A general approach using spiroborate reversible cross-linked Au nanoparticles for visual high-throughput screening of chiral vicinal diols

Article abstract of DOI:10.1039/c2sc21571c

Since enantiopure vicinal diols are important intermediates for the synthesis of numerous pharmaceutical and industrial products, enantioseparation of chiral vicinal diols has received much attention. Here we report a stepwise protocol for creating high-t

Full text of DOI:10.1039/c2sc21571c

Chemical Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
A general approach using spiroborate reversible
cross-linked Au nanoparticles for visual
Cite this: Chem. Sci., 2013, 4, 1156
high-throughput screening of chiral vicinal diols†
*
Weili Wei, Li Wu, Can Xu, Jinsong Ren and Xiaogang Qu
Since enantiopure vicinal diols are important intermediates for the synthesis of numerous pharmaceutical
and industrial products, enantioseparation of chiral vicinal diols has received much attention. Here we
report a stepwise protocol for creating high-throughput screening (HTS) assays for concentration and
enantiomeric excess (ee) of vicinal diols applied to asymmetric dihydroxylation (AD) reactions by using
spiroborate reversible cross-linked Au nanoparticles (AuNPs). The enantioselective assays have been
demonstrated by NMR spectroscopy and successfully used to rapidly analyze the AD reactions of trans-
stilbene with different reaction time and chiral ligands. The first and second steps involve the decoration
of a small library of chiral AuNPs with saccharides that possess chiral cis-vicinal diol sites, and
verification of the borate-directed assembly and disassembly of the chiral AuNPs. The third step
concerns discovery of the optimal chiral AuNPs for a given analyte. The fourth step involves the
evaluation of the accuracy and HTS performance of the method. The errors resulting from the analysis
of true unknowns are remarkably low, within 2.7% for ee and 0.05 mM for total concentration. The
method developed for hydrobenzoin has been applied to analyze the real AD reactions of trans-
stilbene. Since the enantioseparation is based on enantioselective ligand exchange (eLE) principle and
the reversibility of boron chemistry, this proof of concept approach can be easily adapted to other kinds
of asymmetric reactions by using relevant optical nanoprobes.
Received 21st September 2012
Accepted 19th December 2012
DOI: 10.1039/c2sc21571c
www.rsc.org/chemicalscience
While progress in this research area has been rapid, the use of
spiroborate for the construction of nanomaterial-based assem-
Introduction
blies is still scarce,8 particularly those with capability of chiral
recognition has not been reported yet.
The unique reversible properties of boron have attracted great
interest in the eld of dynamic-covalent chemistry.1 Both
boronic acids and the borate ion (B(OH)4ꢀ) are versatile building
blocks for reversible assembly. The use of the borate ion is
particularly appealing because one borate can bind two cis-
vicinal diols forming a spiroborate diester with dynamic-cova-
lent behavior under certain conditions or under action of certain
external chemical stimulus.1,2 This has promoted the production
of a wide range of borate molecular devices, such as membrane
transport carriers,3 chromatography solid supports,4 and che-
mosensors.5 Cross-linking reactions between borate ions and the
cis-diol sites on polysaccharide chains has also been utilized to
prepare polysaccharide-based reversible gels.6 Most recently,
evidence suggests that borate even functions in the primary cell
walls of plants, where it cross-links the pectic polysaccharide
rhamnogalacturonan II (RG-II) by forming spiroborate diesters.7
Among various nanomaterials, gold nanoparticles (AuNPs)
have been touted to be promising and useful in many applica-
tions because of their distinctive physical and optical proper-
ties. In particular, the plasmon resonance band of AuNPs can be
changed by modulating the distance between AuNPs along with
the color change of the AuNP solution.9 Reaction of borate ions
with AuNPs that contain cis-vicinal diols on their surface could
lead to AuNP networks that are composed of covalent, yet
readily reversible spiroborate diester linkages. The dynamic-
covalent nature of the spiroborate cross-links would allow the
AuNP networks to recongure their assembly structures in the
presence of external chemical stimulus such as cis-vicinal diols
that compete for bonding with the borate.1,2c That is, the spi-
roborate linkages between the AuNPs can be induced to disso-
ciate via the competitive reaction of diols with the central
borate. As a consequence, the resulting variation of the
distances among AuNPs would allow the translation of the weak
molecular interaction events into color changes, which can be
monitored by commonly used instruments such as a UV-vis
spectrometer or even the naked eye.
Laboratory of Chemical Biology, Division of Biological Inorganic Chemistry, State Key
Laboratory of Rare Earth Resource Utilization, Changchun Institute of Applied
Chemistry, Chinese Academy of Sciences, Changchun, 130022, P. R. China. E-mail:
xqu@ciac.jl.cn; Fax: +86 431-85262625
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2sc21571c
1156 | Chem. Sci., 2013, 4, 1156–1162
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
To further develop the spiroborate reversible cross-linked
AuNP assemblies, we endowed them with a chiral selective
responding capability through an enantioselective ligand
exchange (eLE) process.10 The process involves a competitive
binding event, between two chiral ligands (dened as chiral
selector and chiral analyte), to a center ion (Scheme S1, ESI†).
This has allowed successful enantioseparation of chiral vicinal
diols in capillary electrophoresis (CE) by using the borate as the
center ion.10 As proved by both experimental and computational
results,10,11 the eLE process relies on the energetic difference of
the two diastereometric ternary spiroborates that are composed
of chiral selector-borate-enantiomer of the analyte (suppose the
analyte has two enantiomers). Similarly, by decorating AuNPs
with chiral diols, the external chiral diol (analyte) triggered
dissociation of the spiroborate cross-links between the AuNPs
should also undergo the eLE process. This means different
enantiomers of the same chiral diol analyte would induce a
different reconguration of the AuNPs assemblies as well as a
different color change of the solution. Consequently, these
thoughts have motivated us to design a general approach for
applying spiroborate cross-linked AuNPs to a high-throughput
screening (HTS) assay of both enantiomeric excess (ee) and total
concentration ([Diol]t) of chiral vicinal diols, which has not
been reported.
Scheme 1 Proposed mechanism for the AuNPs-based determination of [Diol]t
and ee of chiral vicinal diols (appropriate to the work described herein for
hydrobenzoin), and the molecular structures of achiral and chiral modifiers.
D(A700/A520), change of absorbance ratios caused by the shift of AuNP's plasmon
resonance band; [Diol]t, total vicinal diol concentration; ee, enantiomeric excess.
The red frames indicate the cis-vicinal diol sites.15
including D-Lac-AuNPs, D-Mal-AuNPs, D-Ara-AuNPs and L-Ara-
AuNPs, allow us to quantify ee in addition to [Diol]t through the
eLE process (Scheme 1B). We use D-Lac, D-Mal, D-Ara and L-Ara
as chiral modiers of the AuNPs because they are naturally
chiral, readily available, and enantioselectivity of their terminal
residues (D-galactose, D-glucose, D-arabinose, and L-arabinose,
respectively) to chiral vicinal diols has been conrmed in eLE-
CE.10b,c Moreover, reversible cross-linking of saccharides via
spiroborate has been well explored in both articial molecular
devices3–6 and natural plant biology.7 Therefore, HTS assay of
both [Diol]t and ee could be achieved by batch monitoring the
color changes with a 96-well plate analysis system. Compared to
the previously reported chemosensor using enantiopure
boronic acid as chiral selector,17 this AuNPs-based strategy have
certain advantages such as ready availability of chiral reagents
(the natural saccharides) and more distinct color changes which
span a wider range from red to blue. As a result, an instant pre-
screening of both [Diol]t and ee by direct use of the naked eye
could be realized.
Chiral vicinal diols are important intermediates in the
synthesis of numerous pharmaceuticals and industrial prod-
ucts such as chiral drugs, pesticides, liquid crystal materials,
and ne chemicals.12 They can be produced from olens
through the well-known asymmetric dihydroxylation (AD)
reactions.13 To date, the ever-increasing demand for enantio-
pure vicinal diols has been accompanied by progress in the
design of catalytic AD systems and combinatorial methods that
can obtain large numbers of enantiopure products in a short
time. However, monitoring of asymmetric chemical trans-
formations, including ee and [Diol]t determination, remains the
main bottleneck in these processes because it usually entails
laborious and time-consuming chromatographic techniques10
or sophisticated instrument-demanded chiral nuclear magnetic
resonance (NMR) spectroscopy.14 Therefore, development of
spiroborate cross-linked AuNP assemblies for colorimetric
visual HTS assay of both ee and [Diol]t of general chiral vicinal
diols is of great importance.
To realize our strategy, AuNPs were decorated with thioctic
amides terminated in an achiral vicinal diol dopamine (DA) or
chiral saccharides including D-lactose (D-Lac), D-maltose (D-Mal),
D-arabinose (D-Ara), and L-arabinose (L-Ara) and cross-linked
Results and discussion
with borate ions by forming dynamic-covalent spiroborates A small library of chiral saccharide-decorated AuNPs (D-Lac-
between the cis-vicinal diol sites under alkaline conditions AuNPs, D-Mal-AuNPs, D-Ara-AuNPs and L-Ara-AuNPs) were
(Scheme 1).1,15 The AuNPs are cross-linked by multiple spi- prepared because different chiral analytes will not be best
roborate linkages and their gold cores are kept separate by the enantioselectively discriminated by one single chiral selector.
sufficient steric stabilization of the cyclic disulde anchoring of Therefore, we anticipated that it would be advantageous to
thioctic amides allowing reversible assembly.16 For the achiral screen each new analyte (chiral diols) with the library of chiral
DA-AuNPs assemblies, they could be dissociated by vicinal diol AuNPs.
analytes independent of chirality and the resulting blue shi of
Our rst goal was to prove the assembly of the modied
the plasmon resonance band of the AuNPs allows the determi- AuNPs with borate ions. As an example of the investigations,
nation of total concentration of both enantiomers ([Diol]t, spiroborate cross-linked assembly of D-Lac-AuNPs containing
Scheme 1A). On the other hand, assemblies of the chiral AuNPs, surface-bound D-Lac (0.04 mM) was complete at a molar ratio of
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 1156–1162 | 1157

View Article Online
Chemical Science
Edge Article
borate ion to the surface-bound D-Lac ¼ 1 : 2 equiv. (Fig. S1,
With the above results, enantioselective disassembly of the
ESI†). This was in accordance with the stoichiometric ratio of chiral AuNPs clusters with a chiral analyte (hydrobenzoin as a
the borate ion to D-Lac for forming spiroborate diesters (as model) was further investigated. Since the aggregation of AuNPs
cross-linkers) and the previously reported results.2b Thus the generally resulted in a signicant increase in the absorbance at
optimal molar ratio of borate ions to the surface-bound D-Lac 700 nm (A700) and decrease in the absorbance at 520 nm (A520),
(1 : 2 equiv.) was used in all subsequent procedures. Time- the ratio of A700 to A520 (A700/A520) was chosen to reect the
dependent experiments (Fig. 1A) showed that the borate ion- assembly of AuNPs.19 A higher ratio corresponded to assembled
directed assembly of D-Lac-AuNPs clusters took place as a rapid clusters of AuNPs with blue color whereas a lower ratio referred
red-shi of plasmon resonance band in the rst 2 min, and then to dispersed AuNPs with red color. Thus a representative
reached equilibrium over the next 3 min. In the presence of 0.02 enantiodisassembly of the D-Lac-AuNPs aggregates with (R,R)-
mM borate ions, the characteristic assembly time (sasm) of D- and (S,S)-hydrobenzoin was shown in Fig. 2. As expected, the
Lac-AuNPs was about 3 min (Fig. S2, ESI†), which was in A700/A520 values decreased enantiodifferently for the two enan-
accordance with the time scale of the reaction rate of forming tiomers. The enantioselective disassembly of the other designed
borate esters.18 The results indicated that the borate-directed chiral AuNPs with the two enantiomers of hydrobenzoin were
assembly was fast. Both dynamic light scattering (DLS) and also observed (Fig. S5–S7, ESI†).
transmission electron microscopy (TEM) experiments further
Aer succeeding in the test of enantioselective disassembly,
conrmed that borate could induce D-Lac-AuNPs aggregation we intended to screen out which chiral AuNPs could have the
(Fig. 1B–D). In a similar way, the borate-directed assembly of best discrimination between the two enantiomers for the model
other AuNPs (D-Mal-AuNPs, D-Ara-AuNPs, L-Ara-AuNPs, and DA- diol analyte (hydrobenzoin). Traditionally, acquiring this
AuNPs) was obtained at 1 : 2 equiv. of borate ion to each bound information would require time-consuming measurements of
diol (D-Mal, D-Ara, L-Ara, and DA) (Fig. S3A, ESI†). In addition, various UV-vis spectra as described above. For the purpose of
control experiments indicated that the UV-vis spectra of D-Lac- HTS, the screening was performed with a 96-well plate analysis
AuNPs were not affected by other common ions (Fig. S3B, ESI†) system. A “screening plate” was generated with chiral AuNPs
and amino or thiol group containing small molecules (Fig. S3C, (D-Lac-AuNPs, D-Mal-AuNPs, D-Ara-AuNPs, and L-Ara-AuNPs) and
ESI†), and UV-vis spectra of thioctic acid modied AuNPs were the optimum ratio (1 : 2 equiv.) of each AuNP with a borate ion
independent of borate ion (data not shown).
was added to each well of the plate (Fig. 3A).
Next, we tested disassembly of the AuNPs clusters in the
Previous studies17,20 indicated that the degree of enantiose-
presence of diol analytes. Fig. 1D showed characteristic TEM lectivity for chiral recognition system depends on analyte
images of the D-Lac-AuNPs assemblies which were dis- concentration. Thus, 3 concentration levels (0.05, 0.5, and
assembled upon incubation with a vicinal diol hydrobenzoin 5.0 mM) of the analyte were designed. As shown in Fig. 3A, the
(1.0 mM) for 10 min (Fig. 1E). The UV-vis spectrum of the D-Lac- analyte produced visual discernible colors from blue to red,
AuNPs solution was blue-shied, as expected following the which were dependent on [Diol]t and ee of the analyte, on the
dissociation of the assemblies (Fig. S4, ESI†).
chiral AuNPs. Thus the enantioselectivity of the chiral AuNPs
could be directly assessed by visual inspection under a
moderate concentration level (0.5 mM) of the analyte (Fig. 3A,
yellow frame). This was especially helpful to achieve HTS assays
because we might rapidly screen out the best chiral AuNPs by
the naked eye for an unknown analyte in practical applications.
Fig. 1 (A) Variation in the UV-vis spectra of D-Lac-AuNPs (containing surface-
bound D-Lac 0.04 mM) with time after the addition of borate ion at a concen-
tration of 0.02 mM. The measurements were carried out in water solutions con-
taining 20 vol% methanol (pH 9.5). (B) Size distribution of the particles as
determined by DLS for D-Lac-AuNPs (black), D-Lac-AuNPs in the presence of borate
ion (red), and D-Lac-AuNPs in the presence of borate ion and diol (blue); (C) TEM
images of D-Lac-AuNPs; TEM images of the assembly (D) and disassembly (E) of
D-Lac-AuNPs. Scale bar: 100 nm.
Fig. 2 Enantioselective disassembly titration of D-Lac-AuNPs networks with (R,R)-
hydrobenzoin and (S,S)-hydrobenzoin. All titrations were carried out in a mixture
of water solutions containing 20 vol% methanol (pH 9.5).
1158 | Chem. Sci., 2013, 4, 1156–1162
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
best enantioselectivity to diethyl tartrate and 1-phenylpropane-
1,2-diol, respectively. The results indicated that a different
chiral diol might show good response with a different chiral
AuNPs, which could be easily screened out from the “chiral
AuNPs library” with a screening plate as described above.
The enantioselectivity of the chiral AuNPs was further
conrmed by 1H NMR spectroscopy. As shown in Fig. S8 (ESI†),
clear enantio-splitting of the proton NMR signals of the
hydrobenzoin methines was monitored in D-Lac-AuNPs-borate
ion system. The rather large enantiodifference in chemical
shis (up to 0.05 ppm) robustly veried the existence of enan-
tioselectivity of D-Lac-AuNPs to the two enantiomers of hydro-
benzoin. Generally, splitting of the NMR signals of the
enantiomers derives from the formation of energetic different
diastereomers between the chiral selector and enantiomers.21
Thus the NMR results could further indicate the eLE process-
based formation of the ternary spiroborates diastereomers
(D-Lac-borate-hydrobenzoin). The enantio-splitting of the
proton NMR signals of the hydrobenzoin methines in D-Ara-
Fig. 3 (A) Image of the enantioselectivity of the screening plate. The distinct
colored micro-wells indicated the different response of the chiral AuNPs to the
hydrobenzoin with different concentrations (0.05, 0.5, 5.0 mM) of (R,R)-hydro- AuNPs systems was also observed (Fig. S9, ESI†).
benzoin, (S,S)-hydrobenzoin and racemic mixture (Rac as indicated in the image).
Based on the results from the screening plate, the ability of
the best D-Lac-AuNPs to simultaneously determine [Diol]t and ee
was explored. For this purpose, a “standard plate” with chiral D-
UV-vis measurements of the screening plate showing enantiorecognition of D-Lac-
AuNPs (B), D-Mal-AuNPs (C), D-Ara-AuNPs (D), and L-Ara-AuNPs (E) with (R,R)-
hydrobenzoin (0.5 mM), racemic mixture (0.5 mM) and (S,S)-hydrobenzoin (0.5
Lac-AuNPs (Fig. 4, columns 1–11) and achiral DA-AuNPs were
made (Fig. 4, column 12). The standard ee titrations of hydro-
benzoin were carried out at 6 different [Diol]t (0.1, 0.2, 0.4, 0.6,
0.8, and 1.0 mM) and 11 different (R,R)-hydrobenzoin ee values
mM). All the measurements were performed in the mixture of water solutions
containing 20 vol% methanol (pH 9.5).
Of the designed chiral AuNPs, D-Lac-AuNPs showed the highest (ꢀ100, ꢀ80, ꢀ60, ꢀ40, ꢀ20, 0, 20, 40, 60, 80, and 100)% with D-
enantioselectivity (Fig. 3B–E) and D-Mal-AuNPs did not show Lac-AuNPs. The standard total concentration ([Diol]t) titration
apparent enantioselectivity (Fig. 3C). Both D-Ara- and L-Ara- of hydrobenzoin was carried out at 6 different concentrations
AuNPs showed moderate discrimination between the two (0.1, 0.2, 0.4, 0.6, 0.8, and 1.0 mM) with DA-AuNPs. Because DA-
enantiomers of hydrobenzoin.
AuNP was achiral, it responded only to the change in [Diol]t of
Signicantly, the opposite enantiomer-modied chiral the analyte, whereas D-Lac-AuNPs responded to both the change
AuNPs (D-Ara- and L-Ara-AuNPs) showed equal and opposite in [Diol]t and ee of the analyte.
enantioselectivity, which was indicated by the opposite transi-
The A700/A520 signals of each micro-wells of the standard
tion of color (Fig. 3A) and the opposite variation of UV-vis plate were recorded by a plate reader for the [Diol]t and ee values
spectra (Fig. 3D and E). This result was in good accordance with of hydrobenzoin as described above were shown in Fig. 5. As
the rst principle of stereochemistry,17 and indicated the commonly adopted in the previous chiral assays,22 the A700/A520
robustness of our method.
values from ee and [Diol]t titrations were linearly tted against
Given these results, the screening plate was further quanti-
tatively analyzed with a microplate reader by monitoring the
absorbance at 700 and 520 nm. As expected, 96-well plate
analysis showed that the two enantiomers of hydrobenzoin were
highly enantiodiscriminated by D-Lac-AuNPs. An enantiose-
lectivity (dened as (A700/A520)SS/(A700/A520)RR)in the AuNPs
signal (A700/A520) ratio of 4.77 was determined with D-Lac-AuNPs
at 0.5 mM analyte. The enantioselectivity of each chiral AuNPs
was listed in Table S1.† Briey, the results of microplate analysis
and visual inspection suggested the same enantioselectivity
order of the chiral AuNPs: D-Lac-AuNPs > D-Ara-AuNPs ¼ L-Ara-
AuNPs > D-Mal-AuNPs, thereby validating the reliability of the
screening process.
To display the generality of our method, the enantiose-
lectivity of the chiral AuNPs to two other chiral diols (diethyl
tartrate and 1-phenylpropane-1,2-diol) was investigated. As
shown in Table S2,† both pairs of enantiomers were successfully
Fig. 4 Layout of the standard plate. Solvent: water–methanol mixture (80 : 20,
v/v), pH 9.5. The (R,R)-hydrobenzoin ee values of column 1–11 were (ꢀ100, ꢀ80,
enantiorecognized. D-Mal-AuNPs and D-Lac-AuNPs showed the ꢀ60, ꢀ40, ꢀ20, 0, 20, 40, 60, 80, 100)%, respectively.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 1156–1162 | 1159

View Article Online
Chemical Science
Edge Article
Table 1 Determination of reaction yields and ee values of the four catalytic AD
ee and [Diol]t, respectively. The tting results were satisfying
(correlation coefficients R2 > 0.99).
reactions by our HTS and the CE standard methods
With the above results, the ability of our method for HTS
assay of unknown samples for [Diol]t and ee was investigated.
To do this, an “analysis plate” was made where the unknown
samples with varying [Diol]t and ee were tested. By monitoring
the A700/A520 values, the [Diol]t and ee values were predicted
according to the standard equations generated from the stan-
dard plate (Table S3†). The error for the unknown samples was
calculated in the form of average error was determined to be
ꢁ0.05 mM for [Diol]t and ꢁ2.72% for ee. The ee correlation
graph between the determined values and the actual values had
a regression (R2) of 0.9961, indicating good accuracy (Fig. S10,
ESI†).
Overall, the method is rapid once a standard plate has been
developed for a particular analyte. Among all the procedures,
loading of the AuNPs and unknown sample solutions to the 96-
well plate requires about 30 min per 96 samples. Aer that, the
[Diol]t and ee values of 96 unknown samples can be predicted
within 10 min, including z5 min for the 96-well plate reader to
record the absorbance of 96 unknowns samples at two wave-
lengths (520 and 700 nm), and 5 min for data analysis with a
computer program.
Yielda (%)
eeb (%)
Ligands
Time (h)
HTS
CE
HTS
CE
85.7
96.7
ꢀ83.8
ꢀ96.2
(DHQD)2PHAL
(DHQD)2PHAL
(DHQ)2PHAL
(DHQ)2PHAL
7
15
7
81.6
87.5
80.1
90.7
80.4
86.2
78.9
89.2
81.4
92.8
ꢀ88.2
15
ꢀ92.4
The reaction yields were calculated from [Diol]t. b (R,R)-hydrobenzoin
a
ee values.
discovery of more efficient chiral ligands. Two commercially
available chiral ligands, hydroquinidine 1,4-phthalazinediyl
diether (DHQD)2PHAL and hydroquinine 1,4-phthalazinediyl
diether (DHQ)2PHAL were examined. The literature reported
that catalytic AD of trans-stilbene with (DHQD)2PHAL and
(DHQ)2PHAL formed (R,R)- and (S,S)-hydrobenzoin with highly
enantiopurity, respectively.13,23
AD reactions of trans-stilbene with the two chiral ligands
were performed according to the literature procedures.23 Anal-
ysis of the crude reaction products was carried in two steps: the
reaction yield was rstly identied through the determination
of [Diol]t on a DA-AuNPs generated “analysis plate”; and
according to determined [Diol]t, another “analysis plate” was
generated with D-Lac-AuNPs for the assay of ee values. Although
such a “two step” assay was already high-throughput and rapid,
the assay could be further improved on a single “analysis plate”
with the help of Statistica Neural Network soware as reported
by Anslyn's group.17
The nal performance of this method was evaluated with
real samples produced from catalytic AD reactions. The AD
reactions generally involve a chiral ligand which plays crucial
role in reaction yield and enantioselectivity of its chiral diol
products.13 A method which can be used for the rapid assay of
the reaction yield and ee of AD reactions is highly desired in the
Four AD reactions of trans-stilbene were carried out with
different reaction time (7 and 15 h) and chiral ligands
((DHQD)2PHAL and (DHQ)2PHAL). The reaction yield and ee
values predicted by our HTS method and the standard CE
method12 were listed in Table 1. Apparently, the results of the
HTS method compared favorably with those from the standard
CE,12 indicating that the co-reactants of the AD reactions did not
interfere with the determination accuracy of our HTS method.
Presumably, using this method, dozens to hundreds of AD
reactions could be simultaneously analyzed in a few minutes.
However, a single CE separation of hydrobenzoin enantiomers
requires much longer time (Fig. S11, ESI†). We also found that
the sample solutions displayed different colors in the range
from blue to red as the yield and ee varied. This allowed us to
semi-quantitatively screening the AD reactions with the naked
eye.
Conclusions
On the basis of the reversibility of boron chemistry and the
plasmon resonance character of AuNPs, we demonstrate, for the
rst time, that spiroborate cross-linked AuNP hybrids show
highly enantioselectivity and can be a HTS assay of ee and [Diol]t
of chiral vicinal diols. The underlying enantiorecognition events
have been further conrmed by NMR spectroscopy, and the
method has been successfully used to rapidly analyze real AD
Fig. 5 Analysis of the standard plate. (A) [Diol]t titration of DA-AuNPs at 6
different concentrations of racemic hydrobenzoin. (B) (R,R)-hydrobenzoin ee
titration of D-Lac-AuNPs (0.04 mM surface bound D-Lac with 0.02 mM borate ion)
at 6 different [Diol]t: 0.1 (-), 0.2 (C), 0.4 (:), 0.6 (;), 0.8 (A), and 1.0 mM (=).
All samples were made in the mixture of water solutions containing 20 vol%
methanol (pH 9.5).
1160 | Chem. Sci., 2013, 4, 1156–1162
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
reactions of trans-stilbene with different reaction time and
A “screening plate” was designed at 3 different analyte
chiral ligands. Since this proof of concept approach is based on (hydrobenzoin) concentrations 0.05, 0.50, 5.00 mM, and the
enantioselective ligand exchange principle and the reversibility each chiral AuNP and borate ions (surface-bound diols : borate
of boron chemistry, it can be easily adapted to other kinds of ¼ 2 : 1 equiv.) were added to each well of the microplate. The
asymmetric reactions by using relevant optical nanoprobes. On establishment of standard equations for ee and total concen-
the other hand, chiral selective assembly of nanoparticles could tration ([Diol]t) was combined on a “standard plate” to speed up
also be achieved in other manners by modifying with different the analysis. The standard plate was designed with D-Lac-AuNPs
chiral molecules.24
(chiral assays for ee) and DA-AuNPs (achiral assays for [Diol]t).
The layer out for the standard plate was such that the concen-
tration of the analyte (hydrobenzoin) would vary along each row
of the plate, whereas the ee of the solution varied from 100% to
ꢀ100% (column 1–11) and column 12 was racemic hydro-
benzoin for achiral assay (Fig. 4). The “analysis plates” were
prepared for sample analysis, and the amount of chiral/achiral
AuNPs and borate (including other conditions) were the same to
the standard plate.
Experimental section
Materials
Gold(III) chloride trihydrate (HAuCl4), trisodium citrate dihy-
drate (citrate), thioctic acid, boric acid, D-lactose, D-maltose,
were purchased from Sigma Aldrich (St. Louis, MO). D-Arabi-
nose, L-arabinose, (R,R)-hydrobenzoin, (S,S)-hydrobenzoin,
diethyl D-tartrate, diethyl L-tartrate, (1R,2R)-1-phenylpropane-
1,2-diol, and (1S,2S)-1-phenylpropane-1,2-diol were purchased
from Fisher Scientic (Pittsburgh, PA). 1-Ethyl-3(3-dimethyla-
minopropyl) carbodiimide hydrochloride (EDC$HCl), and
N-hydroxysuccinimide (NHS) were purchased from Alfa Aesar
(Ward Hill, MA). Dialysis membrane was purchased from Pierce
(3000 MWCO). All other regents were of analytical reagent
grade. Ultra-pure water (18.2 MU cmꢀ1, Milli-Q, Millipore) was
used for all experiments.
Acknowledgements
This work was supported by 973 Project (2011CB936004,
2012CB720602), NSFC (20831003, 91213302, 20833006,
21210002), and Funds from the Chinese Academy of Sciences.
Notes and references
1 (a) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders
and J. F. Stoddart, Angew. Chem., Int. Ed., 2002, 41, 898–952;
(b) R. Nishiyabu, Y. Kubo, T. D. James and J. S. Fossey, Chem.
Commun., 2011, 47, 1124–1150; (c) R. Nishiyabu, Y. Kubo,
T. D. James and J. S. Fossey, Chem. Commun., 2011, 47,
1106–1123.
2 (a) A. Deutsch and S. Osoling, J. Am. Chem. Soc., 1949, 71,
1637–1640; (b) E. W. Malcolm, A Study of the Borate-
Carbohydrate Complex Formed in an Aqueous Medium,
Appleton, Wisconsin, 1964; (c) A. P. Bapat, D. Roy,
J. G. Ray, D. A. Savin and B. S. Sumerlin, J. Am. Chem. Soc.,
2011, 133, 19832–19838.
Instrumentation
Ultraviolet-visible spectroscopy (UV-vis) measurements were
recorded on a Jasco-V550 UV-vis spectrophotometer. Fourier
transform infrared spectroscopy (FTIR) measurements were
carried out with a BRUKER Vertex 70 FTIR spectrometer. The
sample was prepared as pellets using spectroscopic grade KBr.
Transmission electron microscopic (TEM) experiments were
performed using a Philips Tacnai G2 20 S-TWIN microscope
operating at 200 kV. For visualization by TEM, samples were
prepared by dropping a solution of production on a copper grid.
The dynamic light scattering (DLS) measurements were carried
out using a Zeta PALS analyzer (Brookhaven Instruments Corp.
3 J. A. Riggs, R. K. Litcheld and B. D. Smith, J. Org. Chem.,
1996, 61, 1148–1150.
4 R. Mathur, S. Bohra, V. Mathur, C. Narang and N. Mathur,
Chromatographia, 1992, 33, 336–338.
1
Holtsville, NY). H nuclear magnetic resonance (NMR) spectra
were recorded at 25 ꢂC on a Bruker Avance 600 MHz NMR
Spectrometer. X-ray photoelectron spectroscopy (XPS) data were
obtained with an ESCALab220i-XL electron spectrometer from
VG Scientic using 300 W AlKa radiation. The base pressure was
about 3 ꢃ 10ꢀ9 mbar. The binding energies were referenced to
the C 1s line at 284.8 eV from adventitious carbon. Curve tting
was performed using a Gaussian–Lorentzian peak shape.
5 (a) P. D. Beer and P. A. Gale, Angew. Chem., Int. Ed., 2001, 40,
486–516; (b) N. Kameta and K. Hiratani, Tetrahedron Lett.,
2006, 47, 4947–4950; (c) N. Kameta and K. Hiratani, Chem.
Commun., 2005, 725–727.
6 (a) S. Gao, J. Guo and K. Nishinari, Carbohydr. Polym., 2008,
72, 315–325; (b) E. Pezron, A. Ricard, F. Lafuma and
R. Audebert, Macromolecules, 1988, 21, 1121–1125.
7 (a) M. A. O'Neill, S. Eberhard, P. Albersheim and A. G. Darvill,
Science, 2001, 294, 846–849; (b) T. L. Holdaway-Clarke,
N. M. Weddle, S. Kim, A. Robi, C. Parris, J. G. Kunkel and
P. K. Hepler, J. Exp. Bot., 2003, 54, 65–72.
8 S. Tamesue, M. Numata, K. Kaneko, T. D. James and
S. Shinkai, Chem. Commun., 2008, 4478–4480.
9 (a) Y. Liu, X. Han, L. He and Y. Yin, Angew. Chem., Int. Ed.,
2012, 51, 6373–6377; (b) M.-C. Daniel and D. Astruc, Chem.
Rev., 2004, 104, 293–346.
The 96-well plate analysis
Arrays were made by mixing AuNPs, borate ion, and analyte
stock solutions within Corning 96-well polystyrene plates.
Absorbance spectra were recorded at ambient temperature on a
Bio-Gen Technology PHERAstar FS multidetection microplate
reader. Each well contained a total solution volume of 300 mL.
Aer making the plate, it was sealed to prevent solvent evapo-
ration. All samples were equilibrated for at least 10 min before
ꢂ
measurements. All measurements were taken at 25 C.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 1156–1162 | 1161

View Article Online
Chemical Science
Edge Article
10 (a) S. Kodama, A. Yamamoto, R. Iio, K. Sakamoto,
A. Matsunaga and K. Hayakawa, Analyst, 2004, 129, 1238–
1242; (b) S. Kodama, S. Aizawa, A. Taga, T. Yamashita and
S. H. Shabbir, L. A. Joyce, G. M. da Cruz, V. M. Lynch,
S. Sorey and E. V. Anslyn, J. Am. Chem. Soc., 2009, 131,
13125–13131.
A. Yamamoto, Electrophoresis, 2006, 27, 4730–4734; (c) 18 S. Xing, Y. Guan and Y. Zhang, Macromolecules, 2011, 44,
S. Kodama, S.-I. Aizawa, A. Taga, T. Yamashita, T. Kemmei, 4479–4436.
A. Yamamoto and K. Hayakawa, Electrophoresis, 2007, 28, 19 (a) J. Wang, L. Wu, J. Ren and X. Qu, Small, 2012, 8, 259–264;
3930–3933; (d) M. G. Schmid, N. Grobuschek, O. Lecnik
(b) D. Li, A. Wieckowska and I. Willner, Angew. Chem., Int.
Ed., 2008, 47, 3927–3931; (c) X. Xie, W. Xu and X. Liu, Acc.
Chem. Res., 2012, 45, 1511–1520.
¨
and G. J. Gubitz, J. Biochem. Biophys. Methods, 2001, 48,
143–154.
11 J. A. Raskatov, J. M. Brown and A. L. Thompson, 20 (a) W. Wei, K. Qu, J. Ren and X. Qu, Chem. Sci., 2011, 2, 2050–
CrystEngComm, 2011, 13, 2923–2929.
12 P. Liu, W. He, Y. Zhao, P.-A. Wang, X.-L. Sun, X.-Y. Li and
S.-Y. Zhang, Chirality, 2008, 20, 75–83.
2056; (b) W. Wei, B.-Y. Guo and J.-M. Lin, J. Chromatogr., A,
2009, 1216, 1484–1489; (c) W. Wei, B. Guo and J.-M. Lin,
Electrophoresis, 2009, 30, 1380–1387.
13 H. C. Kolb, M. S. VanNieuwenhze and K. B. Sharpless, Chem. 21 (a) C. Naumann, W. A. Bubb, B. E. Chapman and
Rev., 1994, 94, 2483–2547.
P. W. Kuchel, J. Am. Chem. Soc., 2007, 129, 5340–5341; (b)
G. Uccello-Barretta, F. Balzano, R. Menicagli and
P. Salvadori, J. Org. Chem., 1996, 61, 363–365.
´
14 A. M. Kelly, Y. Perez-Fuertes, J. S. Fossey, S. L. Yeste,
S. D. Bull and T. D. James, Nat. Protoc., 2008, 3, 215–219.
15 (a) T. Ishii and H. Ono, Carbohydr. Res., 1999, 321, 257–260; 22 (a) G. Mirri, S. D. Bull, P. N. Horton, T. D. James, L. Male and
(b) J. A. Riggs, R. K. Litcheld and B. D. Smith, J. Org. Chem.,
1996, 61, 1148–1150; (c) P. Britz-McKibbin and D. D. Y. Chen,
Anal. Chem., 2000, 72, 1242–1252; (d) R. A. Wallingford and
A. G. Ewing, Anal. Chem., 1989, 61, 98–100.
J. H. R. Tucker, J. Am. Chem. Soc., 2010, 132, 8903–8905; (b)
J. F. Folmer-Andersen, V. M. Lynch and E. V. Anslyn, J. Am.
Chem. Soc., 2005, 127, 7986–7987; (c) X. F. Mei and
C. Wolf, J. Am. Chem. Soc., 2006, 128, 13326–13327.
16 (a) S.-Y. Lin, Y.-T. Tsai, C.-C. Chen, C.-M. Lin and C.-H. Chen, 23 (a) E. N. Jacobsen, I. Marko, W. S. Mungall, G. Schroder and
J. Phys. Chem. B, 2004, 108, 2134–2139; (b) J. M. Abad,
K. B. Sharpless, J. Am. Chem. Soc., 1988, 110, 1968–1970; (b)
T. P. Yoon and E. N. Jacobsen, Science, 2003, 299, 1691–
1693.
´
S. F. L. Mertens, M. Pita, V. M. Fernandez and
D. J. Schiffrin, J. Am. Chem. Soc., 2005, 127, 5689–5694.
17 (a) S. H. Shabbir, C. J. Regan and E. V. Anslyn, Proc. Natl. 24 D. Balogh, Z. Zhang, A. Cecconello, J. Vavra, L. Severa,
Acad. Sci. U. S. A., 2009, 106, 10487–10492; (b) F. Teply and I. Willner, Nano Lett., 2012, 12, 5835–5839.
1162 | Chem. Sci., 2013, 4, 1156–1162
This journal is ª The Royal Society of Chemistry 2013