcarbodiimide (40.0 µL,
0.252 mmol) and 1-hydroxybenzotriazole (34.0 mg, 0.252 mmol).
This solution was stirred for 30 min at room temperature and then
added to the resin (0.300 g, 0.049 mmol, 0.164 mmol/g). The mix-
ture was shaken overnight at room temperature. After filtration,
the resin was washed with dimethylformamide (3×5 mL), meth-
anol (3×5 mL) and dichloromethane (3×5 mL). A sample of the
resin was submitted to both colorimetric tests (TNBS and NF31)
and colourless beads were observed. A sample of the resulting resin
(2 mg, 0.320 µmol, 0.150 mmol/g) was suspended in 100 µL aceto-
nitrile and was irradiated with UV light (365 nm) for 3 h. The solu-
tion was analyzed by ES-MS. ES-MS: m/z = 1656 [M+Na]+, 1672
[M+K]+ (see also Supporting Information, ES-MS spectrum 3).
The Fmoc-protecting group was removed according to the general
procedure.
Synthesis of Dipodal Member 14
Removal of the DMPM Protecting Group: To a cooled (0 °C) sus-
pension of resin 12 (0.15 g, 0.023 mmol, 0.152 mmol/g) in a mixture
of dichloromethane/water (1.4 mL/0.1 mL) was added dichlorod-
icyanoquinone (14.3 mg, 0.063 mmol). The mixture was shaken
overnight at room temperature. After filtration, the resin was
washed with an aqueous saturated solution of sodium hydrogencar-
bonate (3×5 mL), phenol/methanol, 1:1 (3×5 mL), toluene
(3×5 mL), acetonitrile (3×5 mL), dimethylformamide (3×5 mL),
methanol (3×5 mL) and dichloromethane (3×5 mL). A sample of
the resulting resin (2 mg, 0.300 µmol, 0.15 mmol/g) was suspended
in acetonitrile (100 µL) and was irradiated with UV light (365 nm)
for 3 h. The solution was analyzed by ES-MS. ES-MS: m/z 1473
[M+Na]+ (see Supporting Information, spectrum 6).
Coupling of FmocPheOH: FmocPheOH (81.4 mg, 0.210 mmol) was
activated with diisopropylcarbodiimide (33.0 µL, 0.210 mmol) and
1-hydroxybenzotriazole (29.0 mg, 0.210 mmol). This solution was
stirred for 30 min at room temperature and then added to the resin
(0.30 g, 0.047 mmol, 0.155 mmol/g). The mixture was shaken over-
night at room temperature. After filtration, the resin was washed
with dimethylformamide (3×5 mL), methanol (3×5 mL) and
dichloromethane (3×5 mL). A sample of the resin was submitted
to both colorimetric tests (TNBS and NF31) and colourless beads
were observed. A sample of 2 mg of the resin (2 mg, 0.300 µmol,
0.147 mmol/g) was suspended in 100 µL acetonitrile and was irradi-
ated with UV light (365 nm) for 3 h. The solution was analyzed by
ES-MS. ES-MS: m/z = 1804 [M+Na]+, 1820 [M+K]+ (see Sup-
porting Information, spectrum 4).The Fmoc-protecting group was
removed according to the general procedure.
Synthesis and Capping of the Tripeptide Strand: The coupling of
FmocGlyOH (Supporting Information, spectrum 7), FmocH-
is(NMtt)OH (Supporting Information, spectrum 8) and FmocA-
laOH was performed as explained for the first peptide strand. After
final Fmoc deprotection the free amine was capped using AcIm as
described above.
Characterisation of 15: A sample of resin 14 (72 mg, 0.010 mmol,
0.142 mmol/g) in 1,4-dioxane (1 mL) containing 1% of dimethyl
sulfoxide was irradiated with UV light for 4 times 2 h. After each
cleavage, the resin was filtered and washed with 1,4-dioxane. After
evaporation of 1,4-dioxane and removal of the dimethyl sulfoxide
with the kugelrohr, the sample was purified by column chromatog-
raphy (dichloromethane/methanol) to provide the cleaved product
Capping of the Free Amine: Acetylimidazole (77 mg, 0.700 mmol)
15. R (dichloromethane/methanol) = 9:1. IR (KBr): ν = 3281, 2953,
˜
f
was added to
a
suspension of resin (0.250 g, 0.038 mmol,
2855, 1736, 1675, 1649, 1551, 1536, 1512, 1452, 1419, 1256, 1093,
1022, 799, 668 cm–1. 1H NMR (500 MHz, CD3OD, 25 °C): δ = 0.96
(s, 3 H), 1.27 (d, 3 H, J = 7.6 Hz), 1.83 (s, 3 H), 1.91(s, 3 H), 2.29
(s, 3 H), 2.40 (2 H, AB-t, J = 7.0 Hz), 2.47 (2 H, AB-t, J = 7.0 Hz),
2.83 (m, 7 H), 2.91 (dd, 1 H, J = 7.0, 14.8 Hz), 3.00 (dd, 1 H, J =
5.7, 14.0 Hz), 3.09 (dd, 1 H, J = 5.7, 14.0 Hz), 3.34 (dd, 1 H, J =
5.5, 9.3 Hz), 3.39 (1 H, J = 5.5, 9.3 Hz), 3.44 (s, 2 H), 3.44 (s, 2
0.152 mmol/g) in dry dichloromethane (3 mL). The mixture was
shaken overnight at room temperature. After filtration, the resin
was washed with dimethylformamide (3×5 mL), methanol
(3×5 mL) and dichloromethane (3×5 mL). A sample of the resin
12 was submitted to both colorimetric tests (TNBS and NF31) and
colourless beads were observed. IR (KBr): ν = 3572, 3278, 2860,
˜
1734, 1669, 1511, 1453, 1370, 1264, 1073, 947, 841, 698 cm–1. 13C H), 3.45 (s, 2 H), 3.79 (m, 4 H), 3.84 (1 H, AB-d, J = 17.8 Hz),
NMR (50 MHz, CD2Cl2, 25 °C): δ = 174.2, 173.3, 172.3, 171.4,
155.5, 150.0, 147,9, 145.1, 141.9, 140.2, 138.1, 137.1, 135.4, 131.2,
130.8, 130.4, 130.2, 129.9, 129.3, 129.0, 121.6, 112.8, 112.7, 111.0,
110.9, 88.3, 74.7, 74.2, 73.7, 72.1 (polyethylene glycol peak), 71.2,
3.93 (1 H, AB-d, J = 17.8 Hz), 4.13 (s, 2 H), 4.14 (t, 2 H, J =
7.1 Hz), 4.23 (m, 1 H), 4.24 (t, 2 H, J = 7.2 Hz), 4.35 (s, 2 H), 4.36
(s, 2 H), 4.39 (s, 2 H), 4.60 (m, 2 H), 4.65 (dd, 1 H, J = 6.1, 9.4 Hz),
6.76 (s, 1 H), 7.18 (m, 53 H), 7.51 (dd, 4 H, J = 1.3, 7.9 Hz) ppm.
4144
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146

Potential Serine Protease Mimics
In view of the limited mass range (m/z Ͻ 2000) of the available ES- mation spectrum 10, after Trt and Mtt removal), Fmoc-Val-OH
FULL PAPER
MS equipment mass spectrometric analysis of 15 and subsequent
products requires removal of the side-chain protection on serine
(O-trityl) and histidine (N-methyltrityl).
(see Supporting Information, spectrum 12, after Trt and Mtt re-
moval) and Fmoc-Asp(O2PhiPr)-OH was performed as explained
before. Resin 22b was then treated treated with a solution of chloro-
acetic acid (1.32 g) in a mixture dichloromethane/H2O (3:1, 1 mL)
overnight at room temperature. The procedure was repeated 3 times
within two hours. A sample of the resulting resin was suspended in
acetonitrile (100 µL) and was irradiated with UV light (365 nm)
for 3 h. The solution was analyzed by ES-MS. ES-MS: m/z = 1591
[M+H]+ (see Supporting Information, spectrum 15).
Therefore, a sample of resin 14 (10 mg, 1.4 µmol, 0.142 mmol/g)
was treated overnight at room temperature with a solution of chlo-
roacetic acid (1.32 g) in a mixture of dichloromethane/H2O (3:1, 1
mL). The procedure was repeated 3 times within two hours. A sam-
ple of the resulting resin (2 mg, 0.280 µmol, 0.140 mmol/g) was sus-
pended in acetonitrile (100 µL) and was irradiated with UV light
(365 nm) for 3 h. The solution was analysed by ES-MS. ES-MS:
m/z = 1538 [M+Na]+ (Supporting Information, spectrum 9).
Supporting Information (see also the footnote on the first page of
this article): Detailed Table with energy values for the Molecular
Modeling calculations related to Figure 1. Top view pictures of rel-
evant conformations. 2D NMR assignment at 500 MHz of the fully
protected tripodal member 17. ES-MS spectra of all intermediate
and final compounds after cleavage from solid support.
Synthesis of Tripodal Member 18
Removal of the TBDPS Protecting Group: To a suspension of resin
14 (150 mg, 0.021 mmol, 0.142 mmol/g) in dimethylformamide
(0.36 mL) was added TASF (0.0975 ) [tris(dimethylamino)sulfur
(trimethylsilyl)difluoride]
in
dimethylformamide
(0.82 mL,
Acknowledgments
0.084 mmol). The suspension was shaken for 3 h at room tempera-
ture. After filtration, the resin was washed with dimethylformamide
(3×5 mL), methanol (3×5 mL) and dichloromethane (3×5 mL).
Financial
assistance
from
GOA
(GOA96009),
TMR
(ERBFMRXCT960011), HPI (HPRN-CT-2000-00014) and the
FWO-Vlaanderen (KAN 1.5.186.03 and G.0347.04) is gratefully
acknowledged.
Synthesis of the Tripeptide Strand: The coupling of FmocGlyOH
(Supporting Information, spectrum 10, after Trt and Mtt removal),
FmocAsp[COO(CH2)SiMe3]OH and FmocValOH was performed
as explained for the first peptide strand. After final Fmoc deprotec-
tion the free amine was capped using AcIm as described above.
[1] a) Y. Murakami, J. Kikuchi, O. Hayashida, Chem. Rev. 1996,
96, 721–758; b) W. B. Motherwell, M. J. Bingham, Y. Six, Tet-
rahedron 2001, 57, 4663–4686.
Characterisation of 17: A sample of resin 16 (36.7 mg, 0.005 mmol,
0.141 mmol/g) in suspension of 1,4-dioxane (1 mL) containing 1%
of dimethyl sulfoxide was irradiated with UV light for 3 times 2 h.
After each cleavage, the resin was filtered and washed with 1,4-
dioxane. After evaporation of 1,4-dioxane and removal of the di-
methyl sulfoxide with the kugelrohr, the sample was purified by
column chromatography (ethyl acetate) to provide 5.6 mg of com-
pound 17 (51%). MW = 2188 g/mol, Rf = 0.52 (methanol/dichloro-
methane, 1:9). See Results and Discussion for 1H NMR characteri-
sation.
[2] A. J. Kirby, Angew. Chem. Int. Ed. Engl. 1996, 35, 707–724.
[3] a) J. M. Lehn, C. Sirlin, J. Chem. Soc., Chem. Commun. J.
Chem. Soc., Chem. Commun 1978, 949–951; b) D. J. Cram, P. Y.
Lam, P. S. Ho, J. Am. Chem. Soc. 1986, 108, 839–841; c) V. T.
D’Souza, M. L. Bender, Acc. Chem. Res. 1987, 20, 146–152; d)
R. Breslow, S. Chung, Tetrahedron Lett. 1989, 30, 4353–4356;
e) R. C. F. Jones, M. Tankart, A. M. Higton, Bioorg. Med.
Chem. Lett. 1991, 1, 353–356; f) L. G. Mackay, R. S. Wylie,
J. K. M. Sanders, J. Am. Chem. Soc. 1994, 116, 3141–3142; g)
R. Breslow, Acc. Chem. Res. 1995, 28, 146–153; h) B. S. Lele,
M. G. Kulkarni, R. A. Mashelkar, React. Funct. Polym. 1999,
39, 37–52.
[4] a) K. W. Hahn, W. A. Klis, J. M. Stewart, Science 1990, 248,
1544–1547; b) M. Z. Atassi, T. Manshouri, Proc. Natl. Acad.
Sci. USA 1993, 90, 8282–8286; c) B. Walse, M. Ullner, C. Lind-
bladh, L. Bulow, T. Drakenberg, O. Teleman, J. Comput.-Aided
Mol. Des. 1996, 10, 11–22; d) R. A. Buono, N. Kucharczyk,
M. Neuenschwander, J. Kemmik, L. Y. Hwang, J. L. Fauchere,
C. A. Venanzi, J. Comput.-Aided Mol. Des. 1996, 10, 213–232;
e) Y. Fukushima, Bull. Chem. Soc. Jpn. 1996, 69, 2269–2274;
f) A. Stavrakoudis, I. N. Demetropoulos, C. Sakarellos, M.
Sakarellos-Daitsiotis, V. Tsikeris, Lett. Pept. Sci. 1997, 4, 481–
487.
[5] M. L. Snapper, A. H. Hoveyda, Combinatorial Chemistry, prac-
tical approach, Oxford University Press, 2003, pp. 433–457.
[6] a) H. De Muynck, A. Madder, N. Farcy, J. P. De Clercq, M. N.
Pérez-Payàn, L. M. Öhberg, A. P. Davis, Angew. Chem. Int. Ed.
2000, 39, 145–148; b) A. Madder, L. Liu, H. De Muynck, N.
Farcy, D. Van Haver, F. Fant, G. Vanhoenacker, P. Sandra,
A. P. Davis, J. P. De Clercq, J. Comb. Chem. 2002, 4, 552–562;
c) G. Vanhoenacker, L. Liu, F. Lynen, A. Madder, P.
De Clercq, P. Sandra, J. Sep. Sci. 2002, 25, 671–676.
[7] N. Farcy, H. De Muynck, A. Madder, N. Hosten, J. P.
De Clercq, Org. Lett. 2001, 3, 4299–4301.
[8] a) M. Pátek, B. Drake, M. Lebl, Tetrahedron Lett. 1994, 35,
9169–9172; b) P. Kocis, O. Issakova, N. Sepetov, M. Lebl, Tet-
rahedron Lett. 1995, 36, 6623–6626; c) G. Wess, K. Bock, H.
Kleine, M. Kurz, W. Guba, H. Hemmerle, E. Lopez-Calla, K.-
H. Baringhaus, H. Glombik, A. Enhsen, W. Kramer, Angew.
Deprotection of O-Trityl (Ser) and N-Methyltrityl (His): A sample
of resin 16 (10 mg, 1.4 µmol, 0.140 mmol/g) was treated with a
solution of chloroacetic acid (1.32 g) in a mixture dichloromethane/
H2O (3:1, 1 mL) overnight at room temperature. The procedure
was repeated 3 times two hours. A sample of the resulting resin
18 (2 mg, 0.280 µmol, 0.140 mmol/g) was suspended in acetonitrile
(100 µL) and was irradiated with UV light (365 nm) for 3 h. The
solution was analyzed by ES-MS. ES-MS: m/z = 869 [M+2Na]2+
,
1713 [M+Na]+ (see Supporting Information, spectrum 11).
Synthesis of Tripodal Member 22a: The synthesis was performed as
described for 18 changing the order of introduction of amino acids
in the third strand to Fmoc-Gly-OH (Supporting Information spec-
trum 10, after Trt and Mtt removal), Fmoc-Val-OH (Supporting
Information spectrum 12, after Trt and Mtt removal) and Fmoc-
Asp(OTMSE)-OH. After final Fmoc deprotection the free amine
was capped using AcIm as described above. The Ser-OTrt and His-
NMtt protections were removed with a solution of chloroacetic
acid as described above for resin 16 (Supporting Information spec-
trum 13). Subsequent treatment with TASF clearly caused asparti-
mide formation. ES-MS analysis proved dehydration to be faster
than deprotection as evidenced from the signals at 1573 (M–
TMSE–H2O+H+) and 1691 (M+H+, still containing OTMSE)
(see Supporting Information, spectrum 14).
Synthesis of Tripodal Member 23: Resin 14 was deprotected as de-
scribed above. Coupling of Fmoc-Gly-OH (Supporting Infor-
Eur. J. Org. Chem. 2006, 4135–4146
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4145

A. Madder et al.
[17] D. Hudson, J. Org. Chem. 1988, 53, 617–624.
FULL PAPER
Chem. Int. Ed. Engl. 1996, 35, 2222–2224; d) T. Wunberg, C.
Kallus, T. Opatz, S. Henke, W. Schmidt, H. Kunz, Angew.
Chem. Int. Ed. Engl. 1998, 37, 2503–2505; e) J. F. Barry, A. P.
Davis, M. N. Pérez-Payan, Tetrahedron Lett. 1999, 40, 2849–
2852; f) C. Kallus, T. Opatz, T. Wunberg, W. Schmidt, S.
Henke, H. Kunz, Tetrahedron Lett. 1999, 40, 7783–7786; g) M.
Pattarawarapan, K. Burgess, Angew. Chem. Int. Ed. 2000, 39,
4299–4302; h) T. Opatz, R. M. J. Liskamp, Org. Lett. 2001, 3,
3499–3502; i) C. Chamorro, R. M. J. Liskamp, J. Comb. Chem.
2003, 5, 794–801; j) For other examples of flexible scaffolds,
see: P. Virta, M. Leppänen, H. Lönnberg, J. Org. Chem. 2004,
69, 2008–2016.
[18] K. Barlos, D. Gatos, S. Koutsogianni, W. Schäfer, G. Stavro-
poulos, Y. Wenging, Tetrahedron Lett. 1991, 32, 471–474.
[19] H. P. Nestler, Mol. Div. 1996, 2, 35–40.
[20] J. C. Mc Auliffe, O. Hindsgaul, Synlett 1998, 307–309.
[21] W. Zhang, M. J. Robins, Tetrahedron Lett. 1992, 33, 1177–1180.
[22] K. A. Scheidt, H. Chen, B. C. Followsn, S. R. Chemler, D. S.
Coffey, W. R. Roush, J. Org. Chem. 1998, 63, 6436–6437.
[23] T. W. Greene, P. G. M. Wuts, Protective groups, in: Organic
Synthesis, 3rd ed., 1999.
[24] a) P. Sieber, Helv. Chim. Acta. 1977, 60, 2711–2716; b) C. K.
Marlowe, Bioorg. Med. Chem. Lett. 1993, 3, 437–440.
[25] K. Wüthrich, NMR of Proteins and Nucleic Acids, Wiley, New
York, USA, 1986.
[26] a) M. A. Ondetti, A. Deer, J. T. Sheehan, J. Plusˇcˇec, O. Kocy,
Biochemistry 1968, 7, 4069–4075; b) M. Bodanszky, S. Natara-
jan, J. Org. Chem. 1975, 40, 2495–2499.
[9] A full conformational analysis was performed using Macro-
Model V6.0, force field MM2*. To reduce calculation times the
carboxylic acid linker and the protected hydroxyethyl chains
were simplified as methylgroups.
[10] a) THP: N. Miyashita, A. Yoshikoshi, P. A. Grieco, J. Org.
Chem. 1977, 42, 3772–3774; b) DMPM: N. Nakajima, M. [27] a) E. Nicolás, E. Pedroso, E. Giralt, Tetrahedron Lett. 1989, 30,
Saito, M. Ubukata, Tetrahedron Lett. 1988, 29, 5565–5568; c)
TBDPS: Y. Guindon, Y. St Denis, S. Daigneau, H. E. Morton,
Tetrahedron Lett. 1986, 27, 1237–1240.
497–500; b) Y. Okada, S. Iguchi, J. Chem. Soc., Perkin Trans. 1
1988, 2129–2136.
[28] a) R. Dölling, M. Beyermann, J. Haenel, F. Kernchen, E.
Krause, P. Franke, M. Brudel, M. Bienert, J. Chem. Soc.,
Chem. Commun. 1994, 853–854; b) M. Quibell, D. Owen, L. C.
Packman, T. Johnson, J. Chem. Soc., Chem. Commun. 1994,
2343–2344; c) M. Mergler, F. Dick, P. Weiler, T. Vorherr, J.
Pept. Sci. 2003, 9, 36–46; d) M. Mergler, F. Dick, B. Sax, C.
Stähelin, T. Vorherr, J. Pept. Sci. 2003, 9, 518–526.
[29] a) T. Opatz, R. M. J. Liskamp, J. Comb. Chem. 2002, 4, 275–
284; b) C. Chamorro, R. M. J. Liskamp, J. Comb. Chem. 2003,
5, 794–801.
[11] a) C. P. Holmes, D. G. Jones, J. Org. Chem. 1995, 60, 2318–
2319; b) S. J. Teague, Tetrahedron Lett. 1996, 37, 5751–5754; c)
C. P. Holmes, J. Org. Chem. 1997, 62, 2370–2380.
[12] H. B. Lee, S. Balasubramanian, J. Org. Chem. 1999, 64, 3454–
3460.
[13] a) J. Meienhofer, M. Waki, E. P. Heimer, T. J. Lambros, R. C.
Makofske, C.-D. Chang, Int. J. Prot. Res. 1979, 13, 35–42; b)
C.-D. Chang, M. Waki, M. Ahmad, J. Meienhofer, E. O. Lund-
ell, J. D. Haug, Int. J. Prot. Res. 1980, 15, 59–66.
[14] J. Coste, D. Le-Nguyen, B. Castro, Tetrahedron Lett. 1990, 31,
205–208.
[30] C. Yue, Tetrahedron Lett. 1993, 34, 323–326.
[31] H. Desvaux, P. Berthault, N. Birlirakis, M. Goldman, J. Magn.
Reson. A 1994, 108, 219–229.
[15] W. S. Hancock, J. E. Battersby, Anal. Biochem. 1976, 71, 260–
264.
Received: February 20, 2006
[16] A. Madder, N. Farcy, N. G. C. Hosten, H. De Muynck, P. J.
De Clercq, J. Barry, A. P. Davis, Eur. J. Org. Chem. 1999, 2787–
2791.
Published Online: July 21, 2006
4146
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 4135–4146