Novel cyclic phosphate-linked oligosaccharides (CyPLOSs) covalently immobilized on solid supports for potential cation scavenging

Article abstract of DOI:10.1002/ejoc.200700203

For potential cation scavenging both from water and from organic solvents, here we propose a synthetic procedure for functionalization of a Tentagel solid support with novel cyclic phosphate-linked oligosaccharide (CyPLOS) analogues. To establish the feas

Full text of DOI:10.1002/ejoc.200700203

FULL PAPER
DOI: 10.1002/ejoc.200700203
Novel Cyclic Phosphate-Linked Oligosaccharides (CyPLOSs) Covalently
Immobilized on Solid Supports for Potential Cation Scavenging
Jennifer D’Onofrio,[a] Cinzia Coppola,[a] Giovanni Di Fabio,[a] Lorenzo De Napoli,[a] and
Daniela Montesarchio*[a]
Keywords: Carbohydrates / Cyclic oligosaccharide analogues / Solid-phase synthesis / Protecting groups / Lipophilicity /
31P NMR spectroscopy
For potential cation scavenging both from water and from
organic solvents, here we propose a synthetic procedure for
functionalization of a Tentagel solid support with novel cyclic
phosphate-linked oligosaccharide (CyPLOS) analogues. To
establish the feasibility of the synthetic strategy, the cyclic
dimer was the model compound selected to be incorporated
onto the solid support. This functionalization was achieved
through a stepwise solid-phase synthesis of the linear dimer,
obtained by standard phosphoramidite protocols, followed by
a synthesis of the cyclic molecule on the resin. The key inter-
mediate in our synthetic strategy was a suitably derivatized
sugar phosphoramidite building block, with the secondary
hydroxy functions masked as TBDMS ethers. This proved to
be an orthogonal protection with respect to the DMT ether,
fully compatible with the phosphoramidite and the phos-
photriester chemistry used for the oligomerization and the
cyclization process onto the solid support, respectively. Con-
ditions for the total unmasking of the hydroxy groups of the
cyclic dimer, not affecting the integrity of the cyclic structure
nor its linkage with the solid matrix, have been optimized.
Gel-phase 31P NMR spectroscopy has been used extensively
here to monitor the efficiency of the reactions carried out on
the solid support.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
logical solutions and also for the transport of radioactive
metals for diagnostic or therapeutic purposes.[5] Insertion of
single carboxylate functions on these modified cyclodex-
trins has resulted in enhanced affinity towards heavy metals,
suggesting that immobilization of these macrocycles on
suitable insoluble supports may produce new materials use-
ful for the elimination or concentration of toxic cations
from aqueous solutions and especially from biological
fluids.[6]
Upon suitable modification, cyclodextrins can also be
converted into lipophilic tools, thus extending their use also
to organic solvents. Particularly interesting are amphiphilic
cyclodextrins, displaying acceptable solubility in a wide
range of solvents.[7] Typical examples of amphiphilic cyclo-
dextrins reported in the literature include preformed oligo-
saccharide macrocycles with covalently attached hydro-
phobic moieties such as cholesterol,[8] or cyclodextrins per-
alkylated at their primary or secondary faces with long ali-
phatic chains, connected through stable ether linkages.[9]
The preparation of modified cyclic oligosaccharides
starting from preformed cyclodextrins, which is still the
most commonly followed route, has a valid, though more
laborious, alternative in the total synthesis approach. This
involves the stepwise synthesis of the linear oligomer, fol-
lowed by appropriate circularization, and points to a higher
degree of molecular diversity in the construction of the
macrocycle, allowing, for instance, the insertion of com-
Introduction
Cyclodextrins exhibit remarkable inclusion properties
towards a large number of chemically different compounds,
due to their characteristic toroidal structure, and for this
reason are among the most studied macrocyclic systems in
supramolecular chemistry.[1] A notable advantage of cyclo-
dextrins over other interesting host molecules such as, for
instance, calyx[4]arenes, is their intrinsic high water solubil-
ity, in conjunction with the hydrophobic nature of the in-
ternal cavity. This allows specific recognition in aqueous
solutions, which is of the utmost importance in mimicking
biological processes or in the development of novel biosen-
sors. Being polyfunctional molecules, cyclodextrins are use-
ful scaffolds that, suitably elaborated, can provide artificial
receptors capable of specifically binding a variety of dif-
ferent ions or neutral molecules in water.[2] Additional
interest in cyclodextrins arises from their potential as ef-
ficient scavengers of environmentally hazardous com-
pounds from water.[3] Among other uses, per(3,6-anhydro)-
cyclodextrins have emerged as good complexing agents for
metals,[4] with potential applications in cleaning of bio-
[a] Dipartimento di Chimica Organica e Biochimica, Università
degli Studi di Napoli “Federico II”, Complesso Universitario
di Monte S. Angelo,
via Cintia, 4, 80126 Napoli, Italy
Fax: +39-081-674393
E-mail: daniela.montesarchio@unina.it
Eur. J. Org. Chem. 2007, 3849–3858
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3849

D. Montesarchio et al.
FULL PAPER
pletely unusual monomeric residues[10] or the replacement
of the natural O-(1,4)-glycosidic linkages with alternative,
more stable chemical bonds.[11]
Aiming at specific cation recognition, we recently de-
scribed novel cyclic oligosaccharides, 4,6-linked through
chemically and enzymatically stable phosphodiester bonds
(1–3, Figure 1), that we have named CyPLOSs (Cyclic
Phosphate-Linked OligoSaccharide analogues).[12] These
molecules were synthesized starting from an appropriate 4-
phosphoramidite derivative of phenyl-β--glucopyranoside
(4, Figure 1). The assembly of the linear precursor was car-
ried out on a DNA synthesizer by standard phosphoramid-
ite protocols. The solid support used was engineered to al-
low an on-resin circularization procedure, which upon a
mild basic treatment released exclusively the cyclic molecule
into solution.[13] The systematic evaluation of the cation-
binding abilities of these cyclic oligomers is currently un-
Scheme 1. General synthetic platform for the functionalization of
Tentagel resin with cyclic disaccharide analogues.
Results and Discussion
Here we propose a simple and versatile protocol for the
derway and will be published in due course. It can be noted functionalization of standard Tentagel supports with novel
here that the cation-scavenging potential of this class of cy- cyclic CyPLOS analogues, based on the initial construction
clic saccharide surrogates, combining some constitutive ele- of the linear counterpart and subsequent on-resin cycliza-
ments of small cyclodextrins and crown ethers, and exhibit- tion. The crucial issue in our synthetic plan was the suitable
ing phosphodiester bonds within the oligosaccharide core selection of the solid support, of the linker and of the func-
as a distinct structural motif, should a priori be widely mo- tional group protection strategy.
dulatable by ad hoc selection of the nature, stereochemistry
and number of the monomeric building blocks.
Tentagel resin (0.29 mequiv. per g of primary amino
groups), widely adopted for the solid-phase synthesis of oli-
gonucleotides, peptides and related analogues, was the sup-
port of choice for its compatibility with standard auto-
mated phosphoramidite protocols and also for its suitability
for applications in water, in view of direct tests of resin-
bound compounds in aqueous environments. In addition,
given the high flexibility of PEG chains attached to the
polystyrene backbone on Tentagel support, it is generally
assumed that the conformational mobility of substrates im-
mobilized onto this copolymer is not substantially altered,
so their complexation abilities should also not be ham-
pered. The bifunctional linker 2-(3-chloro-4-hydroxy-
phenyl)acetic acid was directly introduced onto the solid
support, thus affording a phenolic OH resin, useful for the
covalent attachment of a tailored phosphoramidite saccha-
Figure 1. Dimeric, trimeric and tetrameric CyPLOS (1–3) and
building block 4.
Our goal is indeed to develop a general synthetic plat- ride derivative. In this synthetic context, the previously de-
form for the preparation of solid supports functionalized scribed building block 4 (Figure 1) was of no use, having its
with lipophilic cyclic glycomimetics that, upon a single 2- and 3-OH groups protected in the form of benzoic esters,
treatment, will be easily convertible into highly hydrophilic typically removed under basic conditions, under which
macrocycles, allowing us to sequester cations from both phosphotriester functions would definitely not survive.
aqueous and organic solutions. As a model for potential Therefore, an alternative protection strategy was necessary
cation scavenging, here we describe the synthesis of resin- in our synthetic scheme in order to achieve fully depro-
bound cyclic phosphate-linked disaccharide compounds tected cyclic molecules still anchored to the solid support
with their secondary hydroxy functions decorated with (IV, Scheme 1).
bulky TBDMS groups. This solid matrix, as depicted in
In the literature, the TBDMS group has been widely used
Scheme 1, can switch from an “off-recognition” state, an- to mask the primary hydroxy groups of cyclodextrins,[14]
choring a predominantly lipophilic system, with all the four conferring a high degree of lipophilicity on the final oligo-
hydroxy groups per grafted cycle blocked in the form of saccharides. Here this protecting group was chosen as the
silyl ethers, to an “on-recognition” state, exposing highly appendage to cap the glucoside 2- and 3-hydroxy functions
hydrophilic macrocycles to the solvent. To our convenience, in I in a transient manner, since it satisfies the following
both kinds of cyclic derivatives, the TBDMS-protected or prerequisites: i) it can easily be installed, by standard and
the fully deprotected compounds, can be detached from the high-fidelity reactions, ii) it is orthogonal both to 4,4Ј-di-
solid matrix and recovered in a pure form after simple gel methoxytriphenylmethyl (DMT) and 2-cyanoethyl groups,
filtration chromatography.
typically used for the transient protection of primary hy-
3850
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3849–3858

Novel Immobilized Oligosaccharides
FULL PAPER
droxy and phosphate diester groups, respectively, iii) it can
be selectively detached by a standard fluoride treatment,
expected not to affect the integrity of the circular structure
nor the stability of the phenyl glucoside moieties, iv) it is
highly hydrophobic, able to affect the solubility properties
of the cyclic molecule in organic media significantly, and
v) it is stable to the mild basic treatments required to detach
the phosphodiester-linked cyclic molecule, in case its release
in solution is desired. For our specific purposes, we opti-
mized the conditions for its complete removal in the solid
phase while not affecting the more labile phosphotriester
function linking the cyclic molecule to the solid matrix. As
a further advantage, in the case of experiments carried out
in solution, the treatment with fluoride for TBDMS re-
moval does not involve troublesome purification steps; in
fact, due to the net increase in hydrophilicity, the resulting
macrocycle is typically water-soluble, so that either washing
with organic solvents or a simple gel filtration chromatog-
raphy allow the target molecule to be easily isolated from
the reaction mixture in a very pure form.
Scheme 2. a) α,α-Dimethoxytoluene, PTSA cat., DMF, 12 h, 50 °C
(100%); b) TBDMSCl, imidazole 48 h, DMF, room temp. (87%);
c) TFA/DCM/H2O (1:10:0.5, v/v/v), 4 h, 0 °C (82%); d) DMTCl,
pyridine, 12 h, room temp. (92%); e) 2-cyanoethyl N,N-diisopro-
pylchlorophosphoramidite, DIEA, DCM, 1 h, room temp. (90%).
Phosphoramidite 9, obtained in five easy and high-yield-
ing reaction steps depicted in Scheme 2, was therefore cho-
sen as the key intermediate in our synthetic strategy. This
building block shows the following features: 1) anomeric phoramidite chemistry, involving tetrazole activation, oxi-
position blocked in the form of a phenyl glucoside, stable dation and capping, to afford support 12 with a loading of
to the various treatments required in the synthesis and fur- 0.16 mequiv.g–1, as calculated by the DMT test. Success-
nishing a UV/Vis tag attached to the sugar building block, ively, treatment with triethylamine in pyridine allowed the
2) 6-OH protected in the form of a DMT ether, 3) 4-OH removal, by a β-elimination mechanism, of the 2-cyanoethyl
derivatized with 2-cyanoethyl N,N-diisopropylchlorophos- group, thus converting the phosphotriester into the more
phoramidite, and 4) protected at both the 2- and the 3-posi- stable phosphodiester function and furnishing support 13.
tions as a TBDMS ether. Phosphoramidite 9 was prepared
From resin 13, the oligosaccharide chain was grown by
starting from commercially available phenyl β--glucopyr- exploiting a standard phosphoramidite protocol, well opti-
anoside (5), which was first converted in almost quantita- mized for the solid-phase synthesis of oligonucleotides.[16]
tive yield into the corresponding 4,6-benzylidene derivative, The synthetic procedure involved the following steps: a) re-
by treatment with α,α-dimethoxytoluene in the presence of moval of the DMT group from the 6-OH function on func-
catalytic amounts of PTSA. Next, the hydroxy functions at tionalized support 13 by treatment with DCA in DCM
the 2- and 3-positions were treated with TBDMSCl in (2%), b) coupling with the addition monomer 9, in the pres-
DMF in the presence of imidazole, thus furnishing deriva- ence of tetrazole, c) oxidation of the phosphite triester to
tive 6 in 87% yield.
phosphate triester, by treatment with an iodine solution
After removal of the benzylidene group, achieved in 82% (0.1 ) in THF/H2O/pyridine to give the support 14, d) cap-
yield by treatment with TFA in DCM, the 6-OH of 7 was ping with acetic anhydride in pyridine to block unreacted
selectively protected in the form of an acid-labile DMT hydroxy functions, and e) removal of the DMT group from
ether by addition of DMTCl in pyridine, yielding derivative the 6-OH function of the second added monomer, to afford
8 in 92% yield. Successive phosphitylation at the 4-OH po- support 15. The efficiency of the coupling reaction was
sition by treatment with 2-cyanoethyl N,N-diisopropylchlo- evaluated by spectrophotometric DMT tests, in all cases
rophosphoramidite in the presence of DIEA gave target showing incorporation yields very close to 100%. Once the
compound 9 in 90% yield. The overall yield for the synthe- linear precursor on support 15 had been achieved, the fol-
sis of 9 starting from compound 5 was 60% for the five lowing cyclization step was carried out by exploiting classi-
steps. In all cases, the intermediate compounds and final cal phosphotriester chemistry protocols in the solid phase.
derivative 9 were purified by silica gel chromatography and The linear oligomer was treated with the condensing agent
were then fully characterized by NMR (1H, 13C and also MSNT in pyridine, to afford support 16. This was then
31P in the case of 9) and ESI-MS data.
treated with triethylamine, to induce selective removal of
The functionalization of the solid support was achieved the 2-cyanoethyl group from the phosphodiester intergluco-
by treatment of building block 9 with TentaGel-NH2 resin side linkages, giving support 17. All the crucial steps on the
(0.29 mequiv.g–1) previously derivatized with 2-(3-chloro-4- solid phase, from the functionalization of the support to the
hydroxyphenyl)acetic acid (10), in the presence of DIC, cyclization and deprotection of the dimer were monitored:
HOBt and DIEA, as shown in Scheme 3.[15] The first sugar i) directly on the solid phase, by 31P NMR spectroscopy
building block was anchored to resin 11 by standard phos- performed on the resin suspended in CDCl3, and ii) indi-
Eur. J. Org. Chem. 2007, 3849–3858
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3851

D. Montesarchio et al.
FULL PAPER
Scheme 3. a) DIC, HOBt, DIEA, pyridine, 12 h, room temp.; b) Ac2O/pyridine (1:1, v/v), 1 h, room temp.; c) 18  NH4OH, 1 h, 50 °C;
d) coupling with 9, 0.45  tetrazole in CH3CN, 1 h, room temp.; e) 0.1  I2 in THF/H2O/pyridine, 3 treatments, 5 min each, room temp.;
f) Et3N/pyridine (1:1, v/v), 3 treatments, 6 h each, room temp.; g) 2% DCA in DCM, 5 treatments, 3 min each, room temp.; h) MSNT,
pyridine, 3 treatments, 6 h each, room temp.; i) Et3N·3HF/THF (3:1, v/v), 12 h, 50 °C; l) 0.2  1,1,3,3-tetramethylguanidinium 2-ni-
trobenzaldoximate in H2O/dioxane (1:1, v/v), 12 h, room temp.
rectly, by detachment, after each step, of samples of the the saccharide chain in 14, the following cyclization to give
glycomimetics grown onto the solid support and analysis of 16 and phosphate deprotection to give 17) were directly fol-
the released material by 1H and 31P NMR spectroscopy and lowed in the solid phase by 31P NMR spectroscopy. Typi-
ESI-MS spectrometry.
cally, on oxidation of the PIII atoms to the PV state, a large
Accordingly, all the reactions involving the phosphorus upfield shift is observed, with the two signals centred at δ
centre (i.e., the oxidation of the phosphite triester to phos- = 140 ppm coalescing into a broad resonance shifted at
phate triester in 12, the subsequent deprotection to give the –4 ppm. This signal, upon removal of the 2-cyanoethyl
desired phosphate diester in 13, as well as the growing of group, is characteristically shifted downfield by ca. 2 ppm.
3852
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3849–3858

Novel Immobilized Oligosaccharides
FULL PAPER
An example of 31P NMR spectroscopic monitoring of the
2-cyanoethyl removal from phosphodiester function to ob-
tain support 13 is shown in Figure 2.
Critically, the efficiency of the cyclization reactions in
giving functionalized support 16 was ascertained through
the following data: a) total disappearance of the signal at δ
= –2.52 ppm in the 31P NMR spectrum, attributed to the
phosphodiester function in support 15, with the emergence
of a new, broad signal at δ = –4.17 ppm, consistent with the
desired phosphotriester function (see Figure 3), and b) re-
covery, after detachment with benzaldoximate ions, of the
fully protected cyclic dimer 19, in a pure form (see Figure 4)
and in the expected amounts. By combining these data, it
was concluded that the cyclization yields with the solid-
phase method were in all cases (within the detection limits
of the NMR technique) close, if not superior, to 95%. The
detachment of cyclic dimer 19 from the insoluble matrix
was achieved by treatment of solid support 17 with a
1,1,3,3-tetramethylguanidinium
2-nitrobenzaldoximate
solution in H2O/dioxane (0.2 ), selectively cleaving the 2-
chlorophenate group from a dialkyl-phosphodiester func-
tion. Following this procedure, the cyclic compound is re-
leased into solution exclusively, whereas the unreacted lin-
Figure 2. 31P NMR spectra (161.98 MHz) of resin 12 suspended in
CDCl3 (panel A), after the first 6 h treatment with Et3N/pyridine
(panel B), and after the third and last 6 h treatment with Et3N/
pyridine (panel C), showing the complete conversion of resin 12
into 13.
Figure 3. 31P NMR spectrum (161.98 MHz) of resin 16 suspended
in CDCl3.
Figure 4. Low-field region of the 1H NMR spectrum (400 MHz, CDCl3) of cyclic dimer 19. In the inset, its 31P NMR spectrum
(161.98 MHz, CDCl3).
Eur. J. Org. Chem. 2007, 3849–3858
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3853

D. Montesarchio et al.
FULL PAPER
ear molecules or polymeric chains formed in the solid phase
From the perspective of developing resins for cation ex-
by undesired intramolecular reactions – if any – are not change that can be kept either in the “off-recognition” (i.e.,
affected by this treatment and therefore remain on the solid in the protected form) or in the “on-recognition” (i.e., in
support.[13]
a totally deprotected form) states, the TBDMS ether can
Once the cyclization conditions had been optimized, spe- therefore also be regarded as a useful protection system for
cial attention was devoted to the desilylation reaction on secondary hydroxy groups in the solid phase. In principle,
solid support 17. The complete deprotection of the hydroxy the liberated OH groups in support 18 are easily amenable
groups – under conditions not affecting the stability of the to subsequent reprotection, either with TBDMS or with
phosphotriester bond linking the macrocycle to the solid other protecting groups, if required, so to yield a switchable
resin – was achieved by addition of fluoride; particularly, motif, crucial for potential cation recognition.
the Et3N·3HF complex was found to be preferable to the
Further studies are now in progress to evaluate the po-
more aggressive TBAF, NH4F or BF3·Et2O, though their tential of these Tentagel solid supports functionalized with
use is widely documented for TBDMS removal in per-O- saccharide-based macrocycles for analytical applications in
silylated cyclodextrins. Several tests were carried out to op- different solvent systems or also at the interface between
timize this reaction, by varying the excess, solvent and tem- water and immiscible organic solvents, in analogy with re-
perature conditions. Checking for optimal conditions was cent investigations carried out on ion-exchange resins,[17]
carried out: 1) once the desilylation procedure had been chemically modified silica gels[18] or polystyrene.[19] Appli-
completed, by detaching the glycomimetics from weighed cations to chiral discrimination are also being examined.
samples of the solid support by treatment with benzald-
oximate ions and evaluating, by 1H NMR and ESI-MS
analysis of the material released in solution, the effective
deprotection, and 2) during the desilylation treatments, by
Conclusions
direct analysis, through ESI-MS techniques, of the com-
In principle, grafting of a specific host molecule onto a
bined eluates and washings of the solid support obtained solid support may significantly affect its behaviour and
for each Et3N·3HF treatment, so to ensure that no precious binding properties. However, given the specific features of
dimeric material is cleaved, even in traces, as an undesired Tentagel resin, a polystyrene matrix copolymerized with
side reaction of the desilylation procedure. Full removal of highly flexible and hydrophilic PEG chains, macrocycles
the TBDMS groups in the solid phase, without concomi- immobilized on such a solid support can be assumed to
tant detrimental loss of functionalization on the support, exhibit almost unaltered complexation abilities. Within this
thus affording the target solid support 18, was finally scenario, we have developed a synthetic protocol to obtain
achieved by adopting Et3N·3HF in anhydrous THF (1:1, Tentagel resins functionalized with novel cyclic oligosaccha-
v/v) and carrying out the reaction at 50 °C for 12 h. In all ride analogues (CyPLOS) bearing phosphodiester linkages
our experiments, the treatment with Et3N·3HF at 50 °C, as the interglycosidic bonds, here proposed as synthetic
even if prolonged for 48 h on the functionalized resin, did platforms for potential cation scavenging. Stepwise solid-
not lead to any detectable side reaction; particularly, no de- phase synthesis was adopted here to build the macrocycle,
tachment of saccharidic material from the Tentagel support by growing the linear dimer first and then cyclizing it onto
was observed. Interestingly, only the experiments carried the solid support. The key point in our synthetic strategy
out at high temperature proved to be successful. On the was the selection of suitably derivatized sugar phosphoram-
contrary, when the reactions were carried out at room tem- idite building blocks, with the secondary hydroxy functions
perature, only partial TBDMS deprotection occurred, as masked as TBDMS ethers. This proved to be an orthogonal
shown by the presence in the eluates, after treatment with protection with respect to the DMT ether, fully compatible
oximate ions on the solid phase, of all the possible TBDMS with phosphoramidite and phosphotriester chemistry, used
derivatives of cyclic dimer 1, bearing from zero to four for the oligomerization and the cyclization process onto the
TBDMS groups, in different ratios.
solid support, respectively. Gel-phase 31P NMR spec-
The possibility of releasing both kinds of cyclic oligosac- troscopy has been extensively used here to monitor the effi-
charide derivatives (i.e., lipophilic 19 and the fully depro- ciency of the reactions carried out on the solid support.
tected, hydrophilic 1) from the solid support was then inves- Interestingly, after the crucial cyclization reaction, 31P
tigated. Starting from support 18, an overnight treatment NMR signals diagnostic of the PV atom in an aryl-phos-
with oximate ions at room temp. provided cyclic dimer 1, phodiester function (centered at ca. –2 ppm) were absent,
which proved to be identical to an authentic, previously with the concomitant appearance of two new signals at δ ≈
synthesized sample, in the expected high purity and yields. –4 ppm, attributable to a chiral PV atom in an aryl-phos-
Similar results were obtained when resin 17 was exposed to photriester bond. The obtained results point at nearly
the same treatment, yielding the lipophilic cyclic oligosac- quantitative yields for the circularization of the linear disac-
charide derivative 19 as the sole compound, as ascertained charide mimic, within the detection limit of the NMR tech-
by RP-HPLC analysis of the detached material. The latter nique. It was thus demonstrated that the presence of bulky,
experiments further demonstrated the complete orthogo- lipophilic appendages on the secondary 2- and 3-OH
nality of aryl-phosphotriester functions with respect to groups of the glucose monomer had no adverse effects on
TBDMS ether groups (Figure 4).
the final outcome of the crucial cyclization step. Sub-
3854
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3849–3858

Novel Immobilized Oligosaccharides
FULL PAPER
DMF (10 mL) and treated with tert-butyldimethylsilyl chloride
(3.0 g, 5 equiv.) and imidazole (1.8 g, 6.8 equiv.). The mixture was
left whilst stirring at room temp. and monitored by TLC with
CHCl3/CH3OH 98:2 (v/v) as the eluent system. After 48 h the reac-
tion mixture was diluted with CHCl3, the organic phase was
washed three times with NaHCO3 (5%), filtered and concentrated
under reduced pressure. The resulting crude product was purified
on a silica gel column eluted with benzene/petroleum ether 9:1
(v/v), which gave 1.98 g of target compound 6 (87% yields).
sequently, optimization for the full deprotection of the
macrocycle in the solid phase was carried out, to provide
the desired solid support, functionalized with the cyclic oli-
gosaccharide analogues. Where desired, either the fully de-
protected or the TBDMS-protected cyclic molecules can be
detached from the solid support.
Experimental Section
Compound 6: White amorphous solid. Rf = 0.8 [benzene/AcOEt
1
98:2 (v/v)]. H NMR (CDCl3, 300 MHz): δ = 7.46–6.97 (complex
Abbreviations
signals, 10 H, aromatic H), 5.44 (s, 1 H, Ph–CH), 5.12 (d, J =
9.5 Hz, 1 H, 1-H), 4.31 (dd, J = 3.0 and 12 Hz, 1 H, 4-H), 3.92–
3.55 (overlapped signals, 5 H, 2-H, 3-H, 5-H and 6-H2), 0.88 and
0.82 (2 s, 9 H each, 2ϫ tert-butyl groups of the two TBDMS moie-
ties), 0.19, 0.14, 0.046 and 0.005 (4 s, 3 H each, 2ϫ methyl groups
of the two TBDMS moieties) ppm. 13C NMR (CDCl3, 75 MHz): δ
= 156.6, 137.3, 129.5, 129.2, 128.2, 126.6, 122.3 and 116.2 (aro-
matic C), 102.5 (Ph-CH), 100.2 (C-1), 81.7 (C-5), 76.0 (C-4), 69.1
(C-2 and C-3), 66.0 (C-6), 26.2 (CH3 of the tert-butyl groups), 18.4
and 18.2 (quaternary C of the tert-butyl groups), –3.0 and –3.7 (2ϫ
methyl groups of the TBDMS groups) ppm. ESI-MS (positive
ions): calculated for C31H48O6Si2: 572.884; found: 573.59
[M + H]+, 595.45 [M + Na]+, 611.40 (M + K]+.
Ac2O = acetic anhydride; AcOEt = ethyl acetate; tBuOOH = tert-
butyl hydroperoxide; DCA = dichloroacetic acid; DCM = dichloro-
methane; DCCI = N,N-dicyclohexylcarbodiimide; DIC = N,N-di-
isopropylcarbodiimide; DIEA = diisopropylethylamine; DMF =
N,N-dimethylformamide; DMT = 4,4Ј-dimethoxytriphenylmethyl;
Et3SiH = triethylsilane; Et3N = triethylamine; EtOH = ethanol;
Fmoc = fluorenylmethoxycarbonyl; HOBt = N-hydroxybenzotri-
azole. MSNT = 1-mesitylenesulfonyl-3-nitro-1,2,4-triazole; PTSA
= p-toluenesulfonic acid; TBDMS = tert-butyldimethylsilyl; TFA
= trifluoroacetic acid; THF = tetrahydrofuran.
Materials and Methods: TLC analyses were carried out on silica gel
plates from Merck (60, F254). Reaction products on TLC plates
were viewed under UV light and then by treatment with a 10%
Ce(SO4)2/H2SO4 aqueous solution. For column chromatography,
silica gel from Merck (Kieselgel 40, 0.063–0.200 mm) was used.
Tentagel® NH2 (0.29 mmolg–1 of primary amino groups) was pur-
chased from NovaBiochem. The solid support functionalizations
were carried out in a short glass column (5 cm length, 1 cm i.d.)
fitted with a sintered glass filter, a stopcock and a cap. The linear
oligosaccharide chains were assembled on an Expedite Perseptive
Biosystems DNA synthesizer, using phosphoramidite 9 as the
building block.
Synthesis of Derivative 7: Compound 6 (1.6 g, 2.8 mmol) was
treated with a TFA/DCM/H2O solution [1:10:0.5 (v/v/v), 3 mL].
The reaction was left at 0 °C and monitored by TLC using benzene/
AcOEt 98:2 (v/v) as the eluent system. After 4 h, the reaction mix-
ture was diluted with DCM and the organic phase was washed
twice with water, filtered and concentrated. The resulting crude
product was purified on a silica gel column, eluted with benzene
containing growing volumes of AcOEt (from 0 to 20%), which gave
1.1 g of pure 7 (82% yields).
Compound 7: White amorphous solid. Rf = 0.4 [benzene/AcOEt
95:5 (v/v)]. H NMR (CDCl3, 400 MHz): δ = 7.35–6.96 (complex
HPLC analyses and purifications were performed on a Beckman
System Gold instrument fitted with a UV detector module 166 and
a Shimadzu Chromatopac C-R6A integrator. By HPLC analysis
1
signals, 5 H, aromatic H), 5.08 (d, J = 6.0 Hz, 1 H, 1-H), 3.82
(m, 2 H, 6-H2), 3.71 (overlapped signals, 2 H, 2-H and 3-H), 3.57
(overlapped signals, 2 H, 4-H and 5-H), 2.49 (d, J = 4.0 Hz, 1 H,
4-OH), 1.93 (t, 1 H, 6-OH), 0.99 and 0.87 (2 s, 9 H each, 2ϫ tert-
butyl groups of the two TBDMS moieties), 0.20, 0.16, 0.13 and
0.010 (4 s, 3 H each, 2ϫ methyl groups of the two TBDMS moie-
ties) ppm. 13C NMR (CDCl3, 100 MHz): δ = 156.3, 129.5, 122.1,
115.6 (aromatic C), 99.1 (C-1), 77.0 (C-5), 75.9 (C-3), 74.2 (C-2),
70.9 (C-4), 62.7 (C-6), 26.1 (CH3 of the tert-butyl groups), 18.2 and
18.1 (quaternary C of the tert-butyl groups), –3.2, –3.3, –3.8 and
–3.9 (2 methyl groups of the TBDMS groups) ppm. ESI-MS (posi-
tive ions): calculated for C24H44O6Si2: 484.268; found: 507.39 [M
+ Na]+, 523.48 [M + K]+.
on
a
Nucleosil 100–5 C18 Supelco analytical column
(250ϫ4.6 mm, 5 µm), eluted with a linear gradient from 0 to 100%
of CH3CN in H2O over 30 min, flow: 0.8 mLmin–1, detection at λ
= 264 nm, all the synthesised compounds proved to be more than
98% pure. For the ESI MS analyses, a Waters Micromass ZQ in-
strument – fitted with an Electrospray source – was used in the
positive and/or negative mode. NMR spectra were recorded on
Bruker WM-400 or Varian Gemini 300 spectrometers. All the
chemical shifts are expressed in ppm with respect to the residual
solvent signal; J values are in Hz. The following abbreviations have
been used to explain the multiplicities: s = singlet, d = doublet, t
= triplet, q = quartet, m = multiplet, b = broad, dd = double
doublet. 31P NMR spectra have been registered using 85% H3PO4
as the external reference.
Synthesis of Derivative 8: Compound 7 (1.0 g, 2.06 mmol), dis-
solved in anhydrous pyridine (9 mL), was treated with DMTCl
(850 mg, 1.2 equiv.). The reaction, left whilst stirring overnight at
room temp., was then quenched by addition of CH3OH and con-
centrated under reduced pressure. The crude reaction product was
diluted with CHCl3 and the organic phase was washed twice with
water, filtered and concentrated. The crude product was then puri-
fied on a silica gel column eluting with cyclohexane, treated with a
few drops of pyridine, containing increasing amounts of AcOEt
(from 2 to 20%), to yield 1.5 g of pure target compound 8 (92%
yields).
UV measurements were carried out on a Jasco V-530 UV spectro-
photometer equipped with a Jasco ETC-505T temperature control-
ler unit. The temperature was kept constant with a thermoelec-
trically controlled cell holder (JASCO PTC-348).
Synthesis of Phenyl 4-O-[(2-Cyanoethoxy)(diisopropylamino)phos-
phanyl]-6-O-(4,4Ј-dimethoxytriphenylmethyl)-2,3-di-O-(tert-butyldi-
methylsilyl)-β-D-glucopyranoside (9)
Synthesis of Derivative 6: Phenyl 4,6-benzylidene-β--glucopyrano-
side (1.4 g, 4 mmol), prepared in almost quantitative yields as pre-
viously reported,[12] was repeatedly coevaporated with anhydrous
pyridine, dried under reduced pressure, dissolved in anhydrous
Compound 8: White amorphous solid. Rf = 0.5 [cyclohexane/AcOEt
9:1 (v/v)]. 1H NMR (C6D6, 400 MHz): δ = 7.77–6.76 (complex sig-
nals, 18 H, aromatic H), 5.02 (d, J = 7.5 Hz, 1 H, 1-H), 3.99 (dd,
Eur. J. Org. Chem. 2007, 3849–3858
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3855

D. Montesarchio et al.
FULL PAPER
J = 7.5 and 8.0 Hz, 1 H, 2-H), 3.76 (m, 1 H, 5-H), 3.66 (m, 2 H,
6-H2), 3.58 (overlapped signals, 2 H, 3-H and 4-H), 3.38 and 3.37
(2 s, 3 H each, 2ϫ OCH3 of DMT group), 1.16 and 1.15 (2 s, 9 H
each, 2ϫ tert-butyl groups of the two TBDMS moieties), 0.45, 0.39,
0.37 and 0.36 (4 s, 3 H each, 4ϫ methyl groups of the two TBDMS
moieties) ppm. 13C NMR (CDCl3, 100 MHz): δ = 158.3, 144.8,
135.9, 130.0, 129.3, 128.3, 128.1, 127.3, 126.7, 121.8, 116.1 and
113.0 (aromatic C), 99.4 (C-1), 86.2 (quaternary C of the DMT
group), 78.2 (C-5), 74.8 and 74.7 (C-2 and C-3), 72.1 (C-4), 64.1
(C-6), 55.1 (OCH3 of the DMT group), 26.2 (CH3 of the tert-butyl
group), 18.3 and 18.4 (quaternary C of the tert-butyl groups), –3.0
and –3.5 (CH3 groups directly linked to Si in the TBDMS moie-
ties) ppm. ESI-MS (positive ions): calculated for C45H62O8Si2:
786.398; found 809.56 [M + Na]+, 826.51 [M + K]+.
at r.t. and then with aq. ammonia for 1 h at 50 °C and finally
exhaustively washed with solvents and dried under reduced pres-
sure.
Synthesis of Functionalized Support 13: Phosphoramidite
9
(280 mg, 0.29 mmol), dissolved in a tetrazole solution (0.45 ,
300 µL), was added to support 11 (200 mg). The reaction was left
for 1 h at room temp. whilst stirring and, after repeated washings
with CH3CN, the resin was oxidized with an iodine solution in
THF/H2O/pyridine (0.1 , 3 treatments of 300 µL, 5 min each),
furnishing support 12. The efficiency of incorporation of the first
saccharide residue was spectrophotometrically evaluated by UV
measurements of the DMT cation at
λ = 498 nm (ε =
71700 cm–1 –1), released by acidic treatment (HClO4/EtOH 3:2,
v/v) of weighed aliquots of the dried resin. The obtained function-
alization was 0.16 mequiv.g–1 on average, corresponding to 65%
incorporation of derivative 9. Subsequently, the support was
capped by treatment with Ac2O in pyridine (1:1, v/v, 400 µL total
volume, 1 h). Next, support 12 was treated with Et3N/pyridine (1:1,
v/v, 400 µL total volume, 3 treatments, 6 h each) at room temp. in
order to remove the 2-cyanoethyl group, to afford the desired 13.
Spectrophotometric tests were then carried out to check that no
losses in functionalization had occurred due to this treatment. Af-
ter repeated washings, the resin was dried under reduced pressure,
suspended in CDCl3 and analysed by 31P NMR spectroscopy,
which confirmed the complete conversion of the phosphotriester
into phosphodiester function.
Synthesis of Derivative 9: DIEA (940 µL, 5.34 mmol) and 2-cyano-
ethyl N,N-diisopropyl-chlorophosphoramidite (640 µL, 2.94 mmol,
3.2 equiv.) were sequentially added to compound 8 (700 mg,
0.89 mmol), dissolved in anhydrous DCM (6 mL). The reaction,
monitored by TLC using cyclohexane/AcOEt 9:1 (v/v) as the eluent
system, was left 1 h at room temperature. Then the mixture was
diluted with AcOEt and the organic phase was washed with a satu-
rated NaCl aq. solution, dried with anhydrous Na2SO4, filtered and
concentrated under reduced pressure. The crude product was then
purified on a silica gel column, eluted with cyclohexane (with a few
drops of triethylamine) containing increasing volumes of AcOEt
(from 3 to 10%), which provided 780 mg (0.79 mmol) of pure 9
(90% yields).
Solid Support 12: 31P NMR (CDCl3, 161.98 MHz): δ = –3.98 ppm.
Solid Support 13: 31P NMR (CDCl3, 161.98 MHz): δ = –2.52 ppm.
Compound 9 (as a diastereomeric mixture): White amorphous solid.
Rf = 0.4 [cyclohexane/AcOEt 9:1 (v/v)]. 1H NMR (CDCl3,
400 MHz): δ = 7.46–6.74 (complex signals, 36 H, aromatic H), 5.31
and 5.30 (2 d, J = 7.0 and 7.0 Hz, 2 H, 2ϫ 1-H), 4.02 (m, 2 H, 2ϫ
2-H), 3.90 (m, 4 H, 2ϫ 4-H and 2ϫ 5-H), 3.88 (m, 2 H, 2ϫ 3-H),
3.77 (s, 12 H, 2ϫ OCH3 of the DMT group), 3.84–3.67 (m, 4 H,
OCH2CH2CN), 3.60–3.50 [m, 4 H, CH(CH3)2], 3.35–3.23 (m, 4 H,
6-H2), 2.57–2.41 (m, 4 H, OCH2CH2CN), 1.19–1.12 [overlapped
signals, 24 H, CH(CH3)2], 0.88 and 0.80 (2 s, 18 H each, 2ϫ tert-
butyl groups of the two TBDMS moieties), 0.17, 0.15, 0.12, 0.09,
0.08, 0.07, –0.05 and –0.06 (4 s, 24 H, 4ϫ methyl groups of the two
TBDMS moieties) ppm. 13C NMR (CDCl3, 100 MHz): δ = 158.3,
157.1, 145.0, 136.2, 136.1, 123.0, 129.2, 128.2, 128.1, 127.6, 126.8,
126.5, 121.7, 121.6, 116.1 and 112.9 (aromatic C), 117.5 and 117.4
(CN), 100.1 (C-1), 86.0 (quaternary C of the DMT group), 79.2
and 79.0 (C-4), 77.5 and 77.4 (C-5), 75.6, 75.4, 70.7 and 70.6 (C-2
and C-3), 64.8 (C-6), 58.6, 58.4, 58.1 and 58.0 (OCH2CH2CN),
55.0 (OCH3 of the DMT group), 43.15, 43.07, 43.06 and 42.95
[CH(CH3)2]; 25.9, 25.8, 25.7 and 24.4 [CH3 of the tert-butyl group
and CH(CH3)2], 20.1, 20.0, 18.8 and 18.7 (quaternary C of the tert-
butyl group and OCH2CH2CN), –4.1, –4.2, –4.4, –4.6 and –4.9
(CH3 groups of the TBDMS moieties) ppm. 31P NMR (CDCl3,
161.98 MHz): δ = 152.4 and 150.6 ppm. ESI-MS (positive ions):
calculated for C54H79N2O9PSi2: 986.506; found: 987.66 [M + H]+,
1009.79 [M + Na]+, 1025.70 [M + K]+.
Synthesis of Linear Precursor 15: Solid support 13 (200 mg,
0.02 mequiv.), swollen in anhydrous DCM, was treated with DCA
in CH2Cl2 (2%, 5 treatments of 300 µL each, 3 min). Subsequently,
the resin was exhaustively washed with pyridine, DCM and anhy-
drous CH3CN and treated with phosphoramidite
9 (60 mg,
3 equiv.), dissolved in a tetrazole solution in anhydrous CH3CN
(0.45 , 500 µL). The support, left for 1.5 h at room temp., was
then washed with CH3CN, and then treated with an I2 solution in
pyridine/H2O/THF (0.1 , 3 treatments of 300 µL, 5 min each).
The support, repeatedly washed with CH3CN, was dried under re-
duced pressure. The efficiency of incorporation of the second glyco-
sidic residue in the solid phase, monitored by UV measurements at
λ = 498 nm (ε = 71700 cm–1 –1) of the DMT cation released by
acidic treatments [HClO4/EtOH 6:4 (v/v)] on weighed amounts of
the resin, was 0.10 mequiv.g–1, corresponding to almost quantita-
tive incorporation of derivative 9. After a capping treatment with
Ac2O in pyridine (1:1, v/v, 1 h at room temp.), the resulting support
14 was detritylated by the procedure reported above for 13 and
then exhaustively washed with DCM and CH3CN, giving support
15.
Synthesis of Support 17: Three treatments with MSNT (150 mg
each, 0.51 mmol, 25equiv., 3ϫ 6 h) dissolved in pyridine (1.5 mL)
were performed at room temp. on support 15 (100 mg), previously
swollen in anhydrous pyridine, whilst stirring, to afford support 16.
After repeated washings with pyridine and CH3CN, the resin was
then treated with Et3N/Py (1:1, v/v) for 3 h whilst stirring, yielding
desired support 17. After repeated washings with solvents the resin
was dried, suspended in CDCl3 and analysed by 31P NMR spec-
troscopy, which confirmed that the reaction had gone to comple-
tion.
Solid Support 16: 31P NMR (CDCl3, 161.98 MHz): δ = 0.61, –0.21,
–4.17 and –4.32 ppm.
Solid Support 17: 31P NMR (CDCl3, 161.98 MHz): δ = 2.08, 0.61,
–3.52 and –4.36 ppm.
Synthesis of Solid Support 18. Functionalization of the Solid Support
with Phosphoramidite Derivative 9
Synthesis of Support 11: TentaGel-NH2 resin (500 mg, 0.14 mmol),
swollen in anhydrous DCM, was treated with 2-(3-chloro-4-hy-
droxyphenyl)acetic acid (10, 270 mg, 1.45 mmol), DIC (220 µL,
1.45 mmol), DIEA (250 µL, 1.45 mmol) and HOBt (200 mg,
1.45 mmol) in anhydrous pyridine (4 mL). The reaction was left
overnight at room temp. whilst stirring. The obtained support was
then repeatedly washed with pyridine, DCM, MeOH and CH3CN,
treated with Ac2O/pyridine 1:1 (v/v), 400 µL total volume, for 1 h
3856
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3849–3858

Novel Immobilized Oligosaccharides
FULL PAPER
Trans. 2 1998, 2639–2646. For a recent example of highly elab-
orated cyclodextrins, useful for specific oligosaccharide–pro-
tein recognition, see for example: g) N. Smiljanic, V. Moreau,
D. Yockot, J. M. Benito, J. M. Garcia Fernández, F. Djedaïni-
Pilard, Angew. Chem. Int. Ed. Angew. Chem. Int. Ed. Engl.
2006, 45, 5465–5468.
a) C. Baudin, C. Pean, B. Pellizzari, A. Gadelle, F. Fauvelle, J.-
C. Debouzy, J.-P. Dalbiez, B. Perly, J. Inclusion Phenom.
Macrocyclic Chem. 2000, 38, 287–296; b) C. Baudin, C. Pean,
B. Perly, P. Gosselin, Int. J. Environ. Anal. Chem. 2000, 77, 233–
242; c) B. Perly, C. Baudin, P. Gosselin, Water purification using
cyclodextrins, Int. Patent WO 9818722, 1998, p. 21.
a) P. R. Ashton, G. Gattuso, R. Koniger, J. F. Stoddart, D. J.
Williams, J. Org. Chem. 1996, 61, 9553–9555; b) H. Yamamura,
H. Masuda, Y. Kawase, M. Kawai, Y. Butsugan, H. Einaga,
Chem. Commun. 1996, 1069–1070; c) H. Yamamura, H. Na-
gaoka, M. Kawai, Y. Butsugan, K. Fujita, Tetrahedron Lett.
1995, 36, 1093–1094.
Synthesis of Support 18: Resin 17, bearing the protected cyclic di-
mer, was then subjected to several tests to optimize the full removal
of the TBDMS groups in the solid phase, varying the solvent, tem-
perature, excess and nature of the desilylating agent. In order to
test the efficiency of the adopted desilylating systems, after each
test, samples (30 mg) of the resins were treated with 1,1,3,3-tet-
ramethylguanidinium 2-nitrobenzaldoximate solution (0.2 ,
500 µL) in H2O/dioxane (1:1, v/v) for 12 h at room temp., and the
obtained eluates were then analysed by 1H NMR and ESI-MS
data. Full TBDMS removal was achieved by suspending the resin
in THF (100 µL) and Et3N·3HF (300 µL, 1.8 mmol, 1200 equiv.)
and leaving it for 12 h at 50 °C. Under these conditions, the cyclic
dimer 1[12] was exclusively detached from the solid support by
benzaldoximate treatment and was found to be identical to an au-
thentic, independently synthesized sample.
[3]
[4]
Synthesis of TBDMS-Protected Cyclic Dimer 19: A sample (30 mg)
of functionalized resin 17 was treated with a solution of 1,1,3,3-
tetramethylguanidinium 2-nitrobenzaldoximate (0.2 , 500 µL) in
H2O/dioxane (1:1, v/v) for 12 h at room temp. The detached mate-
rial, after gel filtration chromatography on a Sephadex G25 column
eluted with H2O/EtOH 1:1 (v/v), was then analysed by HPLC on
an analytical RP18 column (Nucleosil 100–5 C18 Supelco,
4.6ϫ250 mm, 5 µm). By using a gradient from 30% to 100% of
CH3CN in TEAB (0.1 , pH 7.0) over 15 min, flow: 0.8 mLmin–1,
detection at λ = 265 nm, a unique main peak was observed in the
mixture, accounting for more than 90% of the total integrated area,
with retention time 17.46 min. This, collected, gave 2 mg (40%
overall yields calculated from support 13) of pure target compound
19.
[5]
[6]
[7]
[8]
a) F. Fauvelle, A. Gadelle, J. C. Debouzy, C. Baudin, B. Perly,
Supramol. Chem. 2000, 11, 233–237.
L. Rambaud, J.-P. Dalbiez, B. Amekraz, C. Moulin, B. Perly,
C. Baudin, Eur. J. Org. Chem. 2006, 1245–1250.
a) A. R. Hedges, Chem. Rev. 1998, 98, 2035–2044; b) B. Perly,
S. Moutard, F. Djedaïni-Pilard, PharmaChem. 2005, 4, 4–9.
a) R. Auzély-Velty, B. Perly, O. Taché, Th. Zemb, P. Jéhan, P.
Guenot, J. P. Dalbiez, F. Djedaıni-Pilard, Carbohydr. Res. 1999,
318, 82; b) R. Auzély-Velty, F. Djedaïni-Pilard, S. Désert, B.
Perly, Th. Zemb, Langmuir 2000, 16, 3727–3734.
a) N. Badi, N. Jarroux, P. Guégan, Tetrahedron Lett. 2006, 47,
8925–8927; b) N. Madhavan, E. C. Robert, M. S. Gin, Angew.
Chem. Int. Ed. 2005, 44, 7584 –7587; c) A. Mazzaglia, R.
Donohue, B. J. Ravoo, R. Darcy, Eur. J. Org. Chem. 2001,
1715–1721; d) M. Wazynska, A. Temeriusz, K. Chmurski, R.
Bilewicz, J. Jurczak, Tetrahedron Lett. 2000, 41, 9119–9123; e)
H. Parrot-Lopez, C.-C. Ling, P. Zhang, A. Baszkin, G. Al-
brecht, C. De Rango, A. W. Coleman, J. Am. Chem. Soc. 1992,
114, 5479–5480.
a) T. Velasco-Torrijos, P. V. Murphy, Tetrahedron: Asymmetry
2005, 16, 261–272; b) B. A. Mayes, R. J. E. Stetz, C. W. G. An-
sell, G. W. J. Fleet, Tetrahedron Lett. 2004, 45, 153–156; c)
D. Q. Yuan, T. Tahara, W. H. Chen, Y. Okabe, C. Yang, Y.
Yagi, Y. Nogami, M. Fukudome, K. Fujita, J. Org. Chem.
2003, 68, 9456–9466; d) T. Kida, A. Kikuzawa, Y. Nakatsuji,
M. Akashi, Chem. Commun. 2003, 3020–3021.
a) K. D. Bodine, D. Y. Gin, M. S. Gin, Org. Lett. 2005, 7,
4479–4482; b) B. Hoffmann, B. Bernet, A. Vasella, Helv. Chim.
Acta 2002, 85, 265–287; c) L. Fan, O. Hindsgaul, Org. Lett.
2002, 4, 4503–4506.
[9]
Compound 19, Triethylammonium Salt: 1H NMR (CD3OD,
400 MHz): δ = 7.29–6.97 (complex signals, 5 H, aromatic H), 5.17
(d, J = 8.0 Hz, 1 H, 1-H), 4.17–3.98 (overlapped signals, 3 H, 4-H
and 6-H2), 3.85 (t, J = 8.0 and 8.0 Hz, 1 H, 3-H), 3.75 (m, 1 H, 5-
H), 3.69 (t, J = 8.0 and 8.0 Hz, 1 H, 2-H), 3.21 [q, 6 H, (CH3CH2)3-
NH]+, 1.32 [t, 9 H, (CH3CH2)3NH]+, 0.97 and 0.87 (2 s, 9 H each,
tert-butyl groups of the TBDMS moieties), 0.23, 0.19, 0.17 and 0.15
(4 s, 3 H each, methyl groups of the TBDMS moieties) ppm. 31P
NMR (CD3OD, 161.98 MHz): δ = 2.66 ppm. UV (CH3OH): λmax
= 267 nm. ESI-MS (negative ions): calculated for C48H86O16Si4P2:
1093.480; found 545.42 (M – 2H]2–; 1092.84 [M – H]–.
[10]
[11]
Acknowledgments
[12]
[13]
G. Di Fabio, A. Randazzo, J. D’Onofrio, C. Ausin, A. Grandas,
E. Pedroso, L. De Napoli, D. Montesarchio, J. Org. Chem.
2006, 9, 3395–3408.
We thank the Ministero dell’Università e della Ricerca (MIUR,
PRIN) and the Regione Campania (LR n.5) for grants in support
of this investigation and the Centro di Metodologie Chimico-
Fisiche (CIMCF), Università di Napoli “Federico II”, for the MS
and NMR facilities.
a) L. Moggio, L. De Napoli, B. Di Blasio, G. Di Fabio, J.
D’Onofrio, D. Montesarchio, A. Messere, Org. Lett. 2006, 8,
2015–2018; b) E. M. Alazzouzi, N. Escaja, A. Grandas, E. Ped-
roso, Angew. Chem. Int. Ed. Engl. 1997, 36, 1506–1508.
a) G. Sicoli, Z. Jiang, L. Jicsinsky, V. Shurig, Angew. Chem.
Int. Ed. 2005, 44, 4092–4095; b) T. Sukegawa, T. Furuike, K.
Niikura, A. Yamagishi, K. Monde, S.-I. Nishimura, Chem.
Commun. 2002, 430–431; c) S. H. Chiu, D. C. Myles, J. Org.
Chem. 1999, 64, 331–333; d) S. Hannessian, A. Benalil, C.
Laferrière, J. Org. Chem. 1995, 60, 4786–4797; e) P. Zhang,
C. C. Ling, A. W. Coleman, H. Parrot-Lopez, H. Galons, Tet-
rahedron Lett. 1991, 32, 2769–2770; f) P. Fügedi, Carbohydr.
Res. 1989, 192, 366–369.
In our previous paper, we first coupled the bifunctional linker
2-(3-chloro-4-hydroxyphenyl)acetic acid (10) with the 4-phos-
phoramidite sugar building block in solution, and then treated
the resulting adduct with the TentaGel-NH2 solid support.
Here this procedure was simplified by directly attaching bifunc-
tional linker 10 to the resin by classical peptide coupling proto-
[14]
[1] J. Szejtli, Chem. Rev. 1998, 98, 1743–1753 and references cited
therein.
[2] a) G. Gattuso, S. A. Nepogodiev, J. F. Stoddart, Chem. Rev.
1998, 98, 1919–1958; b) C. J. Easton, S. F. Lincoln, Modified
Cyclodextrins: Scaffolds and Templates for Supramolecular
Chemistry, Imperial College Press, London, 1999. There is an
impressive number of papers on modified cyclodextrins recog-
nizing neutral or ionic guests; among others, see for example:
c) R. Heck, F. Dumarcay, A. Marsura, Chem. Eur. J. 2002,
8, 2438–2445; d) B. L. May, P. Clements, J. Tsanaktsidis, C. J.
Easton, S. F. Lincoln, J. Chem. Soc. Perkin Trans. 1 2000, 463–
469; e) F. J. Huuskonen, J. E. H. Buston, N. D. Scotchmer,
H. L. Anderson, New J. Chem. 1999, 23, 1245–1252; f) J. Lin,
C. Creminon, B. Perly, F. Djedaïni-Pilard, J. Chem. Soc. Perkin
[15]
Eur. J. Org. Chem. 2007, 3849–3858
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3857

D. Montesarchio et al.
FULL PAPER
cols, as also reported in ref.[13a], and by then treating phenolic
resin 11 with sugar building block 9. On comparing the two
routes for the synthesis of cyclic dimer 1, no significant differ-
ences were found either in yields or in purity of the released
material.
[18] a) R. K. Dey, U. Jha, A. C. Singh, S. Samal, A. R. Ray, Analyt-
ical Sci. 2006, 22, 1105–1110; b) T.-Y. Guo, P. Liu, J.-W. Zhu,
M.-D. Song, B.-H. Zhang, Biomacromolecules 2006, 7, 1196–
1202; c) P. Piatek, D. T. Gryko, A. Szumna, J. Jurczak, Tetrahe-
dron 2004, 60, 5769–5776.
[19] A. Narumi, H. Kaga, Y. Miura, I. Otsuka, T. Satoh, N.
Kaneko, T. Kakuchi, Biomacromolecules 2006, 7, 1496–1501.
Received: March 6, 2007
[16] F. Eckstein, Oligonucleotides and Analogues: A Practical Ap-
proach, IRL Press, Oxford, UK, 1991.
[17] I. H. Lee, Y.-C. Kuan, J.-M. Chern, J. Hazard. Mater. 2006,
B138, 549–559.
Published Online: June 19, 2007
3858
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3849–3858