Catalytic signal amplification for the discrimination of ATP and ADP using functionalised gold nanoparticles

Article abstract of DOI:10.1039/c6ob00993j

Diagnostic assays that incorporate a signal amplification mechanism permit the detection of analytes with enhanced selectivity. Herein, we report a gold nanoparticle-based chemical system able to differentiate ATP from ADP by means of catalytic signal amplification. The discrimination between ATP and ADP is of relevance for the development of universal assays for the detection of enzymes which consume ATP. For example, protein kinases are a class of enzymes critical for the regulation of cellular functions, and act to modulate the activity of other proteins by transphosphorylation, transferring a phosphate group from ATP to give ADP as a byproduct. The system described here exploits the ability of cooperative catalytic head groups on gold nanoparticles to very efficiently catalyze chromogenic reactions such as the transphosphorylation of 2-hydroxypropyl-4-nitrophenyl phosphate (HPNPP). A series of chromogenic substrates have been synthesized and evaluated by means of Michaelis-Menten kinetics (compounds 2, 4-6). 2-Hydroxypropyl-(3-trifluoromethyl-4-nitro)phenyl phosphate (5) was found to display higher reactivity (k_cat) and higher binding affinity (K_M) when compared to HPNPP. This higher binding affinity allows phosphate 5 to compete with ATP and ADP to different extents for binding on the monolayer surface, thus enabling a catalytically amplified signal only when ATP is absent. Overall, this represents a viable new approach for monitoring the conversion of ATP into ADP with high sensitivity.

Full text of DOI:10.1039/c6ob00993j

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Catalytic signal amplification for the
discrimination of ATP and ADP using
Cite this: DOI: 10.1039/c6ob00993j
functionalised gold nanoparticles†
Cristian Pezzato, Jack L.-Y. Chen, Patrizia Galzerano, Michela Salvi and
Leonard J. Prins*
Diagnostic assays that incorporate a signal amplification mechanism permit the detection of analytes with
enhanced selectivity. Herein, we report a gold nanoparticle-based chemical system able to differentiate
ATP from ADP by means of catalytic signal amplification. The discrimination between ATP and ADP is of
relevance for the development of universal assays for the detection of enzymes which consume ATP. For
example, protein kinases are a class of enzymes critical for the regulation of cellular functions, and act to
modulate the activity of other proteins by transphosphorylation, transferring a phosphate group from ATP
to give ADP as a byproduct. The system described here exploits the ability of cooperative catalytic head
groups on gold nanoparticles to very efficiently catalyze chromogenic reactions such as the transphos-
phorylation of 2-hydroxypropyl-4-nitrophenyl phosphate (HPNPP). A series of chromogenic substrates
have been synthesized and evaluated by means of Michaelis-Menten kinetics (compounds 2, 4–6).
2-Hydroxypropyl-(3-trifluoromethyl-4-nitro)phenyl phosphate (5) was found to display higher reactivity
(kcat) and higher binding affinity (KM) when compared to HPNPP. This higher binding affinity allows phos-
phate 5 to compete with ATP and ADP to different extents for binding on the monolayer surface, thus
enabling a catalytically amplified signal only when ATP is absent. Overall, this represents a viable new
approach for monitoring the conversion of ATP into ADP with high sensitivity.
Received 6th May 2016,
Accepted 16th June 2016
DOI: 10.1039/c6ob00993j
www.rsc.org/obc
protein kinases as well as other enzymes which involve ATP
consumption.
Introduction
The detection of low levels of proteins and other biomarkers is
Assays for ATP consumption traditionally rely on the use of
of crucial importance for the early diagnosis of diseases.1–3 In 32P-labeled ATP to ensure high sensitivity.9 However, this
particular, the development of technology that allows naked comes with disadvantages such as the short life-time of 32P-
eye detection of target molecules without the need for complex labelled ATP and the generation of radioactive waste. These tra-
instrumentation is the utmost challenge.4 In particular, assays ditional protocols are also laborious and non-conducive to
that allow for the detection of ATP consumption are important high-throughput screening. Due to these limitations, there has
as ATP is used as a source of energy for numerous biochemical been much research towards non-radioactive assays for ATP
processes. For example, protein kinases are a class of enzymes consumption,10 with particular emphasis on techniques based
that play critical roles in a variety of cellular functions includ- on fluorescence.11–16 While exciting, many of these alternative
ing cell growth, development, differentiation, membrane trans- methodologies rely on elaborate constructs involving fluoro-
port, and cell death.5–8 They exert their function by phores conjugated to enzymes or antibodies. More recent work
transferring the γ-phosphate group of ATP to a serine, threo- has emphasised the need for novel methodologies that rely on
nine or tyrosine residue in specific substrate proteins (forming less-complex elements.17–24 Herein, we present novel methodo-
ADP as a by-product), which are crucially involved in multiple logy using simple and robust components able to visually dis-
signaling pathways within the body. Thus the ability to differ- criminate ATP and ADP at low concentrations by means of
entiate ADP from ATP could provide a universal assay for catalytic signal amplification.
The focus here is on the use of catalytic signal amplifica-
tion.25 This allows for highly sensitive assays where a single
analyte molecule triggers the activation of a signal amplifica-
Department of Chemical Sciences, University of Padova, Via Marzolo 1,
35131 Padova, Italy. E-mail: leonard.prins@unipd.it
†Electronic supplementary information (ESI) available. See DOI:
tion pathway which leads to the generation of a large number
of reporter molecules.4 This marks a strong contrast with con-
10.1039/c6ob00993j
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
ventional chemosensing assays in which a fixed correlation
exists between the amount of analyte and the output signal.
Signal amplification mechanisms typically rely on the use of
multivalent systems or catalysts,25–29 in which the analyte
affects the properties of a multitude of reporter molecules or
activates an (auto)catalytic pathway. Recently, we developed a
catalytic signal amplification assay for the detection of pro-
tease activity that relied on the use of gold nanoparticles
(Au NP 1, Fig. 1) passivated with a monolayer of thiols terminat-
ing with 1,4,7-triazacyclonane (TACN)·Zn2+ head groups.30,31
This system catalyses very efficiently the transphosphoryla-
tion of 2-hydroxypropyl-4-nitrophenylphosphate (HPNPP 2;
Scheme 1a), a model substrate for RNA hydrolysis,32–38 due to
the cooperative action between neighbouring TACN·Zn2+ head
groups in the monolayer.39,40 This reaction liberates p-nitro-
phenol (3a), which is easily detectable by the naked eye. In
addition, negatively charged oligoanions such as peptides and
nucleotides (e.g. ATP and ADP) have been found to have very
Scheme 1 (a) Transphosphorylation of HPNPP 2 catalysed by Au NP 1
to form a cyclic organic phosphate with the release of p-nitrophenol 3a;
(b) the different analogues of HPNPP investigated in this work.
HPNPP that both bound under saturation conditions (as in the
lower hemisphere of Fig. 1). Consequently, ATP → ADP conver-
high affinity for the surface of nanoparticles because of the sion never resulted in reactivation of the catalytic activity (as in
high density of positive charges in the monolayer.41–45
strong correlation between the binding affinity and the
A
the upper hemisphere of Fig. 1). This is regrettable, because
the ability to detect ATP → ADP conversion would provide
number of charges present in the oligoanions was observed.46 access to a universal assay for the detection of protein kinases
Importantly, the binding of negatively charged oligoanions to
using catalytic signal amplification.
the surface of Au NP 1 blocks the binding sites for the sub-
Here, we were interested to investigate whether this supra-
strate HPNPP 2, thus inhibiting catalysis. The working prin- molecular approach could be customized to allow for the dis-
ciple of our previous assay for protease activity relied on the
use of negatively charged peptides which acted both as (1) sub-
crimination of ATP and ADP. As mentioned above, the
addition of either ATP or ADP to a system containing Au NP 1
strates of the target proteases and (2) as inhibitors of the and HPNPP 2 caused an immediate and quantitative inhi-
HPNPP cleavage activity of the gold nanoparticles. Cleavage of
the substrates by proteases resulted in fragments with less
bition of catalytic activity. This implies that the affinity of the
substrate HPNPP 2 for Au NP 1 was just too weak when com-
overall negative charge, thus possessing lower inhibitory power pared to ADP. What is required is an HPNPP analogue with a
for the gold nanoparticles, causing a turn on of catalytic
activity (HPNPP hydrolysis) and the formation of reporter
higher affinity for Au NP 1, strong enough to displace ADP, but
not ATP. Recent studies by us and others have demonstrated
molecules (3a). It was shown that the selectivity of the assay that the binding of small molecules on the monolayer
towards different enzyme targets could be tuned by changing
the peptide sequence. However, this system was not able to
depends on a combination of noncovalent interactions, includ-
ing hydrophobic ones.47–53 Thus, we reasoned that by modify-
differentiate ATP and ADP because ATP and ADP both had ing the chemical structure of HPNPP 2 it could be possible to
such high affinities for the gold nanoparticles compared to
make a new substrate with a higher affinity (KM) for Au NP 1.
Fig. 1 The lower hemisphere shows ATP binding to the surface of Au NP 1, which blocks the catalytic sites (TACN headgroups) and inhibits the cata-
lytic conversion of the substrate to the chromogenic reporter molecule. The upper hemisphere shows the lower binding affinity of ADP, which
allows the catalytic conversion of the substrate to a chromogenic product.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
This would lead to more effective (and different) competition leaving groups in more detail (Fig. 2). 3-Trifluoromethyl-4-
with ATP and ADP for binding with the NP surface, thus nitrophenol (3b) possesses several interesting features: the
enabling their discrimination. In addition, an improvement in presence of a trifluoromethyl group in position 3 of the phenol
the reactivity of the substrate (kcat) and/or chromogenic pro- ring decreases the acidity of the leaving group with respect to
perties (molar extinction coefficient) would permit an improve- 3a (pKa from 7.15 to 6.07),54 while also ensuring a higher
ment in terms of the signal output by the catalytic assay and, hydrophobic contribution to the overall molecular structure.
thus, its sensitivity.
This substrate was thus expected to exhibit enhanced binding
affinity to the nanoparticle surface. On the other hand, despite
the fact that 4-methylumbelliferone (3c) possesses a higher
pKa (7.79),55 we were attracted by this leaving group because of
its fluorescence properties, which would in principle provide a
significant increase in sensitivity. In this context, it is worth
noting how the pKa of phenols determine their optical pro-
perties. In general, phenols show two characteristic absorption
bands in water. The band at shorter wavelengths originates
from the protonated form while the band at longer wave-
lengths is associated to the deprotonated form. At a given pH,
the equilibrium between the two forms is dependent on the
pKa value. In order to quantify how the acidity of the leaving
group influences the optical properties of chromophores 3a–c,
we recorded the UV-Vis spectra as a function of the pH (from
pH 6.0 to 8.0, Fig. 2).
Results and discussion
Substrate analysis
Two different approaches have been taken for the structural
modification of HPNPP (2), acting either on the internal
nucleophile or the chromogenic leaving group (Scheme 1b).
The first approach involves a change in the nucleophilicity of
the hydroxyl group by switching from a secondary to a primary
hydroxyl group (compound 4). This was expected to favourably
affect the cyclization rate because of reduced steric hindrance.
Secondly, changes in the structure of the leaving group (com-
pounds 5 and 6) can be used to affect features such as reactiv-
ity, substrate polarity (with an effect on KM) and signal
strength (molar absorption coefficient). After examination of a
series of potential candidates, we decided to examine two
It can be observed that an increase in the pH causes an
increase in the absorption band originating from the deproto-
Fig. 2 UV-Vis spectra of chromophores 3a–c as a function of pH (arrows show increase in pH). The behaviour of the corresponding molar absorp-
tion coefficients as a function of the pH is reported on the bottom. To account for the inaccuracy of measuring the optical path length in a micro-
titer well we decided to report ε·l. Experimental conditions: [3a] = [3b] = [3c] = 100 μM, [MES] = 10 mM, pH = 6.0 and 6.5, [HEPES] = 10 mM, pH =
7.0, 7.5 and 8.0, 40 °C.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
nated form (λmax = 400 nm for 3a, 392 nm for 3b and 360 nm
for 3c). The trends of the molar absorptivity as a function of
pH are in line with the corresponding pKa’s. In particular, 3b
displayed the best response up to pH 7.5, which illustrates the
important advantage of using more acidic reporter molecules
like 3b at physiological pH’s. Although the protonated forms
also have significant absorptivities at lower pH values (in par-
ticular 3c), from a practical point of view the absorption band
originating from the protonated form is often less useful for
signal generation, because of the frequent presence of other
chromophores that also absorb at lower wavelengths.
Fig. 3 Alkaline hydrolysis study. Experimental conditions: [substrate] =
50 μM, [CAPS] = 10 mM, pH = 11.0, 40 °C. See ESI† for more details.
Finally, all the precursors were deprotected using triethyl-
amine·3HF and passed through a SP-Sephadex column (Na+
form). All substrates 2 and 4–6 were obtained as sodium salts
with moderate yields after preparative reverse phase HPLC.
Substrate synthesis
Based on this initial analysis, three substrates for Au NP 1
have been synthesized according to Scheme 2. Substrate 4 is
an analogue of HPNPP (2) but contains a primary hydroxyl
group rather than a secondary one. Substrates 558 and 6 are
analogues containing leaving groups 3b and 3c, respectively.
The key intermediate in all these syntheses is an aryl phos-
phorodichloridate (7a–c),57 which was synthesized by slowly
adding pyridine to a solution of phosphorus oxychloride and
the corresponding phenol (3a–c) in dry toluene. Coupling of
phosphorodichloridates 7a–c with monoprotected diol 8 or 9
in dry THF with triethylamine as base furnished the corres-
ponding precursors 10 and 11a–c as triethylammonium salts
with moderate to good yield.56 Compound 8 was obtained by
the silyl mono-protection of commercially available (S)-(+)1,2-
propanediol (12), while compound 9 was synthesized by reduc-
ing commercially available ethyl (S)-(−)-2-(tert-butyldimethyl-
silyloxy)propionate (13) with BH3·THF.56
Intrinsic substrate reactivity
The intrinsic reactivity of the new substrates was first evaluated
by observing their base-induced hydrolysis kinetics at pH 11.
Substrates were incubated at 40 °C and the reaction was moni-
tored following the absorbance of the corresponding leaving
group (Fig. 3a). In all cases, the experimental profiles could be
fitted to a first order kinetic model (R2 > 95). The obtained
pseudo-first order constants (kobs) could be directly related to
the molecular structure of the substrates. This was clearly seen
from an analysis of the Brønsted plot (Fig. 3b), in which log
kobs is plotted against the pKa of the corresponding leaving
group (pKArOH). For substrates 2, 5 and 6 a clear linear corre-
lation was observed between the log kobs and pKArOH, support-
ing a mechanism in which the dissociation of the leaving
group is the rate-determining step. In addition, quantitative
analysis revealed that substrate 5 is 3 times more reactive than
2, which, in turn, is 2 times more reactive than substrate 6.
The importance of having a better internal nucleophile is
reflected by the behavior of substrate 4, which is hydrolyzed
7 times faster than 2.
We next investigated the effect of structural changes in the
substrate on the catalytic properties of Au NP 1. Kinetic
measurements were carried out at three different pHs (6.0, 7.0
and 8.0) by monitoring the rate of formation of leaving groups
3a–c as a function of the concentration of substrates 2 and
4–6. At each pH, a calibration curve was used to convert the
measured absorbance originating from 3a–c into the corres-
ponding initial rate (see ESI†). Plots of the obtained initial
reaction rates as a function of substrate concentration demon-
strate that all substrates are cleaved by Au NP 1 obeying
Michaelis-Menten-like enzyme kinetics (Fig. 4). Accordingly,
Scheme 2 General synthesis of HPNPP analogues.56,57 Reaction con- all curves were fitted to the Michaelis-Menten model yielding
ditions: (i) TBDMSCl (dropwise at 0 °C), imidazole, DCM, RT for 2 h
(78%); (ii) pyridine (dropwise at 95 °C), toluene, 110 °C for 15 min; (iii)
BH3·THF (dropwise at 0 °C), THF, 66 °C for 5 h (87%); (iv) 8 or 9 (drop-
wise at 0 °C), 7a–c, THF, RT for 2–5 h (40–90%); (v) 10 or 11a–c
TEA·3HF, THF, 0 °C to RT, overnight; Na+ ion exchange, prep. rev. phase
HPLC (25–30%).
the parameters kcat and KM as listed in Table 1.
The purpose of this study was the identification of
improved alternative substrates to HPNPP 2, both in terms of
reactivity (kcat) and binding affinity (KM), with the aim of
improving both the efficacy of signal generation in a catalytic
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 Initial rates for the cleavage of 2 (dark blue), 4 (red), 5 (light blue) and 6 (green) by Au NP 1 at three different pHs. Experimental conditions:
[TACN·Zn2+] = 20 1 μM, [substrate] = 0–2.0 mM, [HEPES] = 10 mM, pH = 7.0 and 8.0, [MES] = 10 mM, pH 6.0, 40 °C.
Table 1 Michaelis-Menten parameters obtained for substrate 2 and 4–6
pH 6.0
pH 7.0
pH 8.0
Substrate
kcat [×10−3 s−1
]
km [mM]
kcat [×10−3 s−1
]
km [mM]
kcat [×10−3 s−1
]
km [mM]
2(HPNPP)
0.2 0.03
0.5 0.05
0.8 0.1
0.4 0.05
0.4 0.05
0.1 0.03
0.3 0.05
1.4 0.1
3.0 0.2
3.8 0.2
0.1 0.05
0.3 0.04
0.1 0.02
0.2 0.02
0.04 0.01
4.9 0.2
9.2 0.2
7.6 0.2
0.3 0.04
0.5 0.03
0.5 0.03
0.3 0.02
0.07 0.01
4
5
6
0.04 0.01
signal amplification assay and the possibility to differentiate would have been observed. Altogether it indicates that the
between ATP and ADP. Upon examining the Michaelis-Menten interaction between small molecules and the organic mono-
curves in Fig. 4, it can be seen that the observed kcat for 4, the layer is governed by a combination of noncovalent interactions,
substrate with the primary internal nucleophile, was two times which is in line with previous observations made by us and
higher than for HPNPP over the entire pH range tested. While others.47,48,51,53 In light of these observations, together with its
this was exciting in terms of reactivity and the potential improved absorptivity (see Fig. 1), we decided that substrate 5
responsiveness of the system, the exhibited KM values were was the best candidate for pursuing improvements in the cata-
very close to those of HPNPP, which suggested that this sub- lytic signal amplification assay both in terms of sensitivity and
strate was unlikely to exhibit enhanced affinity for the NPs to applicability.
compete with the binding of ADP.
ATP/ADP-inhibition and signal amplification studies
The fluorogenic substrate 6, on the other hand, showed KM
values much lower than that observed for HPNPP (at pH’s
7 and 8). This shows that substrate 6 should exhibit signifi-
cantly higher affinity for Au NP 1 than HPNPP, which would
facilitate the discrimination between ATP and ADP. Unfortu-
nately, substrate 6 was also dramatically less reactive than
HPNPP (up to 15 times at pH 8.0), which rendered the sub-
strate ineffective for a catalytic amplification assay.
Finally, we were pleased to observe that CF3-substituted
substrate 5 showed very promising properties, both in terms of
kcat and KM. Substrate 5 was observed to be up to 3 times more
reactive than HPNPP over the entire range of pH studied and
showed unprecedented high affinity at pH 6.0 (KM = 0.11 mM).
This lower KM is due likely to the increased hydrophobic con-
tribution from the CF3 group, which should increase binding
affinity to the monolayer of the Au NPs. However, a general
analysis of the KM-values measured for all compounds indicate
that hydrophobic interactions alone do not account for the
stability of the interaction with Au NP 1. If that would have
been the case, a lower dependence of the KM-values on the pH
The key hypothesis of this study was that a substrate (like com-
pound 5) with a higher affinity for Au NP 1 than HPNPP (2)
would be able to compete more efficiently with ATP and ADP
for binding to the monolayer of the Au NPs. This idea was
tested by performing a series of inhibition experiments in
which the cleavage of 5 by Au NP 1 was inhibited with increas-
ing amounts of ATP or ADP (Fig. 5). Substrate 5 was added to
give an initial concentration of 1 mM and the release of
chromophore 3b was measured by monitoring the increase in
absorbance at 392 nm. The obtained initial reaction rates
(νinit,1) were normalized over the initial rate in the absence of
inhibitor (νinit,0) and were plotted as a function of the concen-
tration of inhibitor added. As for the Michaelis-Menten kine-
tics, the inhibition experiments were repeated at three
different pHs (6.0, 7.0 and 8.0, Fig. 5a–c). We were very pleased
to observe that in all cases, ATP and ADP inhibited the cataly-
sis of HPNPP to different extents, with a maximum difference
at a concentration of inhibitors equal to 6.4 μM (see the yellow
rectangles). The ratio of the relative initial rates (νADP/νATP) (i.e.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 5 Inhibition curves of ATP (blue) and ADP (red) at pH 6.0 (a), pH 7.0 (b) and pH 8.0 (c). The time course of ΔA corresponding to the concen-
tration at which the difference between ATP (blue) and ADP (red) is maximized (6.4 μM) is reported in the inset. (d) Representative inhibition experi-
ment at pH 6.0. The zoom indicates the wells containing 6.4 μM of ADP (top) or ATP (bottom) after 24 h at pH 6.0. (e) Signal amplification at the
three different pHs in the presence of 6.4 μM of inhibitors (see ESI† for the full kinetics). Experimental conditions: [TACN·Zn2+] = 20 μM, [5] = 1 mM,
[HEPES] = 10 mM, pH = 8.0 or 7.0, [MES] = 10 mM, pH 6.0, [ATP or ADP] = 0–16 μM, 40 °C.
the differentiation power) was observed to be greatest at pH a practical setting, the enzyme alkaline phosphatase was
6.0 (νADP/νATP = 7.2), where the KM of 5 has its minimum chosen. An assay was set up whereby all components of the
(0.11 mM), while the poorest differentiation was observed at system were first pipetted into wells in a 96-well microtitre
pH 8.0 (νADP/νATP = 2.3). However, it is important to remember plate and allowed to stabilise for 30 min to demonstrate the
that the efficacy of the assay for discriminating between ATP lack of signal in the absence of added enzyme (Fig. 6). After
and ADP is determined by the absolute change in output 30 min, the enzyme alkaline phosphatase was added in situ
signal, which depends not only on the discriminatory value,
but also on the reaction rate and the molar extinction coeffi-
cient at a given pH. Interestingly, the analysis of the difference
in output signal for ATP and ADP (expressed by ΔAADP
−
ΔAATP) demonstrates that in this case the combination of a
faster reaction rate and a stronger output signal at higher pH
more than compensates for the lower discriminatory power.
For example, the same level of amplification obtained after
3.0 h at pH 6.0 (ΔΔA ≈ 0.14) is reached almost 10 times faster
(after ca. 20 min) at pH 8.0 (Fig. 5e). This response curve also
demonstrates a key attractive feature of catalytic signal amplifi-
cation, which is the increase in signal intensity as a function
of time. The ability of the assay to convert the difference in
ATP and ADP at 6.4 µM into a difference in absorbance of over
0.14 units in just 20 minutes demonstrates the sensitivity of
the assay. Fig. 5d also illustrates the ease by which the pres-
ence of ADP or ATP can be easily visually distinguished, as well
as the potential use of this methodology for high-throughput
screening.
Fig. 6 The use of the above described gold nanoparticle system for the
in situ detection of alkaline phosphatase activity. The increase in signal
over time demonstrates the utility and advantage of using catalytic
signal amplification for signal generation. Experimental conditions:
[TACN·Zn2+] = 20 μM, [5] = 1 mM, [HEPES] = 10 mM, pH = 7.0, [ATP] =
To demonstrate the potential of the described gold nano-
particle system to detect the enzymatic consumption of ATP in 6.4 μM, 40 °C, [alkaline phosphatase] = 3.3 U mL−1
.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
into the microtitre plate (Fig. 6, see sample – alkaline phos- Instrumentation
phatase). It can be observed that there is significant increase
1H, 13C and 31P-NMR spectra were recorded using a Bruker
AV300 spectrometer operating at 75, 121 and 300 MHz for 13C,
31P and 1H, respectively. Chemical shifts (δ) are reported in
ppm using CDCl3 or D2O residual solvent values as internal
reference. 31P-NMR spectra were calibrated using a solution of
H3PO4 80% as external standard. The coupling constants (J)
are listed in Hertz (Hz), while the multiplicity of the signals
are shown as follow: s: singlet, d: doublet, t: triplet, q: quartet,
sp: septet, m: multiplet.
ESI-MS measurements were performed on an Agilent
Technologies 1100 Series LC/MSD Trap-SL spectrometer
equipped with an ESI source, hexapole filter and ionic trap. All
the sample solutions were prepared in neutral solvents (ACN
or MeOH). In all cases the mass of the anionic species were
characterized operating in the negative mode.
in the absorbance intensity at 392 nm only for the sample of
interest. Controls were performed both in the absence of alka-
line phosphatase, and in the presence of alkaline phosphatase
but without of Au NPs, to show that both the Au NP catalytic
system and the enzyme alkaline phosphatase need to be
present for a signal to be observed. The advantage of using
catalytic signal amplification for the generation of a signal in
an assay is clearly seen in Fig. 6. A signal difference of 0.04
absorbance units after 15 min is translated to a difference of
0.21 absorbance units at 60 min. This generation of signal by
catalytic amplification allows the signal generated to be readily
observed by the naked eye, reinforcing the practicality of this
methodology. Currently we are exploring the possibility to use
this assay for the detection of PK-activity.
HPLC purifications were performed on a preparative HPLC
Shimadzu LC-8A equipped with a Shimadzu SPD-20A UV
detector. The column used for separation was a Jupiter Proteo
4u 90A 250 × 21.20 mm, 4 μm. All the runs were carried out
using a flow of 17 mL min−1. Eluents: neutral H2O (A) and
ACN (B).
The pH of buffer solutions were determined at room temp-
erature using a Metrohm-632 pH meter equipped with a
Ag/AgCl/KCl reference electrode. UV-Vis measurements were
recorded on a TECAN M1000 PRO micro-plate reader using
UV-Star® Greiner Bio-one microtiter plates (cycloolefin 96-
WELL).
Conclusions
In conclusion, we have shown for the first time a system able
to discriminate ATP from ADP by means of catalytic signal
amplification. The approach relies on the use of gold nano-
particles coated which TACN·Zn2+ head groups which are able
to catalyze the transphosphorylation of the compound
2-hydroxypropyl-(3-trifluoromethyl-4-nitro)phenyl phosphate
5, with release of a chromogenic reporter. Phosphate 5 was
found to possess faster kinetics than the more commonly
employed HPNPP, and its increased affinity for the nano-
particle surface allowed for the discrimination of ATP and
ADP. This methodology is conducive to high-throughput
screening with results that can be easily detected, even with
Synthesis of substrates
Synthesis of 2-[(tert-butyldimethylsilyl)oxy]ethyl-1-yl 4-nitro-
the naked eye. Overall, this represents a viable new approach phenyl phosphate, TEA salt (10). In a three-necked flask, a
for monitoring the consumption of ATP with very high solution of alcohol 859 (1.6 g, 6.3 mmol) and TEA (0.88 mL,
sensitivity.
6.3 mmol) in anhydrous THF (20 mL) was added dropwise
over a period of 30 min to a stirred solution of commercially
available 4-nitrophenyldi-chlorophosphate in anhydrous THF
(40 mL) cooled to 0 °C under nitrogen. After the complete
addition, the mixture was stirred at R.T. for 2 hours. Finally, a
mixture of TEA/cold water (3/12 mL) was added to hydrolyze
Experimental
Reagents and materials
Au NP
1 was prepared and characterised as previous the intermediate monochloridate and the solution was stirred
described,46 Zn(NO3)2 was an analytical grade regent and the for a further 30 minutes. THF was evaporated and the product
concentration was determined by atomic absorption spec- extracted into CH2Cl2 and washed with H2O and brine, dried
troscopy. ATP, ADP, 2-(N-morpholino)ethanesulfonic acid (MES), over MgSO4 and the organic solvent removed under reduced
4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), pressure. The product was purified by column chromatography
N-cyclohexyl-3-aminopropanesulfonic acid (CAPS) and alkaline (SiO2, MeOH from 0 to 5%/CHCl3 with 0.5% TEA) to give 10 as
phosphatase from bovine intestinal mucosa (P5521 SIGMA) a pale yellow oil (1.34 g, 43%).
3
were purchased from Sigma Aldrich and used without further
1H-NMR (CDCl3) δ (ppm): 8.12–8.09 (d, 2H, J = 9.1 Hz, Hg),
purification. In all cases, stock solutions were prepared using 7.39–7.36 (d, 2H, 3J = 9.1 Hz, Hf), 4.37 (m, 1H, –CH–CH2–),
3
deionized water filtered with a Milli-Q water-purifier (Millipore) 3.50 (m, 2H, –CH2–CH–), 3.00 (q, 6H, J = 7.3 Hz, –CH2–CH3),
3
3
and stored at 4 °C. For the syntheses of substrates, all commer- 1.27 (t, 9H, J = 7.3 Hz, –CH2–CH3), 1.22–1.20 (d, 3H, J = 6.2
cially available reagents were purchased from Sigma-Aldich and Hz, –CH–CH3), 0.81 (s, 9H, –C(CH3)3), −0.05 (s, 6H, –Si(CH3)2).
used as received. Anhydrous solvents were purchased and used
13C-NMR (CDCl3) δ (ppm): 159.6 (d, 6.2 Hz), 143.1, 125.7,
without further purification. Chromatographic columns were 120.5 (d, J = 5.7 Hz), 74.0 (d, J = 4.8 Hz), 68.0 (d, J = 6.7 Hz),
packed with silica gel Macherey-Nagel 230–400 mesh, while TLC 46.2, 26.3, 18.8 (d, J = 18.8 Hz), 9.2, −4.9.
analyses were performed with silica gel 60-F254 plates.
31P-NMR (CDCl3) δ (ppm): −8.03 (d, J(P,H) = 8.3 Hz).
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
MS-ESI (FIA,(−)-mode, ACN), m/z: Calcd for C15H25NO7PSi−: ated and the product extracted into CH2Cl2 and washed with
390.4; found 390.0. H2O and brine, dried over MgSO4 and the organic solvent
Synthesis of 2-hydroxyethyl-1-yl 4-nitrophenyl phosphate, removed under reduced pressure. The product was purified by
sodium salt (4). To a solution of 10 (1.03 g, 2.1 mmol) in anhy- gravimetric column chromatography (SiO2, MeOH from 0 to
drous THF (10 mL) was added a solution of TEA·3HF complex 5%/CHCl3 with 0.5% TEA) to afford the TEA salt of 11c (2.18 g,
(1.40 mL, 8.4 mmol) in THF (2 mL) at R.T. and the resulting 90%).
mixture was stirred for 5 h before the addition of H2O (15 mL)
1H-NMR (CDCl3), δ (ppm): 7.47–7.44 (d, 1H, 3J = 9.3 Hz, Hf),
3
and TEA until pH ∼ 5. THF was evaporated and the aqueous 7.26 (d, 2H, J = 0.5 Hz, Hd, He), 6.13 (s, 1H, Hh), 3.92–3.67 (m,
solution washed with Et2O to remove tert-butyldimethylsilyl 2H + 1H, –CH–CH2–), 3.02–2.94 (q, 6H, 3J = 7.2 Hz, –CH2–CH3),
fluoride. After evaporation of Et2O, the crude product was 2.38 (s), 1.30–1.25 (t, 9H, 3J = 7.2 Hz, –CH2–CH3), 1.11–1.09
extracted several times into CH2Cl2, washed with H2O to (d, 3H, 3J = 5, 9 Hz, –CH–CH3), 0.82 (s, 9H, –C–(CH3)3), 0,00
remove undesired triethylammonium salts and passed (s, 6H, –Si–(CH3)2).
through an ion exchange resin (Sephadex C-50, Na+). The water
13C-NMR (CDCl3), δ (ppm): 161.2 (d, J = 5.6 Hz), 156.6,
was removed by rotary evaporation. EtOH was added and 154.4, 152.5, 125.0, 116.7, 114.9, 112.5 (d, J = 9.4 Hz), 107.8,
undissolved inorganic salts were removed by filtration. EtOH 71.0, 67.6, 45.6, 25.7, 20.6, 18.3, 8.7, −4.9.
was removed under reduced pressure furnishing the desired
31P-NMR (CDCl3), δ (ppm): −4.39 (t, 3J(P,H) = 6.1 Hz).
product 4 as a white solid (157 mg, 25%).
MS-ESI (FIA,(−)-mode, ACN), m/z: calcd for C19H28O7PSi−:
3
1H-NMR (D2O), δ (ppm): 8.33–8.30 (d, 2H, J = 9.2 Hz, He), 427.5; found 427.0.
7.43–7.40 (d, 2H, 3J = 9.1 Hz, Hd), 4.49 (m, 1H, –CH–CH2–),
Synthesis of 2-hydroxypropyl 4-methylumbelliferyl phos-
3.66 (m, 2H, –CH–CH2), 1.32–1.30 (d, 3H, 3J = 6.4 Hz, –CH– phate, sodium salt (6). To
a solution of 11c (1.22 g,
CH3).
2.85 mmol) in anhydrous THF (10 mL) was added a solution
13C-NMR (D2O) δ (ppm): 158.0 (d, 6.4 Hz), 143.8, 126.3, of TEA·3HF complex (1.86 mL, 11.4 mmol) in anhydrous THF
120.9, 75.3–75.2 (m), 65.8–65.7 (d, J = 8.2 Hz), 17.5.
31P-NMR (D2O), δ (ppm): −4.64 (d, J(P,H) = 7.7 Hz).
(2 mL) at R.T. The resulting pale yellow mixture was stirred
until the disappearance of 11c (followed by ESI–MS†), before
MS-ESI (FIA,(−)-mode, ACN), m/z: Calcd for C9H11NO7P−: the addition of H2O (15 mL) and TEA until pH ∼ 6.5. THF and
276.2, found: 275.8.
the resulting volatile tert-butyldimethylsilyl fluoride were
The synthesis of 2-hydroxypropyl 3-trifluoromethyl-4-nitro- removed by rotary evaporation. The crude product was then
phenyl phosphate, sodium salt (5) was performed accorded to extracted several times into CH2Cl2 (3 × 25 mL), washed with
our previously described procedure.58
H2O to remove undesired triethylammonium salts and passed
Synthesis of 4-methylumbelliferyl phosphodichloridate (7c). through an ion exchange resin (Sephadex C-50, Na+). The water
In a dry 3-necked flask equipped with a condenser, commer- was removed by rotary evaporation. EtOH was added and
cially available 4-methylumbelliferone (1.5 g, 8.5 mmol) was undissolved inorganic salts were removed by filtration. EtOH
suspended in anhydrous toluene (50 mL) and freshly distilled was removed under reduced pressure furnishing 250 mg of the
POCl3 (1.82 mL, 19.6 mmol) was added. The reaction mixture desired product 6 as a white solid (26%).
3
was heated to 95 °C and pyridine (680 μL) was added by
1H-NMR (D2O), δ (ppm): 7.70–7.67 (d, 1H, J = 8.6 Hz, Hf),
syringe pump over a period of 30 min. The reaction was heated 7.23–7.15 (dd, 2H, 3J = 13.4 Hz, 4J = 2.22 Hz, Hd, He), 6.22
at reflux (110 °C) for a further 15 min after the addition was (s, 1H, Hh), 4.10–3.86 (m, 2H + 1H, –CH–CH2–), 2.41 (s, 3H,
completed and the salts were removed by filtration of the –CH3), 1.23–1.21 (d, 3H, –CH–CH3).
cooled reaction mixture. The obtained filtrated solution
13C-NMR (D2O), δ (ppm): 164.3, 156.0, 154.9 (d, J = 6.5 Hz),
(green) was concentrated under reduced pressure and the 153.4, 126.5, 117.3 (d, J = 4.8 Hz), 116.2, 111.8, 107.9 (d, J = 5.2
resulting purple oil was used in the next steps without further Hz), 70.0 (d, J = 6.3 Hz), 66.6 (d, J = 8.2 Hz), 17.8 (d, J =
purification.
14.9 Hz).
1H-NMR (CDCl3), δ (ppm): 7.80 (dd, 1H, 3J = 8.7, 2.4 Hz,
Hc), 7.33–7.25 (m, 2H, Hd, He), 6.32 (s, 1H, Ha), 2.51–2.50 (d, J =
23 Hz, 3H, Hb).
31P-NMR (D2O), δ (ppm): −3.74 (t, J = 9.5 Hz).
MS-ESI (FIA,(−)-mode, ACN), m/z: Calcd for C13H14O7P−:
313.2, found: 312.9.
31P-NMR (CDCl3), δ (ppm): 4.27 (s).
Synthesis of 2-[(tert-butyldimethylsilyl)oxy]propyl 4-methyl-
umbelliferyl phosphate, TEA salt (11c). Phosphodichloridate
7c (5.7 mmol) was dissolved in anhydrous THF (30 mL), and
UV-vis measurements
Calibration curves. The calibration curves were carried out at
cooled to 0 °C under nitrogen. A solution of alcohol 960 40 °C by measuring the absorbance at 400 or 392 nm of
(1.08 g, 5.7 mmol) and TEA (0.79 mL, 5.7 mmol) in anhydrous increasing amounts of 3a, 3b or 3c (final concentration: from
THF (20 mL) was added dropwise over a period of 30 min. 0 to 100 μM) added to a aqueous solution (buffered at the
After addition was complete, the mixture was allowed to stir at desired pH, [buffer] = 10 mM) containing Au NP 1·Zn2+
R.T. for 5 h. Finally, a mixture of TEA/cold water (2/14 mL) was (20 μM) (Vwell = 250 μL). Taking into account the Lambert-Beer
added to hydrolyze the intermediate monochloridate and the law, the slope of the curve represents ε·l. The obtained values
solution was stirred for a further 30 minutes. THF was evapor- are listed in the table below and were used to convert initial
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
rates (vinit) from dA/dt, to dC/dt. See ESI† for full UV-vis spectra volume changes during the analysis, a transparent plastic
and interpolations.
strip on the top of the wells was applied immediately after
the additions of the substrate. Initial rates were scaled
to the initial rate observed in the absence of inhibitor (i.e.
νinit,1/νinit,0). The kinetics at which the differentiation is maxi-
mized (6.4 μM of inhibitors) were obtained using a well-
volume of 50 μL. A shorter optical path length was needed to
avoid saturation of the detector.
List of all the ε·l values used (M−1
)
Leaving group
λ (nm)
pH 6.0
pH 7.0
pH 8.0
pH 11.0
3a
3b
3c
400
392
360
1378
3277
516
5820
7394
2194
10 087
8471
6516
12 966
9745
13 102
Catalytic signal amplification with alkaline phosphatase.
The catalytic activity of Au NP 1·Zn2+ in the presence and
absence of alkaline phosphatase was measured by monitoring
the release rate of chromophore 3b in buffer at pH 7 (HEPES =
10 mM) containing Au NP 1·Zn2+ (20 μM) and 6.4 μM of ATP.
The evolution of the absorbance at 392 nm was monitored at
40 °C after adding substrate 5 (final concentration = 1 mM). In
Fig. 6 – ‘Control – no alkaline phosphatase,’ the above
described solution was observed over time to determine the
background signal. For ‘Control – No Au NPs,’ the above
described solution, but without Au NP 1 was incubated for
30 min before the addition alkaline phosphatase at t = 0. For
‘Sample – Alkaline Phosphatase,’ the above described solution
was incubated for 30 min before the addition of alkaline phos-
phatase at t = 0.
Hydrolysis of the substrates. The hydrolysis was carried out
by monitoring the absorbance of the corresponding leaving
group in an aqueous solution ([CAPS] = 10 mM, pH = 11.0)
containing the substrate (2, 4, 5, or 6) at a final concentration
of 50 μM (Vwell = 250 μL, 40 °C). In order to avoid volume
changes during the analysis, a transparent plastic strip on the
top of the wells was applied before the measurement. The
obtained profiles were nicely fitted with a first order kinetic:
½PNPꢀt ¼ 1 ꢁ eꢁkt
(Origin 8 Pro, function: Mn Molecular 1 A1 − A2e−kt, con-
straints; A1 = A2 = 1). Stock solutions of substrates were pre-
pared by weight from the final product of synthesis. This
experiment was used to determine accurately their concen-
tration using the plateau value (full conversion) and the corres-
ponding ε·l listed in the above table.
Acknowledgements
Michaelis-Menten kinetics. The catalytic activity of Au NP
1·Zn2+ was measured by monitoring the release rate of the
leaving group in different aqueous solutions (buffered at the
desired pH, [buffer] = 10 mM) containing Au NP 1·Zn2+
(20 μM) and increasing amounts of substrate. The evolution of
the absorbance at 400, 392 or 360 nm was monitored at 40 °C
after adding 2, 4, 5, or 6, final concentration: from 0 to
2.0 mM) (Vwell = 250 μL). In order to avoid volume changes
during the analysis, a transparent plastic strip on the top of
the wells was applied immediately after the additions of the
substrates. Initial rates (ν0) were calculated converting the
absorbance signals into concentration using the corres-
ponding ε·l listed in the above table. Michaelis-Menten para-
meters were extrapolated by fitting the corresponding
saturation profile with the Michaelis Menten model:
Financial support from the ERC (StG-239898), University of
Padova (CPDA138148) and COST (CM1304) is acknowledged.
Notes and references
1 S. Naylor, Expert Rev. Mol. Diagn., 2003, 3, 525–529.
2 F. P. Perera and I. B. Weinstein, Carcinogenesis, 2000, 21,
517–524.
3 D. Diamond, Principles of Chemical and Biological Sensors,
Wiley, New York, 1998.
4 P. Scrimin and L. J. Prins, Chem. Soc. Rev., 2011, 40, 4488–
4505.
5 T. Pawson and J. D. Scott, Trends Biochem. Sci., 2005, 30,
286–290.
6 S. A. Johnson and T. Hunter, Nat. Methods, 2005, 2, 17–25.
7 M. B. Yaffe, Nat. Rev. Mol. Cell Biol., 2002, 3, 177–186.
8 L. N. Johnson and R. J. Lewis, Chem. Rev., 2001, 101, 2209–
2242.
9 C. J. Hastie, H. J. McLauchlan and P. Cohen, Nat. Protoc.,
2006, 1, 968–971.
Vmaxꢂ½Substrateꢀ
v ¼
Km þ ½Substrateꢀ
(Origin 8 Pro, function: Hill, constraints; n = 1, i.e. Michae-
lis-Menten kinetic). kcat values were obtained by dividing Vmax
to the concentration of TACN·Zn2+ (20 μM).
Inhibition experiments with ATP and ADP. The catalytic 10 H. C. Ma, S. Deacon and K. Horiuchi, Expert Opin. Drug Dis-
activity of Au NP 1·Zn2+ was measured by monitoring the
covery, 2008, 3, 607–621.
release rate of chromophore 3b in different aqueous solutions 11 N. P. Oien, L. T. Nguyen, F. E. Jernigan, M. A. Priestman
(buffered at the desired pH, [buffer] = 10 mM) containing Au
and D. S. Lawrence, Angew. Chem., Int. Ed., 2014, 53, 3975–
3978.
NP 1·Zn2+ (20 μM) and increasing amounts of ATP or ADP
(final concentration = from 0 to 16 μM). The evolution of the 12 Y. J. Li, W. H. Xie and G. J. Fang, Anal. Bioanal. Chem.,
absorbance at 392 nm was monitored at 40 °C after adding 2008, 390, 2049–2057.
substrate 5 (final concentration = 1 mM). In order to avoid 13 B. E. Turk, Nat. Methods, 2005, 2, 251–252.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
14 K. Kupcho, R. Somberg, B. Bulleit and S. A. Goueli, Anal. 38 W. H. Chapman and R. Breslow, J. Am. Chem. Soc., 1995,
Biochem., 2003, 317, 210–217. 117, 5462–5469.
15 E. A. Gaudet, K. S. Huang, Y. Zhang, W. Huang, D. Mark 39 G. Zaupa, C. Mora, R. Bonomi, L. J. Prins and P. Scrimin,
and J. R. Sportsman, J. Biomol. Screening, 2003, 8, 164–175. Chem. – Eur. J., 2011, 17, 4879–4889.
16 F. S. Wouters and P. I. H. Bastiaens, Curr. Biol., 1999, 9, 40 G. Zaupa, P. Scrimin and L. J. Prins, J. Am. Chem. Soc.,
1127–1130. 2008, 130, 5699–5709.
17 T. Noguchi, B. Roy, D. Yoshihara, Y. Tsuchiya, T. Yamamoto 41 C. Pezzato, P. Scrimin and L. J. Prins, Angew. Chem., Int.
and S. Shinkai, Chem. Sci., 2015, 6, 3863–3867. Ed., 2014, 53, 2104–2109.
18 C. Pezzato, D. Zaramella, M. Martinelli, G. Pieters, 42 C. Pezzato, B. Lee, K. Severin and L. J. Prins, Chem.
M. A. Pagano and L. J. Prins, Org. Biomol. Chem., 2015, 13,
1198–1203.
19 X. Liu, Y. Li, X. H. Xu, P. Li, Z. Nie, Y. Huang and S. Z. Yao,
Trends Anal. Chem., 2014, 58, 40–53.
20 J. Bai, C. H. Liu, T. Yang, F. F. Wang and Z. P. Li, Chem.
Commun., 2013, 49, 3887–3889.
21 S. Martic and H. B. Kraatz, Chem. Sci., 2013, 4, 42–59.
Commun., 2013, 49, 469–471.
43 A. Ojida, K. Honda, D. Shinmi, S. Kiyonaka, Y. Mori and
I. Hamachi, J. Am. Chem. Soc., 2006, 128, 10452–10459.
44 A. Ojida, M. Inoue, Y. Mito-oka, H. Tsutsumi, K. Sada and
I. Hamachi, J. Am. Chem. Soc., 2006, 128, 2052–2058.
45 A. Ojida, Y. Mito-oka, K. Sada and I. Hamachi, J. Am. Chem.
Soc., 2004, 126, 2454–2463.
22 G. Ghale, V. Ramalingam, A. R. Urbach and W. M. Nau, 46 G. Pieters, A. Cazzolaro, R. Bonomi and L. J. Prins, Chem.
J. Am. Chem. Soc., 2011, 133, 7528–7535. Commun., 2012, 48, 1916–1918.
23 A. E. Hargrove, S. Nieto, T. Z. Zhang, J. L. Sessler and 47 B. Perrone, S. Springhetti, F. Ramadori, F. Rastrelli and
E. V. Anslyn, Chem. Rev., 2011, 111, 6603–6782. F. Mancin, J. Am. Chem. Soc., 2013, 135, 11768–11771.
24 T. Z. Zhang, N. Y. Edwards, M. Bonizzoni and E. V. Anslyn, 48 G. Pieters, C. Pezzato and L. J. Prins, Langmuir, 2013, 29,
J. Am. Chem. Soc., 2009, 131, 11976–11984. 7180–7185.
25 L. Zhu and E. V. Anslyn, Angew. Chem., Int. Ed., 2006, 45, 49 D. Zaramella, P. Scrimin and L. J. Prins, J. Am. Chem. Soc.,
1190–1196. 2012, 134, 8396–8399.
26 W. Bai, N. A. Gariano and D. A. Spivak, J. Am. Chem. Soc., 50 X. Liu, Y. Hu and F. Stellacci, Small, 2011, 7, 1961–1966.
2013, 135, 6977–6984.
27 E. Sella and D. Shabat, J. Am. Chem. Soc., 2009, 131, 9934–
51 R. L. Phillips, O. R. Miranda, D. E. Mortenson,
C. Subramani, V. M. Rotello and U. H. F. Bunz, Soft Matter,
2009, 5, 607–612.
9936.
28 N. C. Gianneschi, S. T. Nguyen and C. A. Mirkin, J. Am. 52 G. V. Oshovsky, D. N. Reinhoudt and W. Verboom, Angew.
Chem. Soc., 2005, 127, 1644–1645. Chem., Int. Ed., 2007, 46, 2366–2393.
29 G. Das, P. Talukdar and S. Matile, Science, 2002, 298, 1600– 53 A. V. Ellis, K. Vjayamohanan, R. Goswaimi, N. Chakrapani,
1602.
L. S. Ramanathan, P. M. Ajayan and G. Ramanath, Nano
Lett., 2003, 3, 279–282.
54 B. G. Tehan, E. J. Lloyd, M. G. Wong, W. R. Pitt, E. Gancia
and D. T. Manallack, Quant. Struct.-Act. Rel., 2002, 21, 473–
485.
30 R. Bonomi, A. Cazzolaro, A. Sansone, P. Scrimin and
L. J. Prins, Angew. Chem., Int. Ed., 2011, 50, 2307–2312.
31 F. Manea, F. B. Houillon, L. Pasquato and P. Scrimin,
Angew. Chem., Int. Ed., 2004, 43, 6165–6169.
32 E. Y. Tirel, Z. Bellamy, H. Adams, V. Lebrun, F. Duarte and 55 R. Strachan, J. Wood and R. Hirschmann, J. Org. Chem.,
N. H. Williams, Angew. Chem., Int. Ed., 2014, 53, 8246–8250. 1962, 27, 1074–1075.
33 F. Mancin, P. Scrimin and P. Tecilla, Chem. Commun., 2012, 56 T. Humphry, S. Iyer, O. Iranzo, J. R. Morrow, J. P. Richard,
48, 5545–5559.
P. Paneth and A. C. Hengge, J. Am. Chem. Soc., 2008, 130,
17858–17866.
34 R. Bonomi, F. Selvestrel, V. Lombardo, C. Sissi, S. Polizzi,
F. Mancin, U. Tonellato and P. Scrimin, J. Am. Chem. Soc., 57 J. P. Horwitz and J. V. Freisler, J. Med. Chem., 1970, 13,
2008, 130, 15744–15745. 1024–1025.
35 A. A. Neverov, Z. L. Lu, C. I. Maxwell, M. F. Mohamed, 58 J. L.-Y. Chen, C. Pezzato, P. Scrimin and L. J. Prins, Chem. –
C. J. White, J. S. W. Tsang and R. S. Brown, J. Am. Chem.
Soc., 2006, 128, 16398–16405.
36 N. H. Williams, B. Takasaki, M. Wall and J. Chin, Acc.
Chem. Res., 1999, 32, 485–493.
Eur. J., 2016, 22, 7028–7032.
59 R. S. Perali, S. Mandava and V. R. Chunduri, Tetrahedron
Lett., 2011, 52, 3045–3047.
60 R. A. Pilli, M. M. Victor and A. de Meijere, J. Org. Chem.,
2000, 65, 5910–5916.
37 J. Chin, Curr. Opin. Chem. Biol., 1997, 1, 514–521.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016