Photoinduced enhancement in the luminescence of hydrophilic quantum dots coated with photocleavable ligands

Article abstract of DOI:10.1021/ja209873g

In search of strategies to photoactivate the luminescence of semiconductor quantum dots, we devised a synthetic approach to attach photocleavable 2-nitrobenzyl groups to CdSe-ZnS core-shell quantum dots coated with hydrophilic polymeric ligands. The emission intensity of the resulting nanostructured constructs increases by more than 60% with the photolysis of the 2-nitrobenzyl appendages. Indeed, the photoinduced separation of the organic chromophores from the inorganic nanoparticles suppresses an electron-transfer pathway from the latter to the former and is mostly responsible for the luminescence enhancement. However, the thiol groups anchoring the polymeric envelope to the ZnS shell also contribute to the photoinduced emission increase. Presumably, their photooxidation eliminates defects on the nanoparticle surface and promotes the radiative deactivation of the excited quantum dots. This effect is fully reversible but its magnitude is only a fraction of the change caused by the photocleavage of the 2-nitrobenzyl groups. In addition, these particular quantum dots can cross the membrane of model cells and their luminescence increases by ～80% after the intracellular photocleavage of the 2-nitrobenzyl quenchers. Thus, photoswitchable luminescent constructs with biocompatible character can be assembled combining the established photochemistry of the 2-nitrobenzyl photocage with the outstanding photophysical properties of semiconductor quantum dots and the hydrophilic character of appropriate polymeric ligands.

Full text of DOI:10.1021/ja209873g

Article
pubs.acs.org/JACS
Photoinduced Enhancement in the Luminescence of Hydrophilic
Quantum Dots Coated with Photocleavable Ligands
Stefania Impellizzeri,† Bridgeen McCaughan,‡ John F. Callan,*,‡ and Francisco M. Raymo*,†
̧
†Laboratory for Molecular Photonics, Department of Chemistry, University of Miami, 1301 Memorial Drive, Coral Gables, Florida,
33146-0431
‡Department of Pharmacy and Pharmaceutical Sciences, School of Biomedical Sciences, University of Ulster, Coleraine, BT52 1SA,
Northern Ireland, United Kingdom
ABSTRACT: In search of strategies to photoactivate the
luminescence of semiconductor quantum dots, we devised a
synthetic approach to attach photocleavable 2-nitrobenzyl
groups to CdSe−ZnS core−shell quantum dots coated with
hydrophilic polymeric ligands. The emission intensity of the
resulting nanostructured constructs increases by more than
60% with the photolysis of the 2-nitrobenzyl appendages.
Indeed, the photoinduced separation of the organic
chromophores from the inorganic nanoparticles suppresses
an electron-transfer pathway from the latter to the former and is mostly responsible for the luminescence enhancement.
However, the thiol groups anchoring the polymeric envelope to the ZnS shell also contribute to the photoinduced emission
increase. Presumably, their photooxidation eliminates defects on the nanoparticle surface and promotes the radiative deactivation
of the excited quantum dots. This effect is fully reversible but its magnitude is only a fraction of the change caused by the
photocleavage of the 2-nitrobenzyl groups. In addition, these particular quantum dots can cross the membrane of model cells and
their luminescence increases by ∼80% after the intracellular photocleavage of the 2-nitrobenzyl quenchers. Thus,
photoswitchable luminescent constructs with biocompatible character can be assembled combining the established
photochemistry of the 2-nitrobenzyl photocage with the outstanding photophysical properties of semiconductor quantum
dots and the hydrophilic character of appropriate polymeric ligands.
bleaching resistances compared to most organic dyes and
fluorescent proteins. In addition, their narrow and symmetric
emission bands can be tuned across the visible region with
careful adjustments in elemental composition and physical
dimensions. Thus, the identification of mechanisms to
photoactivate the luminescence of quantum dots can translate
into the development of photocaged probes with improved
performance.
Recently, we developed biocompatible CdSe−ZnS core−
shell quantum dots coated with amphiphilic polymeric
ligands.18 The organic envelope around the inorganic core of
these nanostructured constructs preserves their photophysical
properties, while ensuring aqueous solubility and biocompat-
ibility. In addition, the polymeric ligands permit the covalent
attachment of organic dyes via amide-bond formation. On the
basis of these considerations, we envisaged the possibility of
attaching photocleavable groups to our hydrophilic quantum
dots in order to activate their luminescence under optical
control in aqueous environments. Indeed, a recent report19
demonstrates that the photocleavage of 2-nitrobenzyl groups,
adsorbed on the surface of hydrophilic CdTe−CdS InP−ZnS
or CdHgTe−ZnS core−shell quantum dots, leads to a
INTRODUCTION
■
Photocaged fluorophores1−8 and photoactivatable fluorescent
proteins9−11 switch from a nonemissive to an emissive state
upon illumination at an appropriate wavelength and are
valuable analytical tools for the investigation of biological
samples. Indeed, their introduction into a specimen of interest
permits the local activation of fluorescence under optical
control. In turn, fluorescence photoactivation offers the
opportunity to monitor the diffusion of a labeled target as
well as to resolve in time spatially indistinguishable labels. As a
result, these photoresponsive probes in combination with
fluorescence imaging can be exploited to assess the dynamics of
biological processes and visualize the subtleties of biological
structures.
Photocaged fluorophores generally rely on the photoinduced
cleavage of appropriate functional groups to activate the
fluorescence of organic chromophores.1−8 Similarly, their
genetically encoded counterparts activate the fluorescence of
organic chromophores, embedded within polypeptidic frame-
works, on the basis of photoisomerizations and proton transfer
in the excited state.9−11 The photophysical properties of
organic chromophores, however, are inferior to those of certain
inorganic nanoparticles. Specifically, semiconductor quantum
dots12−17 have huge one- and two-photon absorption cross
sections, long luminescence lifetimes, and excellent photo-
Received: October 20, 2011
Published: January 2, 2012
© 2012 American Chemical Society
2276
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
Figure 1. Structures of the copolymer 1 and model compounds 2−6.
significant luminescence enhancement.20−24 These particular
nanoparticles were prepared relying on the established ability of
dihydrolipoic acid derivatives to anchor organic ligands on the
surface of quantum dots.25 These compounds, however, have
only two thiol groups per molecule, while our polymers
incorporate multiple anchoring groups along the same
macromolecular backbone. In fact, our polydentate ligands
can impose long-term stability on the coated quantum dots in
aqueous environments, while permitting the covalent attach-
ment of 2-nitrobenzyl photocages. In this work, we report an
experimental protocol to couple 2-nitrobenzyl groups to our
polymer-coated CdSe−ZnS core−shell quantum dots, together
with the synthesis of appropriate model systems and the
photochemical and photophysical properties of the resulting
photoresponsive nanostructured assemblies.
characterization of its redox behavior respectively. Instead, the
bisthiol 5 lacks the nitro group necessary to promote photolysis
and, after adsorption on the surface of model quantum dots,
permits the investigation of the influence of the anchoring thiol
groups on the nanoparticles under irradiation.
We synthesized the dithiolanes 2 and 4 in one step from the
corresponding benzylamine and thioctic acid under the
assistance of N,N′-dicyclohexylcarbodiimide (DCC). We then
reduced their disulfide linkage with sodium borohydride to
generate the bisthiols 3 and 5, respectively. Similarly, we
prepared the amide 6 in one step from 2-nitrobenzylamine and
acetic anhydride.
Synthesis and Spectroscopy of the Models. The
anchoring thiol groups of 3 and 5 encourage the adsorption
of both compounds on the surface of preformed CdSe−ZnS
core−shell quantum dots coated with TOPO. Indeed, both
compounds displace the native TOPO ligands of the
nanoparticles, when stirred with the latter in ethanol for 12 h.
In both instances, the modified quantum dots can then be
isolated after reiterative centrifugation and filtration steps and
the presence of the organic ligands on their surface can be
RESULTS AND DISCUSSION
■
Design and Synthesis of the Ligands. The copolymer 1
(Figure 1) incorporates anchoring thiol groups, hydrophilic
poly(ethylene glycol) chains and connecting carboxylic acids
along a common macromolecular backbone. When combined
with preformed CdSe−ZnS core−shell quantum dots, coated
with tri-n-octylphosphine oxide (TOPO), this polymer adsorbs
on the surface of the inorganic nanoparticles, displacing the
TOPO ligands, to generate water-soluble quantum dots.18b The
carboxylic acids appended to the polymeric envelope around
these nanostructured constructs can then be coupled to
chromophores with pendant primary amines, under the
assistance of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
(EDC) and the sodium salt of 3-sulfo-N-hydroxysuccinimide
(sulfo-NHS).18b Thus, our hydrophilic quantum dots can
be reacted with 2-nitrobenzylamine on the basis of this
experimental protocol to generate nanoparticles with photo-
cleavable 2-nitrobenzyl groups on their surface. The inves-
tigation of the photochemical and photophysical properties of
the resulting conjugates, however, requires also the preparation
of appropriate model compounds. In particular, the dithiolane
2, bisthiol 3 and amide 6 incorporate a 2-nitrobenzyl
appendage and permit the investigation of the photolysis of
this particular group in solution, its spectroscopic response after
adsorption on the surface of model quantum dots and the
1
confirmed by H nuclear magnetic resonance (NMR) spec-
1
troscopy. Specifically, the H NMR spectra of the quantum
dots, recorded before and after treatment with the bisthiols,
reveal the appearance of resonances for the aromatic rings of
the two ligands.
The absorption spectrum (a in Figure 2) of the model
compound 2, recorded in tetrahydrofuran (THF), shows a
band for the 2-nitrobenzyl chromophore at 261 nm. Upon
ultraviolet illumination, this particular band decays with the
concomitant growth of the characteristic absorption of 2-
nitrosobenzaldehyde at 286 nm (b in Figure 2). These spectral
changes are consistent with the expected photoinduced
cleavage of the 2-nitrobenzyl group.4
We observed a similar behavior for the ligand 3, after its
adsorption on the surface of preformed CdSe−ZnS core−shell
quantum dots. Specifically, the absorption spectrum (c in
Figure 2) of the modified quantum dots reveals a band for the
2-nitrobenzyl chromophore26 at 256 nm and the band gap
absorption of the CdSe core at 440 nm. The intensities of the
two bands indicate the average number of 2-nitrobenzyl groups
2277
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
Figure 3. Emission spectra of CdSe−ZnS core−shell quantum dots
(1.5 μM, THF, 25 °C, λEx = 380 nm) coated with TOPO before (a)
and after (b) ultraviolet irradiation (365 nm, 0.4 mW cm−2, 5, 10, 15,
20, 25, and 30 min). Emission spectra of CdSe−ZnS core−shell
quantum dots (5 μM, THF, 25 °C, λEx = 400 nm) coated with 3
before (c) and after (d) ultraviolet irradiation (365 nm, 0.4 mW cm−2,
5, 10, 20, 35, 45, and 55 min).
Figure 2. Absorption spectra of 2 (0.2 mM, THF, 25 °C) before (a)
and after (b) ultraviolet irradiation (365 nm, 0.4 mW cm−2, 3, 6, 12,
24, and 35 min). Absorption spectra of CdSe−ZnS core−shell
quantum dots (5 μM, THF, 25 °C) coated with 3 before (c) and after
(d) ultraviolet irradiation (365 nm, 0.4 mW cm−2, 5, 10, 20, 35, and
45 min).
change in luminescence appears to be reversible. Specifically,
the emission intensity of the irradiated quantum dots fades over
the course of 24 h upon storage in the dark.28 This apparent
contradiction indicates that the photolysis of the 2-nitrobenzyl
groups cannot be the sole process responsible for the changes
in luminescence.
per quantum dot to be ∼18. Upon ultraviolet irradiation, the
absorbance at 256 nm decreases and a band for the
photogenerated 2-nitrosobenzaldehyde grows at 288 nm (d
in Figure 2). Thus, the 2-nitrobenzyl ligands can be
photocleaved even when they are adsorbed on the inorganic
nanoparticles.
In order to gain further understanding on these effects, we
examined the influence of ultraviolet irradiation on the very
same batch of CdSe−ZnS core−shell quantum dots before and
after adsorption of the model compound 5. The benzyl
terminus of this particular ligand lacks the nitro group of 3 and
cannot be photocleaved. Consistently, the absorption spectra of
both sets of quantum dots do not change upon ultraviolet
illumination. However, their emission spectra vary significantly
with irradiation. Specifically, the luminescence of the native
quantum dots decreases irreversibly by 44% (a and b in Figure 3),
whereas that of the nanoparticles coated with 5 increases by
43% (a and b in Figure 4) under otherwise identical irradiation
conditions. Furthermore, the photoinduced luminescence
increase observed in the case of 5 is reversible. In fact, the
emission intensity decreases upon storage in the dark and can
be switched for multiple cycles simply by alternating irradiation
and storage steps (c in Figure 4). However, the magnitude of
the luminescence changes decreases significantly with the
number of switching cycles, presumably, as a result of the
gradual degradation of the sample. Furthermore, the
luminescence change observed for the quantum dots coated
The emission spectra of CdSe−ZnS core−shell quantum
dots recorded before and after (a and c in Figure 3) the
adsorption of 3 on their surface reveal a negligible change in
emission wavelength, but a significant decrease in emission
intensity. Indeed, the emission band of the quantum dots is
centered at ∼480 nm in both instances, but the luminescence
quantum yield drops from 0.32 to 0.09 after the adsorption of
3. The pronounced decrease in quantum efficiency is,
presumably, a result of photoinduced electron transfer from
the luminescent inorganic core to the organic ligands. In fact,
the oxidation potential of the quantum dots and the reduction
potential of the 2-nitrobenzyl group suggest this process to be
exergonic with a free energy change of ∼−0.4 eV.27
Upon ultraviolet irradiation, the emission intensity of the
quantum dots coated with 3 increases (d in Figure 3) and,
eventually, approaches that recorded before ligand adsorption
(a in Figure 3). These observations suggest that the photolysis
of the organic ligands removes the quenchers from the surface
of the quantum dots and restores their luminescence in full.
However, the photoinduced conversion of the 2-nitrobenzyl
groups into 2-nitrosobenzaldehyde is irreversible, while the
2278
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
Figure 5. Emission spectra of CdSe−ZnS core−shell quantum dots
(1.5 μM, THF, 25 °C, λEx = 380 nm) coated with n-decanethiol before
(a) and after (b) ultraviolet irradiation (365 nm, 0.4 mW cm−2, 5, 10,
15, 20, 25, and 30 min) and subsequent storage in the dark for
48 h (c).
nanoparticles and, therefore, causes an enhancement in
luminescence.33 The thermal back reduction of the ligands
can then restore gradually the surface defects and lower the
emission intensity back to the original value, leading to the
observed luminescence fading upon storage in the dark of an
irradiated sample.
Synthesis and Spectroscopy of the Hydrophilic
Quantum Dots. The CdSe−ZnS core−shell quantum dots
coated with the copolymer 1 can be dispersed in organic
solvents as well as in aqueous environments.18b The
corresponding absorption spectra (a in Figures 6 and 7)
show the band gap absorption of the nanoparticles to be
centered at 450 nm in both THF and neutral phosphate buffer
saline (PBS). Furthermore, the spectra do not change upon
ultraviolet irradiation in both instances. Instead, the emission
band of the quantum dots shifts from 466 nm in THF to 487
nm in PBS with a concomitant decrease in quantum yield from
0.46 to 0.30. In both instances, the emission intensity increases
significantly with ultraviolet irradiation up to a stationary value
(b and c in Figures 6 and 7) and then decreases after storage in
the dark (d in Figures 6 and 7). These changes in luminescence
parallel those observed for quantum dots coated with the model
ligands 3, 5 or n-decanethiol and, thus, appear to the related to
the presence of anchoring thiol groups in the polymeric
envelope around the inorganic core. Furthermore, essentially
the same behavior can be reproduced in degassed solutions
under an atmosphere of argon, indicating that molecular
oxygen does not participate in these processes. By contrast,
illumination of the very same samples with visible light, rather
ultraviolet radiations, does not cause any change in their
emission intensity.
In order to explore the influence of photocleavable 2-
nitrobenzyl groups on the luminescence of our hydrophilic
nanoparticles, we adapted a literature procedure34 to prepare
two batches of CdSe−ZnS core−shell quantum dots with
sufficiently different core diameters to resolve their emission
bands across the visible region. Then, we coated each set with
the copolymer 1 and illuminated the resulting constructs with
ultraviolet radiations for 30 min. This treatment results in an
increase in luminescence up to a photostationary value (c in
Figure 7), because of the photooxidation of the thiol anchoring
groups in the polymeric envelope. At this point, we reacted
a portion of each set of illuminated quantum dots with
Figure 4. Emission spectra of CdSe−ZnS core−shell quantum dots
(56 nM, THF, 25 °C, λEx = 380 nm) coated with 5 before (a) and
after (b) ultraviolet irradiation (365 nm, 0.4 mW cm−2, 5, 10, 15, and
20 min). Change in the relative emission intensity (c) of CdSe−ZnS
core−shell quantum dots (56 nM, THF, 25 °C, λEx = 380 nm, λEm
=
476 nm) coated with 5 upon ultraviolet irradiation (365 nm, 0.4 mW
cm−2, 10 min) and storage in the dark (24 h).
with 5 is only a fraction of that detected for those coated with
the photocleavable ligand 3. Indeed, the emission intensity of
the nanoparticles coated with 3 increases by 70% upon
ultraviolet irradiation.
The behavior of the quantum dots coated with 5 prompted
us to assess the response of identical nanoparticles coated with
n-decanethiol to ultraviolet irradiation. Once again, the
absorption spectrum does not change upon illumination,
while the emission spectrum reveals an increase in
luminescence of 32% (a and b in Figure 5). As observed for
5, the photoinduced process is reversible and the luminescence
returns to the original value after storing the irradiated
nanoparticles in the dark (c in Figure 5). These observations
suggest that the thiol anchoring groups of 3, 5 and
n-decanethiol are responsible for the photoinduced increase in
emission intensity. Indeed, literature precedents29−32 also
indicate that thiol ligands lead to significant luminescence
enhancements upon prolonged illumination of CdSe quantum
dots with and without protective inorganic shells. These
noticeable changes are, presumably, a result of the photo-
induced oxidation of the thiol groups.29 In fact, the cyclic
voltammogram of 5 shows that the bisthiol appendage can be
oxidized to the corresponding disulfide at +0.45 V vs Ag/AgCl.
This value in combination with the reduction potential of
similar CdSe−ZnS core−shell quantum dots suggests that the
photooxidation of the thiol ligands is exergonic with a free
energy change of ∼−1.5 eV.27 Presumably, this process
eliminates a fraction of the defects on the surface of the
2279
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
2-nitrobenzylamine in PBS, under the assistance of EDC and
sulfo-NHS, purified the final conjugates by size-exclusion
chromatography and confirmed the presence of 2-nitrobenzyl
1
groups on the nanoparticle surface by H NMR spectroscopy.
Finally, we combined irradiated and conjugated quantum dots
of one dimension with irradiated, but unconjugated, nano-
particles of the other and vice versa. The emission spectra (a
and c in Figure 8) of the resulting dispersions reveal bands
Figure 6. Absorption spectra of CdSe−ZnS core−shell quantum dots
(1 μM, THF, 25 °C) coated with 1 (a). Emission spectra of CdSe−
ZnS core−shell quantum dots (1 μM, THF, 25 °C, λEx = 380 nm)
coated with 1 before (b) and after (c) ultraviolet irradiation (365 nm,
0.4 mW cm−2, 10, 20, and 30 min) and subsequent storage in the dark
for 12 h (d).
Figure 8. Emission spectra of mixtures (PBS, pH = 7.4, 25 °C, λEx
=
420 nm) of two sets of CdSe−ZnS core−shell quantum dots, both
coated with 1 but differing in core diameter, recorded after ultraviolet
irradiation for 30 min (365 nm, 0.4 mW cm−2) and conjugation of 2-
nitrobenzylamine to the set emitting at long (a) or short (c)
wavelengths and after further ultraviolet irradiation for 5, 10, and 15
min (b and d). The concentrations of the quantum dots emitting at
short wavelengths are 0.1 (a and b) and 1.8 μM (c and d) and those of
the ones at emitting long wavelengths are 0.8 (a and b) and 5.6 μM (c
and d).
centered at 485 and 531 nm for the two sets of quantum dots
present in each mixture. However, the 2-nitrobenzyl groups are
conjugated to the nanoparticles emitting at long wavelengths in
one instance (a in Figure 8) and to those emitting at short
wavelengths in the other (c in Figure 8). The time (<5 h)
required to perform all of these experimental steps is relatively
short compared to that (>24 h) necessary to revert in full the
luminescence enhancement caused by the thiol groups upon
ultraviolet illumination. As a result, the further irradiation of the
nanoparticles reveals almost exclusively changes in emission
intensity caused by the 2-nitrobenzyl groups. In fact, the
luminescence of the conjugated quantum dots increases
significantly in both mixtures upon illumination (b and d in
Figure 8), as a result of the photoinduced cleavage of their
Figure 7. Absorption spectra of CdSe−ZnS core−shell quantum dots
(4 μM, PBS, pH = 7.4, 25 °C) coated with 1 (a). Emission spectra of
CdSe−ZnS core−shell quantum dots (4 μM, PBS, pH = 7.4, 25 °C,
λEx = 380 nm) coated with 1 before (b) and after (c) ultraviolet
irradiation (365 nm, 0.4 mW cm−2, 10, 20, and 35 min) and
subsequent storage in the dark for 12 h (d).
2280
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
Figure 9. Phase-contrast (a and c) and luminescence (b and d) images (λEx = 800 nm, scale bar = 50 μm), recorded before (a and b) and after (c and
d) irradiation (365 nm, 0.4 mW cm−2, 30 min), of CHO cells incubated with CdSe−ZnS core−shell quantum dots (30 nM), coated with 1 and
conjugated to 2-nitrobenzylamine, for 3 h.
2-nitrobenzyl groups. Instead, the emission intensity of the
unconjugated nanoparticles remains essentially unaffected.
Thus, these results demonstrate unequivocally that the
photolysis of 2-nitrobenzyl quenchers adsorbed on the surface
of hydrophilic CdSe−ZnS core−shell quantum dots can indeed
be exploited to switch the nanoparticle luminescence under
optical control in aqueous environments. Once again, these
changes are most likely a consequence of the ability of the 2-
nitrobenzyl group to quench the luminescence of the CdSe
core on the basis of electron transfer, in agreement with the
behavior of analogous quantum dots coated with the model
ligand 3.27 However, the poly(ethylene glycol) spacers of the
polymeric ligand 1, connecting the quenchers to the inorganic
nanoparticles, are significantly longer than the aliphatic chain of
the model ligand 3. Presumably, the relatively flexible polymeric
chains fold back to bring the 2-nitrobenzyl groups in close
proximity to the ZnS shell and permit the transfer of electrons
upon excitation.
Intracellular Luminescence Photoactivation. The
photoinduced luminescence enhancement observed in aqueous
environment (Figure 8), coupled to the established ability of
our polymeric coatings to promote cellular internalization,18b
encouraged us to investigate the intracellular behavior of
nanoparticles bearing the photocleavable quenchers on their
surface. Specifically, we incubated Chinese Hamster Ovarian
(CHO) cells with CdSe−ZnS core−shell quantum dots,
passivated with 1 and conjugated to 2-nitrobenzylamine, for 3 h.
We then imaged the resulting specimen with a two-photon
fluorescence microscope, operating at an excitation wavelength
of 800 nm. The corresponding phase-contrast image (a in
Figure 9) clearly reveals the contour of individual cells, which,
however, cannot be observed in the luminescence-only
counterpart (b in Figure 9). After ultraviolet illumination, the
emission intensity of the internalized quantum dots increases
significantly and the stained cells become visible in both phase-
contrast and luminescence-only images (c and d in Figure 9).
Thus, the photoinduced cleavage of the 2-nitrobenzyl
quenchers, with the concomitant enhancement in the nano-
particle luminescence observed in organic and aqueous
solutions, occurs also within the intracellular environment. In
order to quantify such luminescent enhancement, we measured
the emission intensity at intervals of 0.3 μm along lines drawn
across individual cells before and after ultraviolet illumination.
The resulting averaged intensities indicate a photoinduced
luminescence increase of 77%.35
CONCLUSIONS
■
Photocleavable 2-nitrobenzyl groups can be attached covalently
to the polymeric coating of hydrophilic CdSe−ZnS core−shell
quantum dots via amide-bond formation. Ultraviolet illumina-
tion of the resulting constructs in neutral buffer cleaves the 2-
nitrobenzyl appendages from the surface of the nanoparticles
and leads to a significant luminescence enhancement. Control
experiments with model systems suggest that the 2-nitrobenzyl
chromophores can accept an electron from the excited
quantum dots and quench their luminescence. Therefore, the
photoinduced removal of the 2-nitrobenzyl quenchers
suppresses this particular electron-transfer pathway and
enhances the nanoparticle luminescence. However, the
behavior of the model systems also reveals that, in addition
to the 2-nitrobenzyl chromophores, the thiol groups respon-
sible for the adsorption of the polymeric coating on the ZnS
shell of the quantum dots affect the excitation dynamics of
these nanostructured constructs. Even in the absence of the
2281
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
extracted with CHCl3 (3 × 50 mL). The organic phase was dried over
Na2SO4 and the solvent distilled off under reduced pressure to afford 3
photocleavable 2-nitrobenzyl quenchers, ultraviolet illumination
of the nanoparticles results in a noticeable luminescence
increase. Nonetheless, the magnitude of the emission change
caused by the thiol groups is only a fraction of that associated
with the cleavage of the 2-nitrobenzyl chromophores.
Furthermore, the luminescence enhancement caused by the
thiol groups is reversible and is not affected by the presence of
molecular oxygen as well as by the nature of the solvent.
Electrochemical data suggest that the photooxidation of these
particular groups and their thermal back reduction are,
presumably, responsible for the reversible change in emission
intensity. In addition, the polymeric coating around the
inorganic core encourages the internalization of these photo-
switchable constructs in model cells. Ultraviolet illumination of
stained cells results in the intracellular cleavage of the 2-
nitrobenzyl quenchers with a luminescent enhancement
approaching 80%. In summary, our results demonstrate that
the photolysis of 2-nitrobenzyl groups can be exploited to
enhance the luminescence of CdSe−ZnS core−shell quantum
dots in organic solvents, aqueous solutions and even inside
living cells. However, they also indicate that the thiol groups
responsible for ligand adsorption contribute to the photo-
induced luminescence enhancement.
1
(200 mg, 89%) as a light yellow oil. ESIMS: m/z = 342 [M]+; H
NMR (CDCl3): δ = 1.38−1.40 (2H, m), 1.59−1.65 (6H, m), 1.84−
1.87 (1H, m), 2.16−2.19 (2H, t), 2.39−2.41 (1H, m), 3.06−3.13 (2H,
m), 3.50−3.52 (1H, m), 4.62−4.64 (2H, s), 7.43−7.45 (1H, m), 7.58−
7.64 (2H, m), 8.00−8.03 (1H, m).
Synthesis of 4. A solution of DCC (0.37 g, 1.8 mmol) in CH2Cl2
(10 mL) was added dropwise over the course of 30 min to a solution
of benzylamine (175 μL, 1.6 mmol) and thioctic acid (0.50 g, 2.4
mmol) in CH2Cl2 (20 mL) maintained at 0 °C under Ar. The reaction
mixture was allowed to warm up to ambient temperature and stirred
for 24 h under these conditions. The resulting precipitate was filtered
off and the solvent was distilled off under reduced pressure. The
residue was purified by column chromatography [CHCl3/MeOH
(22:1, v/v)] to afford 4 (0.29 g, 62%) as a yellow oil. ESIMS: m/z =
1
295 [M]+; H NMR (CDCl3): δ = 1.41−1.43 (2H, m), 1.64−1.66
(4H, m), 1.85−1.87 (1H, m), 2.15−2.19 (2H, t, 8 Hz), 2.38−2.42
(1H, m), 3.07−3.14 (2H, m), 3.50−3.53 (1H, m), 4.36−4.38 (2H, s),
7.26−7.28 (5H, m); 13C NMR (CDCl3): δ = 25.9, 29.4, 35.1, 36.8,
38.9, 40.7, 43.9, 56.9, 126.8, 127.9, 128.2, 128.8, 129.1, 138.9, 173.3.
Synthesis of 5. A mixture of 4 (240 mg, 0.8 mmol) and NaBH4
(25 mg, 0.8 mmol) in MeOH (15 mL) was stirred at ambient
temperature for 3 h, diluted with aqueous NaCl (1M, 85 mL) and
extracted with CHCl3 (3 × 50 mL). The organic phase was dried over
Na2SO4 and the solvent distilled off under reduced pressure to afford 5
1
(210 mg, 88%) as a colorless oil. ESIMS: m/z = 296 [M]+; H NMR
EXPERIMENTAL PROCEDURES
(CDCl3): δ = 1.42−1.43 (2H, m), 1.64−1.66 (6H, m), 1.89−1.94
(1H, m), 2.15−2.18 (2H, t, 8 Hz), 2.39−2.43 (1H, m), 3.07−3.15
(2H, m), 3.52−3.55 (1H, m), 4.36−4.39 (2H, s), 7.29−7.32 (5H, m).
Synthesis of 6. A solution of 2-nitrobenzylamine (62 mg, 0.4
mmol) and acetic anhydride (0.39 mL, 0.4 mmol) in CH2Cl2 (5 mL)
was stirred at ambient temperature for 2 h. The solvent was distilled
off under reduced pressure and the residue was crystallized from EtOH
to give 6 (48 mg, 61%) as a yellow solid. ESIMS: m/z = 194 [M]+; 1H
NMR (CDCl3): δ = 1.97 (3H, s), 4.62−4.64 (2H, s), 7.40−7.46 (1H,
m), 7.61−7.69 (2H, m), 8.01−8.04 (1H, d); 13C NMR (CDCl3): δ =
23.5, 31.7, 113.2, 114.7, 116.1, 117.6, 119.9, 120.6, 165.4.
Adsorption of 3 and 5 on the Quantum Dots. A dispersion of
CdSe−ZnS core−shell quantum dots coated with TOPO in hexane
(0.02 mM, 2 mL) was diluted with EtOH (20 mL) and subjected to
centrifugation. The supernatant was discarded and the solid residue
was dispersed in CHCl3 (10 mL) and diluted with a solution of either
3 or 4 (300 mg) in CHCl3 (10 mL). The solvent was distilled off
under reduced pressure and the residue was dispersed in EtOH (3
mL) and stirred at 70 °C for 12 h in a sealed vial under Ar. After
cooling down to ambient temperature, the mixture was diluted with
EtOH (5 mL) and transferred to a centrifuge tube. Consecutive
aliquots (1 mL) of hexane were added with vigorous shaking until the
formation of a precipitate was observed. After centrifugation, the
precipitate was separated from the supernatant and dispersed in THF
(3 mL) to afford the modified quantum dots.
■
Materials and Methods. Chemicals were purchased from
commercial sources and used as received with the exception of
CH2Cl2 and THF, which were distilled over CaH2 and Na/
benzophenone respectively, and H2O, which was purified with a
Barnstead International NANOpure DIamond Analytical system.
CdSe−ZnS quantum dots coated with either TOPO or 1 were
synthesized according to literature procedures.18b All reactions for the
synthesis of 2−6 were monitored by thin-layer chromatography, using
aluminum sheets coated with silica. Electrospray ionization mass
spectra (ESIMS) were recorded with a Bruker micrOTO-Q II
spectrometer. NMR spectra were recorded with either a Bruker
Avance 300 or a Bruker Avance 400 spectrometer. Absorption spectra
were recorded with a Varian Cary 100 Bio spectrometer, using quartz
cells with a path a length of 0.5 cm. Emission spectra were recorded
with a Varian Cary Eclipse spectrometer in aerated solutions.
Luminescence quantum yields were determined with a fluorescein
standard, following a literature protocol.36 Electrochemical measure-
ments were performed with a CH Instruments 660 electrochemical
analyzer under Ar, using a glassy-carbon working electrode (0.3 cm), a
Pt counter electrode and a Ag/AgCl (3 M KCl) reference electrode.
Samples were irradiated at 365 nm (0.4 mW cm−2) with a Mineralight
UVGL-25 lamp or at 562 nm (0.3 mW cm−2) with a Spectral Energy
LH150/1 light source and their absorption and emission spectra were
recorded immediately after illumination.
Conjugation of 2-Nitrobenzylamine to the Quantum Dots. A
dispersion of CdSe−ZnS core−shell quantum dots coated with 1 in
PBS (3.4 μM, 400 μL, pH = 7.4) was combined with solutions of 2-
nitrobenzylamine in DMSO (0.66 mM, 10.3 μL), EDC in PBS (10.4
mM, 25.9 μL, pH = 7.4) and sulfo-NHS in PBS (9.2 mM, 146.8 μL,
pH = 7.4). The mixture was stirred at ambient temperature for 4 h and
purified by size-exclusion chromatography [GE Healthcare PD-10,
PBS (pH = 7.4)] to afford the modified quantum dots.
Intracellular Luminescence Photoactivation. CHO cells were
cultured in F-12 nutrient mixture and supplemented with fetal bovine
serum (10%, v/v), penicillin (200 U mL−1), streptomycin (200 μg mL−1)
and glutamine (2 mM). After reaching confluency, the cells were
harvested by trypsinization and seeded at a density of 5 × 104
cells mL−1 in a six-well plate containing one sterile cover slide (22 mm
×22 mm) per well. The cells were incubated at 37 °C with O2/CO2/
air (20:5:75, v/v/v) overnight and then in the presence of CdSe−ZnS
core−shell quantum dots (30 nM), coated with 1 and conjugated to 2-
nitrobenzylamine, for a further 3 h. The coverslips were removed,
Synthesis of 2. A solution of DCC (0.32 g, 1.5 mmol) in CH2Cl2
(9 mL) was added dropwise over the course of 30 min to a solution of
2-nitrobenzylamine (0.22 g, 1.4 mmol) and thioctic acid (0.44 g,
2.1 mmol) in CH2Cl2 (20 mL) maintained at 0 °C under Ar. The
reaction mixture was allowed to warm up to ambient temperature and
stirred for 24 h under these conditions. The resulting precipitate was
filtered off and the solvent was distilled off under reduced pressure.
The residue was purified by column chromatography [CHCl3/MeOH
(22:1, v/v)] to afford 2 (0.27 g, 56%) as a brown oil. ESIMS: m/z =
1
340 [M]+; H NMR (CDCl3): δ = 1.37−1.39 (2H, m), 1.58−1.64
(4H, m), 1.79−1.84 (1H, m), 2.15−2.19 (2H, t, 8 Hz), 2.39−2.40
(1H, m), 3.05−3.12 (2H, m), 3.49−3.51 (1H, m), 4.62−4.63 (2H, s),
7.40−7.44 (1H, m), 7.57−7.63 (2H, m), 7.99−8.02 (1H, d, 8 Hz); 13C
NMR (CDCl3): δ = 25.3, 27.5, 34.2, 36.3, 38.4, 39.0, 40.2, 56.3, 126.2,
129.7, 131.2, 132.5, 135.9, 146.9, 173.1.
Synthesis of 3. A mixture of 2 (240 mg, 0.7 mmol) and NaBH4
(25 mg, 0.7 mmol) in MeOH (15 mL) was stirred at ambient
temperature for 3 h, diluted with aqueous NaCl (1M, 85 mL) and
2282
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283

Journal of the American Chemical Society
Article
washed with PBS (pH = 7.2, 3 × 1 mL) and fixed onto a glass slide for
imaging. The images were recorded on an inverted Leica SP5
confocal/multiphoton microscope, using a two-photon excitation
wavelength of 800 nm and collecting the emission between 416 and
540 nm.
Menendez, G. O; Etchehon, M. A.; Giordano, L.; Jovin, T. M.; Jares-
Erijman, E. A. ACS Nano 2011, 5, 2795−2805.
́
(24) (a) Tomasulo, M.; Yildiz, I.; Raymo, F. M. Aust. J. Chem. 2006,
59, 175−178. (b) Erno, Z.; Yildiz, I.; Gorodetsky, B.; Raymo, F. M.;
Branda, N. R. Photochem. Photobiol. Sci. 2010, 9, 249−253.
(25) Mattoussi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.;
Sundar, V. C.; Mikulec, F. V.; Bawendi, M. G. J. Am. Chem. Soc. 2000,
122, 12142−12150.
AUTHOR INFORMATION
■
Corresponding Author
*J.Callan@ulster.ac.uk; fraymo@miami.edu
(26) The small difference in wavelength between the 2-nitrobenzyl
absorption of 2 in solution and that of 3 adsorbed on the quantum
dots is indicative of environmental differences around the photo-
cleavable groups in the two instances and confirms the attachment of 3
to the nanoparticles.
ACKNOWLEDGMENTS
■
We thank the National Science Foundation (CAREER Award
CHE-0237578, CHE-0749840 and CHE-1049860) for support-
ing our research program and for providing funds to purchase a
mass spectrometer (CHE-0946858).
(27) The free energy changes (ΔG°) for the photoinduced electron
transfer processes were estimated with equation 1, where e is the
elementary charge, EOx is the oxidation potential of the species
donating the electron, ERed is the reduction potential of the species
accepting the electron, ΔE00 is the optical band gap of the quantum
dots, ε0 is the vacuum permittivity, εr is the dielectric constant of the
medium, and d is the distance between donor and acceptor
(Kavarnos, G. J. Fundamentals of Photoinduced Electron Transfer;
VCH: New York, 1993). The redox potentials of the quantum dots
are +1.35 and −0.91 V vs Ag/AgCl (ref 34). Their optical band gap is
2.82 eV (c in Figure 2). The reduction potential of the 2-nitrobenzyl
group is −1.10 V vs Ag/AgCl and the oxidation potential the bisthiol
anchoring group is +0.45 V vs Ag/AgCl, according to the cyclic
voltammograms of the model compounds 6 and 4, respectively. The
distance between the CdSe core and the 2-nitrobenzyl acceptors as
well as that between the core and the thiol donors can be estimated to
be 3.5 and 2.4 nm, respectively.
REFERENCES
■
(1) Adams, S. R.; Tsien, R. Y. Annu. Rev. Physiol. 1993, 55, 755−784.
(2) Nerbonne, J. M. Curr. Opin. Neurobiol. 1996, 6, 379−386.
(3) Mitchison, T. J.; Sawin, K. E.; Theriot, J. A.; Gee, K.; Mallavarapu,
A. Methods Enzymol. 1998, 291, 63−78.
(4) Pelliccioli, A. P.; Wirz, J. Photochem. Photobiol. Sci. 2002, 319,
441−458.
(5) Goeldner, M.; Givens, R., Eds. Dynamic Studies in Biology:
Phototriggers, Photoswitches and Caged Biomolecules; Wiley−VCH:
Weinheim, 2005.
(6) Ellis-Davies, G. C. Nat. Methods 2007, 4, 619−628.
(7) Thompson, M. A.; Biteen, J. S.; Lord, S. J.; Conley, N.; Moerner,
W. E. Methods Enzymol. 2010, 475, 27−59.
(8) Johnson, I.; Spence, M. T. Z. The Molecular Probes HandbookA
Guide to Fluorescent Probes and Labeling Technologies, 11th ed.; Life
Technologies: Carlsbad, 2010.
2
e
ΔG° = eE − eE
Ox Red
− ΔE
−
00
4πε ε d
0 r
(1)
(9) (a) Lippincott-Schwartz, J.; Altan-Bonnet, N.; Patterson, G. H.
Nat. Cell Biol. 2003, S7−S14. (b) Lippincott-Schwartz, J.; Patterson,
G. H. Science 2003, 300, 87−91. (c) Lippincott-Schwartz, J.; Patterson,
G. H. Trends Cell Biol. 2009, 19, 555−565. (d) Manley, S.; Gillette,
J. M.; Lippincott-Schwartz, J. Methods Enzymol. 2010, 475, 109−120.
(28) The emission intensity of identical quantum dots that were not
irradiated remained unaffected after storage for days in the dark,
indicating that the observed changes are a consequence of
illumination.
(29) Aldana, J.; Wang, Y. A.; Peng, X. J. Am. Chem. Soc. 2001, 123,
́ ́
(10) Fernandez-Suarez, M.; Ting, A. Y. Nat. Cell Biol. 2008, 9, 929−
8844−8850.
943.
(30) Zhelev, Z.; Jose, R.; Nagase, T.; Ohba, H.; Bakalova, R.;
Ishikawa, M.; Baba, Y. J. Photochem. Photobiol. B 2004, 75, 99−105.
(31) Wang, Y.; Tang, Z.; Correa-Duarte, M. A.; Pastoriza-Santos, I.;
(11) Wiedenmann, J.; Gayda, S.; Adam, V.; Oswald, F.; Nienhaus, K.;
Bourgeois, D.; Nienhaus, G. U. J. Biophotonics 2011, 4, 377−390.
(12) Bawendi, M. G.; Steigerwald, M. L.; Brus, L. E. Annu. Rev. Phys.
Chem. 1990, 41, 477−496.
(13) Alivisatos, A. P. Science 1996, 271, 933−937.
(14) Yoffe, A. D. Adv. Phys. 2001, 50, 1−208.
(15) Efros, Al. L.; Rosen, M. Annu. Rev. Mater. Sci. 2000, 30, 475−
521.
Giersig, M.; Kotov, N. A.; Liz-Marzan
108, 15461−15469.
(32) Suffern, D.; Cooper, D.; Carlini, L.; Parbhoo, R.; Bradforth, S.;
Nadeau, J. L. Proc. SPIE 2009, 7189, 718905−1−10.
́
, L. M. J. Phys. Chem. B 2004,
(33) Similar effects were observed for CdSe quantum dots upon
illumination in the presence of elemental sulfur. Rene-Boisneuf, L.;
Scaiano, J. C. Chem. Mater. 2008, 20, 6638−6642.
(16) Burda, C.; Chen, X. B.; Narayana, R.; El-Sayed, M. A. Chem. Rev.
2005, 105, 1025−1102.
(17) Semiconductor Nanocrystal Quantum Dots; Rogach, A. L., Ed.;
Springer: Wien, 2008.
(18) (a) Yildiz, I.; McCaughan, B.; Cruickshank, S. F.; Callan, J. F.;
Raymo, F. M. Langmuir 2009, 25, 7090−7096. (b) Yildiz, I.; Deniz, E.;
McCaughan, B.; Cruickshank, S. F.; Callan, J. F.; Raymo, F. M.
Langmuir 2010, 26, 11503−11511.
(19) Han, G.; Mokari, T.; Ajo-Franklin, C.; Cohen, B. E. J. Am. Chem.
Soc. 2008, 130, 15811−15813.
(34) Impellizzeri, S.; Monaco, S.; Yildiz, I.; Amelia, M.; Credi, A.;
Raymo, F. M. J. Phys. Chem. C 2010, 114, 7007−7013.
(35) The magnitude of the luminescence change is comparable to
that reported for photoswitchable proteins specifically developed for
super-resolution imaging applications. Stiel, A. C.; Andresen, M.;
Bock, H.; Hilbert, M.; Schilde, J.; Schonle, A.; Eggeling, C.; Egner, A.;
̈
Hell, S. W.; Jakobs, S. Biophys. J. 2008, 95, 2989−2997.
(36) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Springer:
New York, 2006.
(20) In alternative to photocleavable groups, photochromic ligands
can be exploited to switch the luminescence of quantum dots with
optical stimulations. Representative examples of such photoswitchable
constructs are reported in refs 21−24.
(21) Medintz, I. L.; Trammell, S. A.; Mattoussi, H.; Mauro, J. M.
J. Am. Chem. Soc. 2004, 126, 30−31.
(22) Zhu, L.; Zhu, M.; Hurst, J. K.; Li, A. D. Q. J. Am. Chem. Soc.
2005, 127, 8968−8970.
(23) (a) Jares-Erijman, E.; Giordano, L.; Spagnuolo, C.; Lidke, K.;
Jovin, T. M. Mol. Cryst. Liq. Cryst. 2005, 430, 257−265. (b) Díaz, S. A.;
2283
dx.doi.org/10.1021/ja209873g | J. Am. Chem.Soc. 2012, 134, 2276−2283