Water soluble heptakis(6-deoxy-6-thio)cyclomaltoheptaose capped gold nanoparticles via metal vapour synthesis: NMR structural characterization and complexation properties

Article abstract of DOI:10.1016/j.carres.2011.02.001

The complexation of heptakis(6-deoxy-6-thio)cyclomaltoheptaose to gold nanoparticles prepared by using the Metal Vapour Synthesis (MVS) led to water soluble gold nanoaggregates, thermally stable at 25 °C. The role of gold concentration in the MVS-derived starting solution as well as of the cyclodextrin to gold molar ratio on the size of cyclodextrin-capped gold nanoparticles were investigated. The ability of cyclodextrin bonded to gold nanoparticles to include deoxycytidine was also probed in comparison with that of 1-thio-β-d-glucose sodium salt.

Full text of DOI:10.1016/j.carres.2011.02.001

Carbohydrate Research 346 (2011) 753–758
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Water soluble heptakis(6-deoxy-6-thio)cyclomaltoheptaose capped gold
nanoparticles via metal vapour synthesis: NMR structural characterization
and complexation properties
a,
⇑
a
a
a
a
Gloria Uccello-Barretta , Claudio Evangelisti , Federica Balzano , Letizia Vanni , Federica Aiello ,
Laszlo Jicsinszky
a
b
Dipartimento di Chimica e Chimica Industriale, Università degli Studi di Pisa, via Risorgimento 35, I-56126 Pisa, Italy
CYCLOLAB Ltd, Illatos út 7, 1097 Budapest, Hungary
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The complexation of heptakis(6-deoxy-6-thio)cyclomaltoheptaose to gold nanoparticles prepared by
using the Metal Vapour Synthesis (MVS) led to water soluble gold nanoaggregates, thermally stable at
Received 24 November 2010
Received in revised form 27 January 2011
Accepted 1 February 2011
25 °C. The role of gold concentration in the MVS-derived starting solution as well as of the cyclodextrin
to gold molar ratio on the size of cyclodextrin-capped gold nanoparticles were investigated. The ability of
cyclodextrin bonded to gold nanoparticles to include deoxycytidine was also probed in comparison with
Available online 28 February 2011
that of 1-thio-b-D-glucose sodium salt.
Keywords:
Ó 2011 Elsevier Ltd. All rights reserved.
NMR spectroscopy
Gold
Nanoparticles
Cyclodextrins
Inclusion compounds
1
. Introduction
NH2
SH
6,6'
4
g
f
N
O
The unique properties of metal nanoparticles (M-NPs) account
for their widespread applications in material science, catalysis,
2
5
1
HO
N
O
HO
1
–5
3
OH
and biomedicine. Their biotechnological applications as artificial
receptors, drug delivery systems, or biosensors mainly rely on the
fine tuning of their sizes. Thus several interdisciplinary investiga-
tions have been addressed to the development of efficient and
reproducible methods for controlling nanoparticles growth in a
hydrophilic environment.
Among M-NPs, gold nanoparticles (Au-NPs) are endowed with
considerable potential to be a biomedicine by virtue of their im-
proved stability and their optical and electronic properties.
O
O
a
7
HO
β-CDSH
DC
OH
4
6,6'
O
HO
HO
2
S
5
Na
1
3
OH
6
–10
β-GluSNa
The supramolecular aggregation of gold nanoparticles to host
compounds such as the cyclodextrins gives the opportunity to con-
jugate their biosensing properties to drug transport and controlled
release for developing new strategies of specific drug targeting. In
view of such a kind of attractive application, the affinity of sulfur
to gold has been already exploited by using chemical reduction
Chart 1.
Among the various preparative routes Metal Vapour Synthesis
MVS) provides a valuable way to obtain metal nanoparticles of tai-
(
1
5,16
1
1–14
lored size.
The co-condensation of metal vapours with vapours
methods
in order to obtain heptakis(6-deoxy-6-thio)cyclom-
of weakly stabilizing organic solvents on the cold walls (ꢀ196 °C)
of a reactor affords solvent-stabilized metal nanoparticles (Sol-
vated Metal Atoms, SMAs), which are soluble in excess of organic
solvent. Metal nanoparticles obtained this way are very small
altoheptaose (b-CDSH) (Chart 1) capped gold nanoparticles, which
are water soluble.
(
diameter <5 nm) and their sizes can be strictly controlled by
⇑
Corresponding author. Fax: +39 050 2219260.
E-mail address: gub@dcci.unipi.it (G. Uccello-Barretta).
means of selected experimental factors (i.e., kind of solvent, metal
0
008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.02.001

7
54
G. Uccello-Barretta et al. / Carbohydrate Research 346 (2011) 753–758
concentration or keeping temperature) which affect the metal
kT
D ¼ 6
pgRH
ð1Þ
1
7–19
clustering processes.
Remarkable potentialities of MVS in
exerting a fine tuning of nanoparticles sizes prompted us to exploit
MVS technique in the preparation of water soluble Au-NPs stabi-
lized by heptakis(6-deoxy-6-thio)cyclomaltoheptaose ((Au) /b-
n
CDSH). The NMR DOSY (Diffusion-Ordered SpectroscopY) technique
was employed as an effective, reliable and fast way to analyse the
role of the total concentration of gold in SMA solutions as well as of
the Au/b-CDSH molar ratio on nanoparticles sizes. The nature of
where k is the Boltzmann constant, T the absolute temperature and
is the solution viscosity.
g
A rough correlation of the diffusion coefficients to the molecular
weights allowed us to establish that about a two to three-fold in-
crease of the molecular weight occurred. Both the loss of proton
isochrony and the increase of the molecular sizes pointed out the
presence of oligomeric aggregates of the cyclodextrin due to the
formation of disulfide bridges between primary groups of the
cyclodextrin. Any further oligomerization was not detected at
longer times.
The precipitates, which were formed as a consequence of the
addition of the cyclodextrin to Au/acetone SMAs, were re-dissolved
2
in D O and the corresponding limpid dark-purple solutions were
stable at room temperature and could be analysed directly by
NMR without any kind of preliminary manipulation. The signifi-
cant changes detected in the spectral parameters of the cyclodex-
trin in (Au)
us to investigate the nature of aggregation processes involved in
the formation of D O soluble gold nanoparticles. The supramolec-
supramolecular aggregation processes and the ability of (Au)
CDSH nanoaggregates to act as hosts for deoxycytidine (DC)
Chart 1) were also investigated by NMR in order to obtain some
n
/b-
(
preliminary information on Au-NPs potentiality in the field of drug
transport and release.
To mimic the effect of individual units of the cyclodextrin and
thus assess the role played by the structural pre-organization of
the cyclodextrin into supramolecular aggregation processes, we se-
lected 1-thio-b-
D-glucose sodium salt (b-GluSNa) (Chart 1) as a
structural analogue of glucopyranose units of the host.
n
/b-CDSH NPs relative to the pure cyclodextrin allowed
2
2
. Results and discussion
2
ular assembly showed a marked dependence on the gold to cyclo-
dextrin molar ratio and in some cases on the gold concentration in
.1. Preparation of Au-NPs
the starting Au/acetone SMA.
In agreement with the previously reported procedure,17 gold
The 1H NMR spectrum of the diluted preparations which con-
vapours (atoms) were co-condensed with vapours of acetone at
the temperature of the liquid nitrogen. By further melting of
the frozen gold/acetone matrix, a deep purple solution of sol-
vated gold atoms was obtained that is stable at low temperature
tained gold 0.4 mg/mL in the presence of a large excess of cyclo-
dextrin (molar ratio Au/b-CDSH = 1:5) resembled the spectrum of
monomer cyclodextrin. A nearly complete isochrony of corre-
sponding protons of the glucopyranose units was observed, but
remarkable shifts of the signals (Table 1) were detected to be
attributed to metal complexation. The major effects were detected
for the higher frequency shifted methylene proton (H-6) belonging
to the primary site and the proton H-4 which is located on the
external surface of the cyclodextrin and is adjacent to the primary
site.
(
ꢀ20 °C). The addition of b-CDSH to the Au/acetone solution gave
a dark-purple powder of (Au) /b-CDSH which was insoluble in
n
acetone and could be easily re-dissolved in water. Au/acetone
SMAs with different Au concentration (0.4 mg Au/mL (LC) and
.7 mg Au/mL (HC)) were prepared and these systems were used
as starting materials for the preparation of b-CDSH-capped Au
nanoparticles (Chart 2).
3
The complexation of the metal to the thiol moieties of the cyclo-
dextrin fully accounts for the complexation induced shifts of the
cyclodextrin resonances.
n
In all cases, the aggregates of (Au) /b-CDSH were easily sepa-
rated from the eventual excess of b-CDSH by precipitation from
the reaction mixture. Gold nanoparticles stabilized by b-GluSNa
were prepared by a similar procedure (see Section 3).
ꢀ10
2 ꢀ1
The diffusion coefficient of 1.7 ꢁ 10
m s was measured,
from which nanoparticles diameters about of 2.9 nm were calcu-
lated. On assuming a single shell of cyclodextrin units around
Au-NPs, the average diameter of 1.3 nm was then estimated for
gold core nanoaggregates. On the basis of the previously reported
2
.2. Characterization of (Au)
n
/b-CDSH NPs
O) spectrum of b-CDSH was in
1
The H NMR (600 MHz, 25 °C, D
2
20
theoretical study, it was possible to estimate that in our system
agreement with the presence of a symmetric structure where all
primary sites were derivatized. During the time the spectrum of
the cyclodextrin underwent relevant changes: each kind of gluco-
pyranose proton produced a wide distribution of resonances which
reflected the loss of symmetry. A significant decrease of the diffu-
sion coefficient (D) was also detected in the DOSY maps: D value
about 80 atoms of gold were present in each nanoparticle and
about the 75% of them were on the surface of the nanoparticle.
Interestingly, the behaviour of the residual acetone in the nano-
particles solution gave a further confirmation of the gold to cyclo-
dextrin supramolecular assembly. As a matter of fact the residual
ꢀ10
ꢀ10
2 ꢀ1
acetone showed a diffusion coefficient of 7 ꢁ 10
m s which
ꢀ
ꢀ1
10
2 ꢀ1
lowered from 2.4 ꢁ 10
m s in the freshly prepared sample
2 ꢀ1
was remarkably lower than the value (15 ꢁ 10
m s ) mea-
ꢀ
10
2
to 1.7–1.8 ꢁ 10
longer times, which was in agreement with the increase of the
hydrodynamic radius (R ) during the time. Eq. 1 gives the depen-
m s in the samples which were analysed at
sured for the pure acetone in water. In the ROESY map, ROE effects
were detected between acetone and the protons H-3–H-5 of the
cyclodextrin which were located inside the cyclodextrin cavity.
Reasonably the complexation of gold to the cyclodextrin drives
the residual acetone inside the cyclodextrin cavity where a
remarkable slowing down of its translational diffusion is produced.
H
dence of the diffusion coefficient on the hydrodynamic radius
based on the Stokes–Einstein relationship, which strictly holds in
the spherical approximation.
O
O
1
) -196°C
) -20°C
β-CDSH, H2O
25°C
Au/β-CDSH
Au(v)
+
Au
(V)
T
2
Soluble in water
Stable at low
Insoluble in Acetone
temperature (-20°C)
Chart 2.

G. Uccello-Barretta et al. / Carbohydrate Research 346 (2011) 753–758
755
Table 1
1
H NMR chemical shifts (d, ppm, 600 MHz, D
CDNPs ꢀ dCD, ppm) of pure b-CDSH (dCD) and of b-CDSH in (Au)
2
O, 25 °C) and complexation shifts
/b-CDSH NPs (dCDNPs
obtained from 0.4 mg/mL of Au/acetone SMA (molar ratio Au/b-CDSH = 1:5)
Dd
)
(d
n
Proton
d
CD
d
CDNPs
D
d
1
2
3
4
5
6
5.03
3.52
3.83
3.62
3.83
2.94
2.83
5.03
3.50
3.80
3.68
3.78
2.86
2.81
0.00
ꢀ0.02
ꢀ0.03
0.06
ꢀ0.05
ꢀ0.08
ꢀ0.02
0
6
It is noteworthy that the NMR spectrum of the solution did not
change with time, thus oxidation processes were not observed as
in the case of pure cyclodextrin.
When the assembly of the gold nanoparticles and the cyclodex-
trin occurred in the presence of high Au/b-CDSH molar ratios
Figure 1. 1D ROESY (600 MHz, 25 °C, D
perturbation of H-1 nuclei of (a) b-CDSH and (b) (Au)
acetone SMA, molar ratio Au/b-CDSH = 1:5)
2
O, mix = 0.3 s) spectra obtained by selective
n
/b-CDSH NPs (3.7 mg/mL Au/
(
1:0.3), a remarkably different behaviour was found: less soluble
0
indicated nearly equal H-1–H-2 and H-1–H-4 distances. In the
nanoparticles were formed, which probably quickened the precip-
itation at very early stages of the metal nucleation. Accordingly,
very small gold nanoparticles of 2.4 nm were formed and the
NMR spectrum was very similar to that of the oxidized cyclodex-
trin, with the loss of symmetry. Thus, it would be concluded that
excess of gold atoms favours the oxidation of cyclodextrin.
The behaviour observed at high Au/b-CDSH molar ratios was
reproduced till the one to one molar ratio, that is, complexed oxi-
dized cyclodextrin with formation of small nanoparticles. In all
cases, very small nanoparticles sizes were obtained.
n
solution containing (Au) /b-CDSH NPs (3.7 mg/mL Au/acetone
SMA, molar ratio Au/b-CDSH = 1:5), the ROE H-1–H-2 was higher
than the ROE H-1–H-4 was (Fig. 1b), in agreement with the fact
that the aggregation of the cyclodextrin to the gold nanoparticles
caused a significant rotation of the glucopyranose rings about the
glycosidic linkages, which turned the proton H-1 of one unit away
1
H
0
0
from the proton H-4 of the adjacent one.
2
.3. NMR investigation of the complexation phenomena
between deoxycytidine and (Au) /b-CDSH NPs
In order to study the effect of the gold concentration on the
nanoparticles sizes distribution, we analysed the NMR spectra of
the solutions obtained by adding different amounts of the cyclo-
dextrin to a 3.7 mg/mL Au/acetone solution, that is, 10-fold more
concentrated with respect to the cases just discussed. Four differ-
ent molar ratios Au/b-CDSH were probed (1:5, 1:1, 1:0.3, 1:0.2)
spanning from a molar excess of the cyclodextrin to a molar excess
of gold atoms. The gold concentration did not affect the kind of
cyclodextrin-to-gold assembly, which only depended on the gold
to cyclodextrin molar ratio. As a matter of fact complexed mono-
mer cyclodextrin was detected at high cyclodextrin/gold molar ra-
tios, whereas oxidized cyclodextrin was present in the
nanoparticles formed at high gold/cyclodextrin molar ratios. In
the last case smaller nanoparticles of about 2.5 nm (Table 2) were
found, irrespective of the total metal concentration. Only at high
cyclodextrin/gold molar ratio, where the cyclodextrin was mono-
mer, the nanoparticles sizes were sensitive to the concentration
with an increase at higher gold concentrations and a wider sizes
distribution, as revealed by the analysis of the diffusion coefficients
n
Nucleoside analogues are potential anticancer and/or antiviral
agents, the solubility and transport of which could be improved
by inclusion inside the cyclodextrin cavity. In view of such kind
n
of applications we compared the ability of b-CDSH and (Au) /b-
CDSH nanoparticles to include the deoxycytidine (DC). Complexa-
tion phenomena were detected by measuring the diffusion coeffi-
cients and the selective relaxation rates (R) of DC protons in the
presence and in the absence of the cyclodextrin or cyclodextrin-
capped gold nanoparticles.
The diffusion coefficients are very sensitive to complexation
phenomena which slow down the translational diffusion of com-
plexed species and, hence, bring about decreases of the diffusion
parameters, which reflect the increase of the apparent sizes of
complexed species with respect to uncomplexed ones (see Eq. 1).
The selective relaxation rates (R) of protons of the ligand are more
ns
sensitive indicators of the binding than nonselective rates (R ) are.
In fact, methods based on the determination of the selective relax-
ation rates take advantage of the favourable dependence of R on
the reorientational correlation time (s ) in the region of the slow
c
molecular motions, in which the small molecule is forced by the
(Table 2).
It is noteworthy that the pattern of intra-ring ROEs H-1–H-2
0
versus inter-ring ROEs H-1–H-4 (the apex indicates the glucopyr-
anose rings which is adjacent to the observed ones) was remark-
ably different in the solution containing the nanoparticles in
comparison with the pure cyclodextrin. As shown in the Figure 1a,
comparable effects were detected in the pure cyclodextrin, which
interaction with the cyclodextrin. In the fast-motion region
2
2
c
(
x
s
ꢂ 1;
x = Larmor frequency), both the selective and nonse-
lective relaxation rates increase progressively with increasing
2
1
s
c
.
When the molecular motion of the ligand is slowed down to
x
2
2
c
the
s
ꢃ 1 region as the consequence of the interaction with
ns
Table 2
the host, R shows a sharp increase, whereas R reaches a maxi-
mum for
In the presence of well-resolved proton resonances, the selective
relaxation rates can be easily determined by selective excitation
of only one spin, leaving unperturbed the other ones, and by fol-
lowing the magnetization recovery in the time.
Diffusion coefficients (D, ꢁ10 m2
1
0
s
ꢀ1
, D
O, 600 MHz, 25 °C) of b-CDSH in (Au)
/b-
2
n
2
2
c
2
2
c
x
s
ffi 1 and then decreases with further
x
s
increases.
CDSH NPs obtained from the LC or HC Au/acetone SMA solutions at different Au/b-
CDSH molar ratio and calculated hydrodynamic diameters (d
H
, nm)
Au/b-CDSH LC = 0.4 mg/mL Au HC = 3.7 mg/mL Au
D
d
H
D
d
H
1
1
1
1
:5
:1
:0.3
:0.2
1.7
1.8
2.0
2.9
2.7
2.4
0.9 Ä 1.6
2.0
1.9
5.4 Ä 3.0
2.4
2.6
In the fast-exchange limit, both parameters (P), that is, the dif-
fusion coefficients and the selective relaxation rates, are the
b f
weighted average of the values in the bound (P ) and free (P ) states
2.3
2.1
(
Eq. 2):

7
56
G. Uccello-Barretta et al. / Carbohydrate Research 346 (2011) 753–758
P
obs ¼ x
b
P
b
þ x
f
P
f
ð2Þ
the metal, which was confirmed also by the relevant broadening of
the NMR resonances of the b-GluSNa due to the presence of the
metal. In order to probe the ability of the corresponding water sol-
uble nanoparticles to bind DC, the solution containing the b-
GluSNa-capped gold nanoparticles was added to a four molar
excess of DC. Both the selective relaxation rates and the diffusion
coefficient of DC (Tables 3 and 4) were measured in order to detect
their eventual perturbations due to the presence of the nanoparti-
cles. However both kinds of NMR parameters remained nearly un-
changed as in the presence of pure b-GluSNa (Tables 3 and 4).
The potentialities of the MVS technique for a very fine control of
the gold nanoparticles sizes were confirmed, which was exploited
in aqueous environment by the use of heptakis(6-deoxy-6-
thio)cyclomaltoheptaose as the nanoparticles ligand by virtue of
the high sulfur to gold affinity. A very useful procedure was devel-
oped, where the addition of the cyclodextrin to acetone-solvated
nanoparticles led to the precipitation of cyclodextrin-bound gold
nanoparticles, which were easily re-dissolved in water. The proce-
dure not only was very simple, but also the use of potential chem-
ical contaminants was avoided.
At high gold/cyclodextrin molar ratios, the oxidation of the
cyclodextrin was favoured and under these conditions smaller
nanoparticles were selected irrespective of the gold concentration
in the starting Au/acetone SMA solution, probably due to the fact
that the disulfide bridges have a lower affinity to gold than that
the thiol groups bound to monomer cyclodextrin have. Further-
more the oxidized cyclodextrin has lower solubility which quicken
the nanoparticles precipitation. This last feature intrinsically limits
the use of gold nanoparticles to very diluted solutions. By contrast,
in the solutions containing high molar excesses of the cyclodextrin
to gold, only monomer cyclodextrin was present and the sizes of
the gold aggregates could be modulated by means of the concen-
tration of the solvated gold atoms coming from the MVS reactor.
Such kinds of cyclodextrin-capped soluble nanoparticles are very
suitable for the complexation of guests. The deoxycytidine was
complexed with very good affinity by inclusion in spite of the fact
that the original truncated cone shape of the cyclodextrin was
strongly perturbed as the consequence of the metal complexation
to the primary thiolated sites. The high degree of structural pre-
organization determined by the cyclic assembly of the glucopyra-
nose units of the cyclodextrin and the occurrence of inclusion phe-
where x
species.
b
and x
f
are the molar fractions of the bound and free
For the pure DC (100 mM) in the absence of complexing agents
ꢀ10
2 ꢀ1
we measured the diffusion coefficient of 4.8 ꢁ 10
Table 3) and the proton selective relaxation rates of the protons
H-a, H-f and H-g (see numbering in Chart 1) were 0.30, 0.39 and
m s
(
ꢀ1
0
.22 s (Table 4), respectively. In the presence of the cyclodextrin
(
DC/b-CDSH = 4:1) only a small decrease of the diffusion coefficient
ꢀ
10
2 ꢀ1
to the value of 4.3 ꢁ 10
m s was detected. Taking into ac-
count that the diffusion coefficient is the weighted average of its
value in the free and bound states (Eq. 2) and that the translational
diffusion of DC is mainly controlled by the cyclodextrin, we could
assume the diffusion coefficient of the bound DC equal to the dif-
fusion coefficient of the cyclodextrin and, hence, the DC bound mo-
lar fraction of 20% was calculated from Eq. 3.
D
D
obs ꢀ D
f
D
obs ꢀ D
f
x
b
¼
¼ D
ð3Þ
b
ꢀ D
f
bCDSH ꢀ D
f
The proton selective relaxation rates increased to the values of
ꢀ1
2
.16, 2.14 and 1.43 s for the protons H-a, H-f and H-g, respec-
tively, which pointed out about a six-fold increase of the NMR
parameter (see normalized R/R in the Table 4).
The solution of (Au) /b-CDSH nanoparticles which contains
D
n
monomer cyclodextrin (3.7 mg/mL of metal and a five molar excess
of the cyclodextrin with respect to the metal) was very stable dur-
ing the time and no oxidation/precipitation occurred. Thus, the
nanoparticles solution (25 mM in cyclodextrin) was added to DC
(
DC/b-CDSH = 4:1) and the NMR parameters were measured. The
diffusion coefficient remarkably lowered to the value of
ꢀ
10
2 ꢀ1
3
.5 ꢁ 10
m s (Table 3). Thus, a very high bound molar frac-
tion of 40% was obtained from Eq. 3 (using an averaged diffusion
ꢀ10
2 ꢀ1
coefficient for b-CDSH of 1.3 ꢁ 10
m s ). Relaxation measure-
ments confirmed the enhanced ability of the cyclodextrin in the
capped gold nanoparticles to bind the deoxycytidine, in fact a
ten-fold increase of the proton selective relaxation rates was de-
tected (Table 4), in agreement with the high complexation degree.
It is noteworthy that ROE measurements in the mixture DC/
(
n
Au) /b-CDSH NPs pointed to the occurrence of inclusion phenom-
ena as several dipolar interactions were detected between DC pro-
tons and the protons H-3 and H-5 of the cyclodextrin, which were
located in the inner part of the cyclodextrin.
Finally, starting from the more concentrated gold–acetone solu-
tion (3.7 mg/mL of metal), b-GluSNa-capped metal nanoparticles
were isolated by adding a five molar excess of the ligand and easily
re-dissolved in water. Hence the monosaccharide was able to bind
nomena played
processes. Accordingly, the 1-thio-b-
a
fundamental role in guests complexation
-glucose sodium salt, which
D
was by itself able to bind and stabilize the gold and which led to
the formation of water soluble gold nanoparticles, however, did
not demonstrate any ability to act as a complexing agent for the
deoxycytidine as in the case of the cyclodextrin.
Table 3
Diffusion coefficients (D, ꢁ10 m2
1
0
s
ꢀ1
2 n n
, D /b-CDSH NPs or b-GluSNa or (Au) /b-GluSNa NPs
O, 600 MHz, 25 °C) of DC alone (100 mM, Dfree) and in mixture with b-CDSH or (Au)
b
(complexing agent/DC = 1:4). Calculated bound molar fractions (x )
DC
4.8
b-CDSH + DC
(Au)
n
/b-CDSH NPs + DC
b-GluSNa + DC
n
(Au) /b-GluSNa NPs + DC
D
4.3
0.2
3.5
0.4
4.8
/
4.6
/
x
b
Table 4
Selective relaxation rates (R, sꢀ1, D
O, 600 MHz, 25 °C) and normalized selective relaxation rate variations (DR/R, D
R = Rmix ꢀ R) of selected protons of DC alone (100 mM, R) and in
2
mix
mixture (R , complexing agent/DC = 1:4) with b-CDSH or (Au)
n
/b-CDSH NPs or (Au)
n
/b-GluSNa NPs
DC
b-CDSH + DC
(Au)
n
/b-CDSH NPs + DC
(Au)
n
/b-GluSNa NPs + DC
R/R
R
Rmix
D
R/R
Rmix
D
R/R
Rmix
D
H-a
H-f
H-g
0.30
0.39
0.22
2.16
2.14
1.43
6.20
4.49
5.50
3.45
3.81
3.05
10.50
8.77
12.86
0.31
0.43
0.23
0.03
0.10
0.05

G. Uccello-Barretta et al. / Carbohydrate Research 346 (2011) 753–758
757
3
3
. Experimental
to 80 °C. The reaction mixture was stirred for 3 h at 80 °C, then
cooled to 40 °C and the major part of DMF was removed under
reduced pressure at 40 °C. The obtained crude product was recrys-
tallized several times from 96% EtOH (100 mL). The obtained or-
ange coloured product (13.1 g, 75%) was free from thiourea and
was found to be appropriate for further reactions. RF = 0.35–0.40
.1. General methods
The (Au)
scribed procedure was dissolved in D
dark-purple solution was transferred into a NMR tube under nitro-
gen atmosphere. NMR measurements were performed on
spectrometer operating at 600 and 150 MHz for H and C, respec-
tively. The temperature was controlled to ±0.1 °C. The 2D NMR
spectra were obtained by use of standard sequences and employ-
ing the minimal spectral width in both dimensions. g-COSY
n
/b-CDSH NPs solid obtained following the above de-
2
O (0.7 mL) and then the
10:7 dioxane/cc. NH
3
.
a
1
13
3.4. Heptakis(6-deoxy-6-thio)cyclomaltoheptaose (modified
2
3
method of Rojas et al.)
The thiuronium salt (12.2 g, 0.005 mol) was heated around
100 °C in deoxygenated (He/ultrasound) 0.25 M NaOH (120 mL)
for 4 h. As the reaction was completed the reaction mixture was
cooled to around 5–10 °C and the pH was adjusted to 1.0–1.5 with
3 M HCl solution, when the product started to crystallize. As the pH
became stable, the suspension was allowed to crystallize for
15 min, the solids were removed by filtration, washed with deoxy-
genated deionized water to around pH 5-6, then dried in vacuo at
(
gradient-COrrelation SpectroscopY) maps were acquired with a
relaxation delay of 2 s and 2–4 scans of 256 increments each were
collected. 2D TOCSY (TOtal Correlation SpectroscopY) spectra were
recorded acquiring 2–4 scans with 256 increments, 2 K data points,
2
s relaxation delay and a mixing time of 80 ms. The gradient
1
13
H, C-HSQC (Heteronuclear Single Quantum Correlation) maps
were obtained in 16 transients of 128 increments, with a relaxation
delay of 1 s. The 2D ROESY (Rotating-frame Overhauser Enhance-
ment SpectroscopY) spectra were recorded in the phase-sensitive
mode, with a mixing time of 300 ms. The pulse delay was main-
tained at 5 s; 256 increments of 4 scans and 2 K data points were
collected. Proton 1D ROESY spectra were recorded by using selec-
tive pulses generated by the Varian Pandora Software. The selec-
tive 1D ROESY spectra were acquired with 64 scans in 32 K data
points with a 3 s relaxation delay and a mixing time of 0.3 s. DOSY
2 5
rt in the presence of P O and KOH, and yielded 6.5 g of dark
orange, glassy crystals. The obtained product was recrystallized
by dissolution in deoxygenated 1 M NaOH solution (12 mL) and
precipitated with 1 M HCl (18 mL). Filtration and washing to
neutral pH with deoxygenated water afforded 3.8 g (61% theor.)
3
light orange crystals. RF = 0.00–0.05 10:7 dioxane/cc. NH .
3.5. Regeneration of thiol group from its oxidized state
(
Diffusion Ordered SpectroscopY) experiments were carried out by
using a stimulated echo sequence with self-compensating gradient
schemes, the minimal spectral width and 64 K data points. The val-
ues of
Partially oxidized heptakis(6-deoxy-6-thio)cyclomaltoheptaose
(6.2 g, approx. 0.010 mol) was suspended in deoxygenated water
and deoxygenated (He/ultrasound) 3 M NaOH solution (30 mL)
was added at rt. Sodium borohydride (1.1 g, 0.030 mol) is added
and stirred overnight. The occasionally found solids were filtered
through a sintered glass funnel, the filtrate was cooled with an
ice-water bath and the pH was adjusted to around 1.5–2 with
3 M HCl solution (approx. 45 mL). Upon acidification the product
started to crystallize, and as the pH became stable and foaming
stopped, the suspension was allowed to crystallize for an addi-
tional 15 min. The product was removed by filtration and washed
to pH 5–7 with deoxygenated deionized water, then dried in vacuo
D and of d were optimized for each sample and g was varied
in 30 steps to obtain an approximately 90–95% decrease in reso-
nance intensity at the largest gradient amplitudes. The baselines
of all arrayed spectra were corrected prior to processing the data.
After data acquisition, each FID was apodized with 1.0 Hz line-
broadening and Fourier transformed. The data were processed with
the DOSY macro (involving the determination of the resonance
heights of all the signals above a pre-established threshold and
the fitting of the decay curve for each resonance to a Gaussian
function) to obtain pseudo-two-dimensional spectra with NMR
chemical shifts along one axis and calculated diffusion coefficients
along the other. The selective relaxation rates were measured in
at rt in the presence of P
2
O
5
and KOH, yielded 3.2 g (ꢄ50%) b-CDSH
as a yellowish substance.
the initial rate approximation22 by employing a selective
at the selected frequency. After the delay , a non-selective
pulse was employed to detect the longitudinal magnetization.
p-pulse
1
H NMR (D O, 25 °C): d (ppm) 5.03 (H-1, d, J1,2 3.6 Hz, 7H), 3.83
2
s
p/2
(H-3, dd, J3,2 10.1 Hz, J3,4 9.3 Hz, 7H), 3.83 (H-5, m, 7H), 3.62 (H-4,
dd, J4,3 = J4,5 = 9.3 Hz, 7H), 3.52 (H-2, dd, J2,3 10.1 Hz, J2,1 3.6 Hz, 7H),
0
2
4
4
.94 (H-6, d, J6,6
0
13.6 Hz, 7H), 2.83 (H-6 , dd, J
6
0
,6 13.6 Hz, J
6
0
,5
1
3
.8 Hz, 7H). C NMR (D
), 72.9 (C-3/C-5), 72.3 (C-2), 26.0 (C-6).
2
O, 25 °C): d (ppm) 101.5 (C-1), 82.5 (C-
3
.2. Materials
1
-Thio-b-
agents, except
CycloLab), were from Sigma–Aldrich and were used without any
D
-glucose sodium salt (b-GluSNa) and all other re-
3
n
.6. (Au) /b-CDSH NPs from LC Au/acetone SMA
heptakis(6-deoxy-6-iodo)cyclomaltoheptaose
(
In a typical experiment a portion of low concentrated (LC,
additional purification. TLC plates were from Merck, type 5554
Silica Gel 60 F254. All the operations involving the use and the
preparation of the solvated metal atoms (SMAs) were performed
under dry argon atmosphere. Acetone was purified by conven-
tional methods, distilled and stored under argon. Gold pellets
0
.4 mg Au/mL) Au/acetone SMA (11.8 mL, 0.024 mmol Au) ob-
17
tained as previously described in the literature was added to a
solution of b-CDSH in water (5 mL). The mixture was then stirred
at rt and in the dark for about 30 min. During this period it was ob-
served the formation of a dark-purple solid which was isolated and
dried under reduced pressure. To obtain the samples containing
different Au/b-CDSH ratio different amounts of b-CDSH were used:
150 mg (Au/b-CDSH = 1:5); 30 mg (Au/b-CDSH = 1:1) and 10 mg
(Au/b-CDSH = 1:0.3).
(
99.999%) were purchased from Chimet S.p.A.
3
.3. Heptakis(6-deoxy-6-thiuronium)cyclomaltoheptaose
heptaiodide (modified method of Rojas et al.)
23
n 2
Au) O, 25 °C): d
/b-CDSH NPs (Au/b-CDSH = 1:5): 1H NMR (D
(
Dried heptakis(6-deoxy-6-iodo)cyclomaltoheptaose (14.3 g,
.0075 mol) was dissolved in DMF (85 mL), approx. 15 mL of
(ppm) 5.03 (H-1, d, J1,2 3.1 Hz, 7H), 3.80 (H-3, dd, J3,2 9.8 Hz, J3,4
9.3 Hz, 7H), 3.78 (H-5, m, 7H), 3.68 (H-4, dd, J4,3 = J4,5 = 9.3 Hz,
0
DMF was removed under reduced pressure and dried thiourea
8.0 g, 0.105 mol) was added at room temperature and heated up
7H), 3.50 (H-2, dd, J2,3 9.8 Hz, J2,1 3.1 Hz, 7H), 2.86 (H-6, dd, J6,6
0
0
(
13.6 Hz, J6,5 2.8 Hz, 7H), 2.81 (H-6 , br d, J
6 6
13.6 Hz, 7H).
0

7
58
G. Uccello-Barretta et al. / Carbohydrate Research 346 (2011) 753–758
2. Daniel, M. C.; Astruc, D. Chem. Rev. 2004, 104, 293–346.
3
n
.7. (Au) /b-CDSH NPs from HC Au/acetone SMA
3
4
.
.
Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2004, 43, 6042–6108.
Schmid, G. In Cluster and Colloids: From Theory to Applications; VCH: New York,
In a typical experiment a portion of high concentrated (HC,
.7 mg Au/mL) Au/acetone SMA (1.3 mL, 0.024 mmol Au) was
1994.
3
5. Nanotechnology in Catalysis; Ziou, B., Hermans, S., Somorjai, G. A., Eds.; Springer:
New York, 2004.
added to a solution of b-CDSH in water (2 mL). The mixture was
then stirred at rt and in the dark for about 30 min and 10 mL of
acetone was added to the mixture to facilitate the formation of a
dark-purple precipitate which was isolated and dried under re-
duced pressure. To obtain the samples containing different Au/b-
CDSH ratio different amounts of b-CDSH were used: 150 mg (Au/
b-CDSH = 1:5); 30 mg (Au/b-CDSH = 1:1); 10 mg (Au/b-CDSH =
6
7
.
.
Boisselier, E.; Astruc, D. Chem. Soc. Rev. 2009, 38, 1759–1782.
De, M.; Ghosh, P. S.; Rotello, V. M. Adv. Mater. 2008, 20, 4225–4241.
8. Kim, C. K.; Ghosh, P.; Rotello, V. M. Nanoscale 2009, 1, 61–67.
9.
Murphy, C. J.; Gole, A. M.; Stone, J. W.; Sisco, P. N.; Alkilany, A. M.; Goldsmith, E.
C.; Baxter, S. C. Acc. Chem. Res. 2008, 41, 1721–1730.
1
0. Rosi, N. L.; Mirkin, C. A. Chem. Rev. 2005, 105, 1547–1562.
11. Grabar, K. C.; Freeman, R. G.; Hommer, M. B.; Natan, M. J. Anal. Chem. 1995, 67,
35–743.
2. Liu, J.; Mendoza, S.; Roman, E.; Lynn, M. J.; Xu, R. L.; Kaifer, A. E. J. Am. Chem. Soc.
999, 121, 4304–4305.
7
1
1
:0.3) and 6 mg (Au/b-CDSH = 1:0.2).
.8. (Au) /b-GluSNa NPs from Au/acetone SMA
.8 mL (0.052 mmol Au) of high concentrated (HC, 3.7 mg Au/
1
1
1
3. Liu, J.; Alvarez, J.; Kaifer, A. E. Adv. Mater. 2000, 12, 1381–1383.
4. Park, C.; Youn, H.; Kim, H.; Noh, T.; Kook, Y. H.; Oh, E. T.; Park, H. J.; Kim, C. J.
Mater. Chem. 2009, 19, 2310–2315.
3
n
15. Klabunde, K. J. In Free Atoms, Clusters and Nanoscale Particles; Academic Press:
2
San Diego, 1994.
mL) Au/acetone SMA was added to a cooled solution (ice bath) of
b-GluSNa in water (2 mL). The mixture was stirred at rt for about
16. Vitulli, G.; Evangelisti, C.; Caporusso, A. M.; Pertici, P.; Panziera, N.; Bertozzi, S.;
Salvadori, P. Metal vapour-derived nanostructured catalysts in fine chemistry:
the role played by particle size in the catalytic activity and selectivity In Metal
Nanoclusters in Catalysis, Materials Science. The Issue of Size-Control; Corain, B.,
Schmid, G., Toshima, N., Eds.; Elsevier: Amsterdam, 2007. Chapter 32.
1
h and then left over night to facilitate the formation of a dark-
purple precipitate which was isolated and dried under reduced
pressure. To obtain a sample containing 1:5 Au/b-GluSNa ratio
17. Aronica, A.; Schiavi, E.; Evangelisti, C.; Caporusso, A. M.; Salvadori, P.; Vitulli,
G.; Bertinetti, L.; Martra, G. J. Catal. 2009, 266, 250–257.
5
8 mg of b-GluSNa were used.
18. Evangelisti, C.; Raffa, P.; Balzano, F.; Uccello-Barretta, G.; Vitulli, G.; Salvadori,
P. J. Nanosci. Nanotechnol. 2008, 8, 2096–2101.
1
H NMR (D
2
O, 25 °C): d (ppm) 4.40 (H-1, br d, J1,2 6.2 Hz, 1H),
0
0
19. Uccello-Barretta, G.; Evangelisti, C.; Raffa, P.; Balzano, F.; Nazzi, S.; Martra, G.;
3
4
.72 (H-6, d, J6,6
12.7 Hz, 1H), 3.53 (H-6 , dd, J
6
0
,6 12.7 Hz, J
6
0
,5
Vitulli, G.; Salvadori, P. J. Organomet. Chem. 2009, 694, 1813–1817.
.6 Hz, 1H), 3.29-3.27 (H-3/H-4/H-5, m, 3H), 2.88 (H-2, dd,
2
2
0. Hostetler, M. J.; Wingate, J. E.; Zhong, C. J.; Harris, J. E.; Vachet, R. W.; Clark, M.
R.; Londono, J. D.; Green, S. J.; Stokes, J. J.; Wignall, G. D.; Glish, G. L.; Porter, M.
D.; Evans, N. D.; Murray, R. W. Langmuir 1998, 14, 17–30.
J2,1 = J2,3 = 6.2 Hz, 1H).
1. Gaggelli, E.; Lepri, A.; Marchettini, N.; Ulgiati, S. Selective 1H NMR Relaxation
Delineation of Receptor Binding Equilibria In Advanced Magnetic Resonance
Techniques in Systems of High Molecular Complexity; Niccolai, N., Valensin, G.,
Eds.; Birkhauser: Boston, 1986; pp 109–117.
Acknowledgements
The work was supported by the MIUR (FIRB Project
RBPR05NWWC).
2
2
2. Freeman, R.; Wittekoek, S. J. Magn. Reson. 1969, 1, 238–276.
3. Rojas, M. T.; Koeniger, R.; Stoddart, J. F.; Kaifer, A. E. J. Am. Chem. Soc. 1995, 117,
336–343.
References
1
.
Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc.,
Chem. Commun. 1994, 7, 801–802.