Surface-functionalized fluorescent silica nanoparticles for the detection of ATP

Article abstract of DOI:10.1039/c1cc10419e

The design of two-dyed fluorescent silica nanoparticles for ATP detection is presented. The indicator dye possesses a dipicolyl-amine (DPA) unit complexed with Zn(ii) as a receptor function for ATP while a rhodamine derivative is used as the reference dye. The nanoparticles were fully characterized regarding analytical performance, morphology and cytocompatibility.

Full text of DOI:10.1039/c1cc10419e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6066–6068
www.rsc.org/chemcomm
COMMUNICATION
Surface-functionalized fluorescent silica nanoparticles for the detection of
ATPw
Artur J. Moro, Jennifer Schmidt,* Tristan Doussineau, Alejandro Lapresta-Fernande
´
z,d
a
b
c
e
b
Joachim Wegener and Gerhard J. Mohr
Received 21st January 2011, Accepted 8th April 2011
DOI: 10.1039/c1cc10419e
4
The design of two-dyed fluorescent silica nanoparticles for ATP
detection is presented. The indicator dye possesses a dipicolyl-
amine (DPA) unit complexed with Zn(II) as a receptor function
for ATP while a rhodamine derivative is used as the reference
dye. The nanoparticles were fully characterized regarding
analytical performance, morphology and cytocompatibility.
as the receptor. Apart from its sensitivity to ATP, this
indicator provides very favorable optical properties, such as
high quantum yields and a large Stokes’ shift.
The immobilization of fluorescent indicator dyes onto
nanoparticles is of significant importance since plain organic
dyes are often toxic to cells. Moreover, many indicator dyes
only show changes in fluorescence intensity making them
unreliable for quantitative measurements of analyte concen-
tration, since effects such as dye aggregation or fluctuations in
the light source intensity may influence their fluorescence. The
latter problem can be overcome by immobilizing a second but
inert fluorescent dye (reference) in the nanoparticles. By relating
fluorescence changes of the indicator dye to changes of the
reference dye, ratiometric measurements can be achieved. The
success of this methodology was already demonstrated for
The development of new analytical systems for small bio-
molecules of significant medical interest has increased over the
past decades. Within this scope, one of the mostly studied
0
biomolecules is adenosine-5 -triphosphate (ATP). Based on its
vital role for energy production and storage in living cells, it can
1
be used as an indicator for cell viability or cell damage. Thus,
the detection of ATP levels can be extremely useful for metabolic
studies in cell biology as well as clinical diagnostics. In this sense,
fluorescent nanoparticles are regarded as one of the most
promising approaches for analytical applications in biology
and medicine due to several of their unique characteristics,
namely their size, chemical stability and versatility as well as
5
several analytes, including pH and chloride.
The design of the ATP-sensitive nanoparticles started with
the synthesis of the silica core nanoparticles according to a
¨
modified Stober method, using tetraethoxyorthosilicate
2
relatively low cytotoxicity when using a biocompatible matrix.
(TEOS) as the precursor. The reference dye consisted of a
silylated-sulforhodamine (see ESIw for details) which was also
added to the reaction mixture. Consequently, the silyl ether
group provided covalent attachment to the silica matrix of the
nanoparticles. An undoped silica shell was deposited around
the nanoparticle cores. The purpose here was to prevent any
One of the most promising approaches for ATP fluorescence
3
indicator dyes was introduced by Hamachi et al., who used
Zn(II) complexes for the fluorescent detection of phosphorylated
peptides and other species, in an attempt to mimic the binding
mechanism of phosphatases to their substrates. With this in
mind, our group developed a fluorescent indicator for ATP, 4ꢀ
Zn, based on naphthalimide as the fluorophore and a Zn(II)
metal cation complexed with a dipicolylamine (DPA) moiety
¨
effects of Forster resonance energy transfer (FRET) between
the reference and the indicator dye, as well as obtaining an
6
homogeneous reactive surface for further functionalization.
The outer surface of the nanoparticles was then functionalized
with amine groups using 3-aminopropyltriethoxysilane (APTEOS)
which finally allowed for the covalent attachment of 4ꢀZn to
the surface of the nanoparticles via N-hydroxy-succinimidyl
ester activation. A full scheme of the synthesis is found in
Fig. 1. The size and polydispersity of the nanoparticles were
determined by dynamic light scattering (DLS). In order to
assess the effect on the hydrodynamic diameter of the nano-
particles, the size of the nanoparticles was measured before
and after functionalization with 4ꢀZn and found to be 120 and
126 nm, respectively (see Fig. S1 in ESIw). This slight difference
is expected if we take into account that the size of the indicator
dye as well as the alteration of chemical properties on the
surface of the nanoparticles may affect their solvation sphere.
Furthermore, in both cases, the polydispersity index revealed a
a
REQUIMTE, Departamento de Quı´mica, Faculdade de Cieˆncias e
Tecnologia, Universidade Nova de Lisboa, 2829-516 Caparica,
Portugal
Fraunhofer Research Institution for Modular Solid State
Technologies (EMFT), Department of Polytronic Systems,
Workgroup Sensor Materials, Josef-Engert-Straße 9, D-93053
Regensburg, Germany. E-mail: jennifer.schmidt@emft.fraunhofer.de
Universite´ Lyon 1, CNRS, UMR 5579, LASIM, F-69622
Villeurbanne, France. E-mail: tdoussineau@lasim.univ-lyon.fr
Instituto de Ciencia de Materiales de Sevilla,
centro mixto CSIC-Univ. Sevilla, Avda. Americo Vespucio 49,
b
c
d
41092, Sevilla, Spain. E-mail: lapresta@ugr.es
e
Institute of Analytical Chemistry, Chemo- and Biosensors, University
of Regensburg, D-93040 Regensburg, Germany
w Electronic supplementary information (ESI) available: Experimental
procedures, synthesis and characterization of ATP-sensing nano-
particles and cellular studies. See DOI: 10.1039/c1cc10419e
6
066 Chem. Commun., 2011, 47, 6066–6068
This journal is c The Royal Society of Chemistry 2011

View Article Online
dye and ATP. It is assumed that this interaction is not only
due to the binding of the phosphate groups in ATP to the
Zn(II)–DPA receptor unit, but also due to p–p stacking effects
between the adenine moiety of ATP and the naphthalimide
fluorophore in the indicator dye. The immobilization of 4ꢀZn
onto the surface of the silica nanoparticles may introduce
some sterical hindrance which prevents a more efficient binding
to the analyte, especially in the case of the stacking interactions.
This translates into a loss of sensitivity of the indicator dye
towards ATP. Such p–p stacking effects have previously been
made responsible for the differences in interaction between
closely related nucleotides (which differed only in the nucleic
7
base) and acridine bearing a Zn(II)–cyclen unit. It is assumed
to be the main reason for the observed selectivity towards ATP
4
(and also ADP, as reported previously) when compared with
Fig. 1 Design and architecture of ATP-sensitive nanoparticles.
Description: Light blue—silica core nanoparticle; dark blue—
undoped silica shell; —reference dye; —indicator dye.
the other nucleotides having different bases (see Fig. S2 in
ESIw).
The fluorescence of the nanoparticles was also measured in
the presence of other nucleotides, namely guanosine triphos-
phate (GTP), uridine triphosphate (UTP) and cytidine tripho-
sphate (CTP), at three different analyte concentrations. A
summary of these results is represented in Fig. 3. Although
all of these nucleotides present very similar structures, differing
only in the nucleobase, ATP induces a considerably stronger
response in the fluorescence of the indicator dye than any of
the other analytes under study. Nevertheless, the difference in
sensitivity between ATP and the other nucleotides provided by
the covalently immobilized indicator dye is decreased when
compared to the free indicator dye in solution. This could
indicate that the electrostatic interactions between the
Zn(II)–DPA receptor group of the indicator dye and the
phosphate groups of the analytes become more relevant than
the aforementioned p–p stacking effects when 4ꢀZn is immo-
bilized onto nanoparticles. This is also consistent with the
previous explanation for the recorded loss of sensitivity.
In order to confirm that the Zn(II)–DPA centre endured the
synthetic procedure and was, in fact, the recognition element
homogenous size distribution, with values of 0.05 and 0.07, for
prior and post functionalization, respectively.
The analytical performance of the nanoparticles towards
ATP was studied in aqueous buffer at pH 7.4 for different
analyte concentrations using fluorescence spectroscopy. The
resulting fluorescence emission spectra are shown in Fig. 2.
Upon addition of 330 mM ATP to a suspension of the
ꢂ1
fluorescent nanoparticles (1.0 ꢁ 0.2 mg mL ), an increase
in the emission of the indicator dye (up to 0.5 fold) was
observed while the signal of the reference dye remained almost
unaffected (around 11% when compared to the fluorescence of
the indicator dye). This observation is in accordance with
results for the free indicator dye in solution. However, the
signal changes observed for 4ꢀZn when immobilized onto the
nanoparticles are smaller compared to the values obtained in
solution (around 2.3 fold). In fact, the apparent association
constant (Kass) of the fluorescent nanoparticles with ATP,
which was determined by plotting the fluorescence intensity
changes against the logarithm of analyte concentration
3
ꢂ1
(
see the inset in Fig. 2), was found to be 3.3 ꢃ 10
M
,
towards ATP, a simple experiment was performed. To a cuvette
ꢂ1
roughly 50 times smaller than for the free dye in solution
5
ꢂ1
4
containing a suspension of nanoparticles (1.0 ꢁ 0.2 mg mL
)
(
Kass(4ꢀZn) = 1.5 ꢃ 10 M ). This loss of sensitivity can be
with ATP (333 mM), an excess amount (1667 mM) of a
complexing agent, ethylenediaminetetraacetic acid (EDTA),
was added. The resulting spectra show a remarkable decrease
in the fluorescence of the indicator dye, with no visible changes
explained by the interaction mechanism between the indicator
Fig. 2 Changes in fluorescence of the nanoparticles at different ATP
concentrations: 0, 8, 17, 33, 83, 167 and 333 mM (lexc (4ꢀZn) = 450 nm,
Fig. 3 Influence of analytes on the fluorescence properties of the
0
ATP-sensitive nanoparticles. Adenosine-5 -diphosphate (ADP) and
l
exc (Rhod.) = 530 nm).
pyrophosphate (PPi) were also added to the graph for comparison.
This journal is c The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6066–6068 6067

View Article Online
in the reference dye (see Fig. S3 in ESIw). This result supports
that Zn(II) still has been present in the indicator dye molecules
which are attached to the nanoparticles’ surface and that,
upon removal of Zn(II) from the DPA centre using EDTA,
sensitivity towards ATP is no longer observed. With these
results in mind, the possibility of removal and subsequent
rebinding of Zn(II) to the nanoparticles and how this process
affected their sensitivity towards ATP was investigated. Proven
successful, this approach would provide a mean of ‘‘recycling’’
the nanoparticles, suggesting their possible application in
8
microfluidics, if the nanoparticles were physically attached
to such a system. Accordingly, an aliquot of nanoparticles was
suspended in a vial containing an aqueous solution of EDTA
and stirred overnight. The nanoparticles were then washed
thoroughly with water. Upon addition of ATP to a suspension
of nanoparticles, no changes were observed in the recorded
spectra. However, when Zn(II) was added to the same cuvette,
a considerable increase of the indicator fluorescence (0.5 fold)
occurred, while the reference dye’s emission remained unaltered
Fig. 4 Confocal fluorescence micrographs of NRK cells incubated
ꢂ1
(
Fig. S4 in ESIw). This proves not only that Zn(II) was
with nanoparticles (0.2 mg mL ) for 48 hours: (a) rhodamine-
removed and rebound, but also that a full regeneration of
the detection capacity of the nanoparticles towards ATP was
achieved.
reference dye, (b) naphthalimide-indicator dye, (c) merge and (d)
rhodamine-reference dye after cell fixation and DAPI-nuclear staining.
cells without using additional agents and without exerting
cytotoxic effects, they are superior to most organic fluoro-
phores and may pave the way to applications in biomedical
studies.
In order to evaluate the potential of the fluorescent nano-
particles for biomedical applications we studied cellular uptake
and cytotoxic effects using epithelial normal rat kidney (NRK)
cells. Nanoparticle cytotoxicity was investigated by impedimetric
monitoring of the cell response during nanoparticle exposure for
We gratefully acknowledge the financial support from the
project MO 1062/5-2 of Deutsche Forschungsgemeinschaft, and
from the EU Projects NASCENT (MRTN-CT-2006-033873)
and NIR NanoBioSens (PIEF-GA-2008-220775). This work
48 hours. Impedimetric monitoring was the method of choice as
it works label-free and the readout does not interfere with the
nanoparticles’ luminescence. The bar chart in Fig. S5 (ESIw)
shows cell viability for different nanoparticle concentrations.
Untreated cells were set to 100%, as positive control cells were
treated with 0.1% (w/v) Saponin which is known to permeablize
the plasma membrane leading to immediate cell lysis. Results
indicate that cell viability was not impaired at all by exposure
¨
was also supported by the Bayerische Staatsministerium fur
Wirtschaft, Infrastruktur, Verkehr und Technologie within
project AZ-Nr.: 20.10-3410-2 (Projekt Sensormaterialien).
Notes and references
to ATP-sensing nanoparticles at concentrations ranging from
ꢂ1
1 (a) A. Bianchi, K. Bowman-James, E. Gracıa-Espana, Supramolecular
´
0.1 to 0.4 mg mL
.
Chemistry of Anions, Wiley-VCH, New York, 1997; (b) E. Takeda,
H. Yamamoto, K. Nashiki, T. Sato, H. Arai and Y. Taketani, J. Cell.
Mol. Med., 2004, 8, 191; (c) C. Spangler, M. Schaeferling and
O. S. Wolfbeis, Microchim. Acta, 2008, 161, 1.
To study nanoparticle incorporation NRK cells were
ꢂ1
incubated with 0.2 mg mL of nanoparticles for 48 hours.
Afterwards, cells were extensively washed using PBS buffer
and the uptake of particles was studied by confocal laser
scanning microscopy. Cells were imaged in physiological
buffer at pH 7.4. Micrographs in Fig. 4 show typical optical
xy-sections through the NRK cell bodies. The fluorescence of
the rhodamine dye (a) and the naphthalimide dye (b) were
recorded separately using individual excitation and emission
filters. As depicted in the merged image (c) the reference
and indicator dyes perfectly co-localize in the cytoplasm of
the cells. Mixed colors indicate that particles are present
in different cellular compartments. For a better evaluation
of intracellular localization of nanoparticles a DAPI-based
staining of the nucleus was performed in addition. Subsequent
microscopic inspection reveals that nanoparticles are present
in the cytoplasm but not in the nucleoplasm (Fig. 4d). Further
research regarding nanoparticle uptake mechanism and
detailed intracellular localization are currently ongoing. Since
nanoparticles were incorporated in non-phagocytosing animal
2
(a) Y. Jin, S. Kannan, M. Wu and X. Zhao, Chem. Res. Toxicol.,
2007, 20, 1126; (b) A. Dahm and H. Eriksson, J. Biotechnol., 2004,
111, 279.
3 (a) A. Ojida, Y. Mito-oka, M. Inoue and I. Hamachi, J. Am. Chem.
Soc., 2002, 124, 6256; (b) T. Sakamoto, A. Ojida and I. Hamachi,
Chem. Commun., 2009, 141.
4
A. J. Moro, P. J. Cywinski, S. Ko
Commun., 2010, 46, 1085.
(a) T. Doussineau, S. Trupp and G. J. Mohr, J. Colloid Interface
Sci., 2009, 339, 266; (b) A. Grafe, S. E. Stanca, S. Nietzsche,
L. Kubicova, R. Beckert, C. Biskup and G. J. Mohr, Anal. Chem.,
008, 80, 6526.
¨
rsten and G. J. Mohr, Chem.
5
¨
2
6 (a) D. R. Larson, H. Ow, H. D. Vishwasrao, A. A. Heikal,
U. Wiesner and W. W. Webb, Chem. Mater., 2008, 20, 2677;
(b) S. Roy, R. Woolley, B. D. MacCraith and C. McDonagh,
Langmuir, 2010, 26, 13741.
7 M. W. Hosseini, A. J. Blacker and J.-M. Lehn, J. Am. Chem. Soc.,
1990, 112, 3896; M. Shionoya, T. Ikeda, E. Kimura and M. Shire,
J. Am. Chem. Soc., 1994, 116, 3848.
8
(a) K. R. Ackermann, T. Henkel and J. Popp, ChemPhysChem,
007, 8, 2665; (b) A. Funfak, J. Cao, O. S. Wolfbeis, K. Martin and
2
J. M. Ko
¨
hler, Microchim. Acta, 2009, 164, 279.
6
068 Chem. Commun., 2011, 47, 6066–6068
This journal is c The Royal Society of Chemistry 2011