Photomodulation of fluorescent upconverting nanoparticle markers in live organisms by using molecular switches

Article abstract of DOI:10.1002/chem.201103767

In vivo upconversion: Photoresponsive dithienylethene ligands decorated onto lanthanide-doped, upconverting nanoparticles are used to modulate the amount of emission from the nanoparticles. The hybrid system can be toggled between 'bright' and 'dark' states by using two different types of light in living nematodes (see figure). Copyright

Full text of DOI:10.1002/chem.201103767

DOI: 10.1002/chem.201103767
Photomodulation of Fluorescent Upconverting Nanoparticle Markers in Live
Organisms by Using Molecular Switches
John-Christopher Boyer,[a] Carl-Johan Carling,[a] Shu Yi Chua,[b] Danielle Wilson,[a]
Bob Johnsen,[b] David Baillie,*[b] and Neil R. Branda*[a]
Although fluorescence microscopy continues to be the
method of choice by biomedical researchers to reconstruct
images of cells and tissues containing optical labels,[1] there
still exist many limitations. The diffraction of light imposes
the major drawback and reduces the spatial resolution ach-
ievable by conventional microscopes, which cannot distin-
guish between molecular or nanoscale probes residing in
close proximity to each other. This limitation can be over-
come by multiphoton approaches and the help of molecular
switches that, by undergoing reversible photoreactions be-
tween two isomers with different absorption properties, can
turn ꢀoffꢁ and ꢀonꢁ the fluorescent signal by using light sour-
ces different from the one used to initiate the emission.[2] In
this way, temporal control of emission offers heightened spa-
tial resolution. Addition of the component of time provides
better tracking of specific biological species in vivo and can
potentially identify false positive signals at subdiffraction
scales.[3]
Several issues must be considered when designing systems
for use in this technology and many challenges must be
overcome. The type of light used to induce emission of the
probe cannot be the same as those that trigger the forward
and reverse isomerization reactions of the photoswitch.
(Typically, it must lie outside the UV and visible regions of
the spectrum.) This is a particularly difficult issue to avoid
when two types of lights are needed to drive the photo-
chemistry along the two reaction pathways. The fluorescent
probes cannot be prone to photodegradation, which short-
ens the fluorescence lifetime, as is often the case for organic
systems. The probes should also lack undesirable optical and
biochemical properties, such as the ꢀblinkingꢁ and cytotoxici-
ty observed for inorganic quantum dots.[4] Herein we high-
light how the combination of lanthanide-doped, NaYF4, up-
converting nanoparticles and photochromic dithienylethene
dyes is an excellent choice to overcome these issues.
Upconverting nanoparticles (UCNPs) composed of
NaYF4 doped with Er3+ and Yb3+ absorb multiple near-in-
frared (NIR) photons and convert them into light, which is
emitted as relatively narrow bands in several regions of the
visible spectrum.[5] (NIR light is not capable of triggering
the photoreactions of typical photoswitches.) Because they
have long-lived excited states when stimulated with NIR
light, continuous-wave diode lasers can be utilized as excita-
tion sources that reduce autofluorescence and background
interference. They have already been successfully applied to
cell and whole animal imaging, drug release, photodynamic
therapy, and remote-control photorelease/photoswitching.[6]
Their nonblinking behavior and resistance to luminescence
bleaching make them the state of the art in upconverting
technology and are the systems of choice for our studies.
Dithienylethenes (DTEs) are one of the most appealing
photoswitching components. They can be efficiently toggled
between two optically unique, stable isomers with a high
degree of durability by using UV and visible light,[7] which
can be tuned by tailoring the molecular backbone. Herein,
we demonstrate how we can photomodulate the fluores-
cence from UCNPs in vitro and in vivo by using DTE-deco-
rated nanoparticle biolabels. In vivo experiments of this
type have not been shown previously. By using the photores-
ponsive hybrid system shown in Figure 1 (1[NaYF4:ErYb]),
we can change the fluorescence output from a higher inten-
sity green/red emission to a lower intensity red-dominate
fluorescence state. This modulation is feasible because only
the ring-closed isomer of the photoresponsive dye (1c[NaY-
F4:ErYb]) possesses a strong absorption in the appropriate
regions of the visible spectrum to overlap with the green
emission of UCNPs through energy transfer.[8] The ring-
open isomer, on the other hand, does not absorb in the visi-
ble region of the spectrum and is incapable of selective
quenching.
[a] J.-C. Boyer,+ C.-J. Carling,+ D. Wilson, Prof. N. R. Branda
Department of Chemistry and 4D LABS
Simon Fraser University
8888 University Drive, Burnaby, BC (Canada)
Fax : (+001)778-782-3765
E-mail: nbranda@sfu.ca
The doped, upconverting, NaYF4:ErYb nanoparticles
(2 mol% Er3+, 20 mol% Yb3+) were synthesized by using
high-temperature colloidal methods, which result in UCNPs
coated in oleate ligands.[5] We chose to omit the step in
which an undoped, NaYF4, protective shell is grown onto
the outside surfaces of the nanoparticles to maximize the
energy transfer between the two components. Whereas this
[b] S. Y. Chua, B. Johnsen, Prof. D. Baillie
Department of Molecular Biology and Biochemistry
Simon Fraser University
8888 University Drive, Burnaby, BC (Canada)
[+] Authors contributed equally to this work
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103767.
3122
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3122 – 3126

COMMUNICATION
Figure 1. Synthesis of the photoresponsive hybrid system (1o[NaYF4:ErYb]) and a conceptual description of selective quenching of the emission from
the nanoparticles by ring-closing the photoswitch. The images are TEM micrographs that show a) the oleate-coated nanoparticle starting material, b) the
azide-coated nanoparticles, and c) the nanoparticles decorated with the ring-open form of the photoswitch (1o[NaYF4:ErYb]); this illustrates the uni-
form size and shape.
shell is typically used to increase the luminescence of the
UCNP, it would also increase the distance between the lan-
thanide emitters and the DTE photoswitches decorated
onto their surfaces, thus decreasing the quenching effect and
minimizing any differences in luminescence between the on
and off states. Figure 1a is a representative TEM image of
the oleate-coated UCNPs that shows the uniform size (ca.
26 nm) and shape.[9]
The oleate ligands were displaced by azidopropylphospho-
nate ligands, which were prepared according to published
procedures.[10] The phosphonate group was chosen because
of the high affinity for the surface of lanthanide-based nano-
particles, whereas the azide provides a convenient way to
attach the photoswitches through a copper-catalyzed cyclo-
addition (CuAAC) “click” reaction with an alkyne-function-
alized DTE. The appeal of the use of “click” chemistry lies
in the tolerance to a wide range of functional groups, which
will allow for the eventual modular assembly of multifunc-
tional nanoparticles by including additional alkynes in the
coupling steps.[11]
original oleate-coated nanoparticles. Strong absorption
bands were also observed from 1250 to 990 cmꢀ1 corre-
ꢀ
ꢀ ꢀ
sponding to the P O and P O M stretches of the phospho-
nate group, respectively.[9] TEM images of the azide-coated
UCNPs show only small observable changes in the size (ca.
25 nm) and shape of the nanoparticles compared with the
oleate-coated ones (Figure 1b). Dynamic light scattering
(DLS) experiments also showed small change in the diame-
ter of UCNPs as the oleate ligands (ca. 32 nm) are replaced
with azidopropylphosphonate ligands (ca. 33 nm).[9]
The photoresponsive DTE 1o/1c possesses suitable opti-
cal properties for controlling energy transfer from the
UCNPs. It also contains the alkyne necessary for the “click”
chemistry and a methoxypolyethylene glycol (mPEG) chain
to provide water dispersibility and eventual biocompatibility
for future studies. DTE 1o was prepared in four steps as de-
scribed in the Supporting Information. The final decorated
hybrid nanoparticles (1o[NaYF4:ErYb]) were prepared by
reacting the two components (1o and NaYF4:ErYb) in the
presence of a copper catalyst (Figure 1). All reactions were
left a minimum of 12 h before they were diluted with
MeOH and centrifuged to isolate the UCNPs. Any unreact-
ed ligands could be easily removed by washing once more
with water. The FTIR spectrum of 1o[NaYF4:ErYb] shows
a reduction in the azide stretch at approximately 2100 cmꢀ1
Ligand displacement of the oleates for the azidopropyl-
phosphonates was confirmed by FTIR spectroscopy, which
revealed the strong absorbance at approximately 2100 cmꢀ1
attributed to the N=N=N antisymmetric stretch of the azide
group, a feature that is completely absent in the case of the
Chem. Eur. J. 2012, 18, 3122 – 3126
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3123

N. R. Branda et al.
when compared with that for the azide-functionalized ana-
logues along with the appearance of a peak at 1554 cmꢀ1 at-
tributed to the 1,2,3-triazole ring system.[12] The absorption
bands attributed to the phosphonate group (1250–990 cmꢀ1)
were also still present in the spectrum.
TEM images of the UCNPs after the CuACC reaction
showed that they were well dispersed with no changes in
particle morphology (Figure 1c). A size analysis of several
TEM images generated an average particle size of about
26 nm, which corresponds well to the sizes of the original
oleate and azide-coated UCNPs. DLS results showed a
more significant increase in particle diameter to about
42 nm after attachment of DTE 1o to the nanoparticles; we
attribute this to the longer length of the photoresponsive
molecule compared with the original oleate ligand. As ex-
pected, the hydrodynamic diameter of the DTE-functional-
ized nanoparticles is larger than the diameter measured with
TEM, as DLS also takes into account the length of the li-
gands on the nanoparticle surface.
The decorated 1o[NaYF4:ErYb] nanoparticles are highly
stable in water and we observed no settling over a period
longer than a month. We also observed them to be stable in
10 mm phosphate-buffered saline (pH 7.2) for several hours
before settling occurred, an important property for biologi-
cal applications. By using the intensity of the bands in the
UV/Vis absorption spectra corresponding to the ring-open
isomer of the free DTE ligand (1o) and in the hybrid com-
plex 1o[NaYF4:ErYb], and the results from the particle-size
analysis of the decorated nanoparticles, we estimate an ap-
proximate loading of the photoresponsive component on the
surface of each individual nanoparticle as approximately 600
molecules per particle.
Figure 2. a) UV/Vis absorption and emission spectra (lex =980 nm,
150 W/cm2) of an aqueous solution (9.0ꢃ10ꢀ6 m) of 1o[NaYF4:ErYb]
before (c, dark shaded areas) and after irradiation with 365 nm light
(a, light shaded areas) for 2.5 min (1.3 mW/cm2). b) Changes in the
~
*
absorbance at 530 nm ( ), and emission intensities at 538 nm ( ) and
^
652 nm ( ) of a similar solution of 1o[NaYF4:ErYb] when it is alternate-
ly irradiated with UV and visible light. c) Changes in the absorbance at
530 nm of a similar solution of 1c[NaYF4:ErYb] when it is continuously
irradiated with 980 nm light.
Figure 2a shows the optical properties of aqueous solu-
tions of the hybrid system (1[NaYF4:ErYb]) and justifies the
choice of nanoparticle and photoresponsive ligand. The or-
ganic chromophores on the decorated nanoparticles have
similar optical properties and photochromic behavior as the
free ligands (1o/1c):[9] they absorb UV light (365 nm) and
undergo an immediate change in their UV/Vis absorption
spectra. As is typical for the DTE chromophore, the absorp-
tion bands in the high-energy region of the spectrum of both
1o[NaYF4:ErYb] (Figure 2a and S6 in the Supporting Infor-
mation) and 1o (Figure S6) decrease in intensity and new
broad bands in the visible region appear. These are the char-
acteristic changes that occur when the ring-open isomer of
the free ligand (1o) is converted into the ring-closed coun-
terpart (1c). In the present case, the conversion is 1o[NaY-
F4:ErYb]!1c[NaYF4:ErYb] and results in a visual color
change of the solutions from colorless to red at the photo-
stationary state, which contains 82% of the ring-closed
isomer as measured by 1H NMR spectroscopy of the free
ligand (1o/1c). These red solutions can be converted back
to the colorless versions by exposing them to visible light of
wavelengths greater than 434 nm; this triggers the reverse,
ring-opening reaction and regenerates the original spectra.
The upconversion luminescence of 1o[NaYF4:ErYb]
under 980 nm laser-diode excitation (150 Wcmꢀ2) exhibits
two main sets of emission peaks in the visible portion of the
spectra (Figure 2a), from 510–560 (green) and 635–680 nm
(red), corresponding to the [4H11/2, S3/2]! I15/2 and F9/2! I15/2
transitions, respectively. Whereas there is no significant
overlap between the absorption bands of 1o, whether
alone[9] or decorated onto the nanoparticles, the broad,
long-wavelength absorption band (450–650 nm) of the ring-
closed isomer (1c and 1c[NaYF4:ErYb]) completely over-
laps with the green emission of the NaYF4:ErYb UCNPs
and only slightly with the red emission. This implies that sig-
nificant luminescence quenching will only be observed after
the system is exposed to UV light to convert 1o into the
ring-closed counterpart on the nanoparticles. Figure 2a,b
show this to be the case.
5
4
4
4
The green emission (538 nm)[13] of an aqueous solution of
the decorated nanoparticles is 3.5 times less intense (29% of
the original value) when the photoswitch is first converted
to the photostationary state.[14] The red emission (651 nm) is
also quenched, albeit to a lesser extent (to 75% of the origi-
nal value) as can be expected from the lower overlap be-
tween the absorption bands of the DTE and the emission
bands of the UCNP. The red-emission quenching is also ex-
4
plained by the fact that the F9/2 excitation state of the nano-
3124
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3122 – 3126

Nanoparticle Markers in Live Organisms
COMMUNICATION
particles is partially fed through
nonradiative decay from the
5
[4H11/2, S3/2] levels.
We are attributing the fluo-
rescence quenching of the
nanoparticles decorated with
the ring-closed photoisomer
(1c[NaYF4:ErYb]) to a combi-
nation of fluorescence reso-
nance energy transfer (FRET)
and an inner filter effect (the
absorption of the visible light
emitted from the nanoparticle
by the colored ligands, not nec-
essarily on the same nanoparti-
Figure 3. Optical (left) and two-photon upconversion fluorescence (middle and right) microscopy images of
wild-type N2 C. elegans incubated with 1o[NaYF4:ErYb] (0.25 mg of a 0.5 mgmLꢀ1 solution in M9 buffer) that
show the changes in fluorescence due to the photoswitching of the DTE component within the bodies of the
worms. The middle panel shows the strong initial fluorescence prior to exposure to UV light. The right panel
shows the reduced emission that is a result of the ring-closing of the photoswitch with 365 nm light for 2.5 min.
cle). The dominance of the latter mechanism is reflected in
triggered by irradiating the nematodes with 365 nm light for
150 s, which produced a significant reduction in the observa-
ble fluorescence (Figure 3). The changes in the photolumi-
nescence spectra of the hybrid nanoparticles inside the nem-
atodes before and after UV irradiation corresponds to a 50–
60% decrease in emission intensity.[9] The reduced quench-
ing can be explained by a reduction in the inner filter effect
in these more dilute conditions. Importantly, all nematodes
showed little evidence for sensitivity (toxicity to the nano-
particles, damage from the light source) before, during, and
after the imaging experiments, and were still mobile after ir-
radiation. We also noted that the nematodes were still
viable and had progeny after incubation.
We have successfully demonstrated the ability to modu-
late the luminescence from upconverting nanoparticles in
vitro and in vivo by taking advantage of the well-understood
photochemistry of DTE molecular switches decorated onto
nanoparticle surfaces by the use of CuACC chemistry. By
reversibly converting the photoswitch back and forth be-
tween the two isomers by using UV and visible light, we are
able to toggle the system between on (bright fluorescence)
and off (quenched fluorescence) states. This process can be
cycled multiple times before any degradation of the nano-
particles or the photoswitch renders the system impractical.
Our system has an advantage over ꢀcagedꢁ fluorophores as
the luminescence of our designed nanoparticles can be re-
peatedly cycled through on and off states. Systems such as
the one described in this manuscript not only have the ca-
pacity to reversibly control the intensity of the probe emis-
sion, through selective absorption, they can also alter the
color of the emitted light from multicolored probes; this fur-
ther enhances the imaging capabilities.
4
the reduction of the [4H11/2
,
5S3/2]! I15/2 upconversion life-
time of the nanoparticles when the ring-open photorespon-
sive ligands are converted to the ring-closed counterparts
(74 ms for 1o[NaYF4:ErYb] and 61 ms for 1c[NaYF4:ErYb]);
this corresponds to a FRET efficiency of 18%.[15]
Exposing the decorated nanoparticles that contain the
ring-closed photoswitches (1c[NaYF4:ErYb]) to visible light
at wavelengths greater than 434 nm (107 mW/cm2) restores
the original emission intensities for both the green and red
emissions. Although, as can be seen from the results of sub-
jecting them to several photochemical cycles by alternating
between UV and visible light (Figure 2b), the quenching ef-
ficacy gradually diminishes. Because the absorptions that
correspond to the ring-closed isomer of the ligand (Fig-
*
ure 2b, ) also slightly change after each cycle, we believe
that the degradation of the photoswitch is the culprit. This
claim is supported by the fact that prolonged irradiation of
the nanoparticles (1c[NaYF4:ErYb]) with 980 nm light
(150 W/cm2 for 90 min) resulted in negligible ring-opening
of the photoresponsive ligand (Figure 2c) and no loss in the
upconversion efficiency of the nanoparticles. This minor
degradation (likely due to photochemical side reactions in
the oxygen rich aqueous environment) will not hinder the
use of the decorated nanoparticles in biological application
for which the photoswitch is likely to only be subjected to a
few cycles. Future generations of systems may have en-
hanced performance by employing fluorinated DTE photo-
switches because they demonstrate superior stability in
oxygen and water environments.
The regulation of fluorescence imaging in live organisms
was demonstrated in Caenorhabditis elegans N2 hermaphro-
dites, which were incubated for 3 h with the decorated nano-
particles 1o[NaYF4:ErYb] dispersed in M9 buffer. After im-
mobilizing the nematodes with Levamisole in M9 buffer,
they were isolated by centrifugation, mounted on agarose
gel pads and imaged by using 2-photon fluorescence micros-
copy (Figure 3).
Experimental Section
Biological experiments: The worm handling methods described by Bren-
ner were used.[16] C. elegans N2 hermaphrodites were incubated on
OP50-seeded plates at 208C. After four days, the nematodes were lightly
washed off the plates with M9 buffer and aliquoted into separate 200 mL
microtubes that contained 0.25 mg of hybrid nanoparticles (1o[NaY-
F4:ErYb]) in 0.5x M9 buffer. After 3 h incubation, the nematodes were
immobilized by using 2.5 mm Levamisole, mounted on 2% agarose gel
The upconverting fluorescence from 1o[NaYF4:ErYb]
could be clearly observed in the digestive tract of the nemat-
odes, indicating that they had ingested the nanoparticles.
The ring-closing reaction of the photoresponsive ligands was
Chem. Eur. J. 2012, 18, 3122 – 3126
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3125

N. R. Branda et al.
pads, and imaged by using the two-photon upconversion imaging micro-
scope. Samples with reduced background fluorescence, due to presence
of the nanoparticles, were prepared by washing the nematodes in 100 mL
of 2.5 mm Levamisole in M9 buffer and then centrifuged at 1200 rpm for
5 min. This wash was repeated twice before the nematodes were immobi-
lized and imaged.
Zhang, M. Hong, X. Liu, Nature 2010, 463, 1061–1065; c) X. Yea,
J. E. Collins, Y. Kang, J. Chen, D. T. N. Chen, A. G. Yodh, C. B.
Murray, Proc. Natl. Acad. Sci. USA 2010, 107, 22430–22435.
[6] a) D. K. Chatterjee, Y. Zhang, Nanomedicine 2008, 3, 73–82; b) C.-J.
Carling, J.-C. Boyer, N. R. Branda, J. Am. Chem. Soc. 2009, 131,
10838–10839; c) C.-J. Carling, F. Nourmohammadian, J.-C. Boyer,
N. R. Branda, Angew. Chem. 2010, 122, 3870–3873; Angew. Chem.
Int. Ed. 2010, 49, 3782–3785; d) J.-C. Boyer, C.-J. Carling, B. D.
Gates, N. R. Branda, J. Am. Chem. Soc. 2010, 132, 15766–15772;
e) H. S. Mader, P. S. Kele, S. M. Saleh, O. S. Wolfbeis, Curr. Opin.
Chem. Biol. 2010, 14, 582–596; f) F. Wang, D. Banerjee, Y. Liu, X.
Chen, X. Liu, Analyst 2010, 135, 1839–1854.
[7] a) B. L. Feringa, Molecular Switches, Wiley-VCH, Weinheim, 2010;
b) H. Tian, S. Yang, Chem. Soc. Rev. 2004, 33, 85–97; c) L. Ubaghs,
D. Sud, N. R. Branda in Handbook in Thiophene-Based Materials:
Applications in Organic Electronics and Photonics, Vol. 2 (Eds: I. D.
Perepichka, D. Perepichka), John Wiley & Sons, Chichester, 2009,
pp. 783–812.
[8] For examples of photoswitches that have been engineered to reversi-
bly regulate emission from neighboring inorganic and organic
probes, see: a) F. M. Raymo, M. Tomasulo, Chem. Soc. Rev. 2005,
34, 327–336; b) Y. L. Feng, Y. L. Yan, S. Wang, W. H. Zhu, S. X.
Qian, H. Tian, J. Mater. Chem. 2006, 16, 3685–3692; c) H. Zhao, U.
Al-Atar, T. C. S. Pace, C. Bohne, N. R. Branda, J. Photochem. Pho-
tobiol. A-Chem. 2008, 200, 74–82; d) Z. Erno, I. Yildiz, B. Gorodet-
sky, F. M. Raymo, N. R. Branda, Photochem. Photobiol. Sci. 2010, 9,
249–253.
Acknowledgements
This research was supported by the Natural Sciences and Engineering
Research Council (NSERC) of Canada, the Canada Research Chairs
Program, and Simon Fraser University through funding from the Com-
munity Trust Endowment Fund. This work made use of 4D LABS shared
facilities supported by the Canada Foundation for Innovation (CFI), Brit-
ish Columbia Knowledge Development Fund (BCKDF) and Simon
Fraser University. J.-C. Boyer thanks the Michael Smith Foundation for
Health Research for support. C.-J. Carling thanks Simon Fraser Universi-
ty for a Graduate Fellowship.
Keywords: fluorescence imaging · nanoparticles · optical
probes · upconversion
[1] a) J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer,
New York, 2006; b) D. B. Murphy, Fundamentals of Light Microsco-
py and Electronic Imaging, John Wiley & Sons, New York, 2001.
[2] a) I. Yildiz, E. Deniz and F. M. Raymo, Chem. Soc. Rev. 2009, 38,
1859–1867; b) J. Cusido, E. Deniz, F. M. Raymo, Eur. J. Org. Chem.
2009, 2031–2045; c) Y. Yan, M. E. Marriott, C. Petchprayoon, G.
Marriott, Biochem. J. 2011, 433, 411–422; d) C. Petchprayoon, Y.
Yan, S. Mao, G. Marriott, Bioorg. Med. Chem. 2011, 19, 1030–1040.
[3] a) L. Cheng, K. Yang, M. Shao, S.-T. Lee, Z. Liu, J. Phys. Chem. C
2011, 115, 2686–2692; b) L. Zhu, W. Wu, M.-Q. Zhu, J. J. Han, J. K.
Hurst, A. D. Q. Li, J. Am. Chem. Soc. 2007, 129, 3524–3526; c) M.-
Q. Zhu, L. Zhu, J. J. Han, W. Wu, J. K. Hurst, A. D. Q. Li, J. Am.
Chem. Soc. 2006, 128, 4303–4309; d) I. Yildiz, S. Impellizzeri, E.
Deniz, B. McCaughan, J. F. Callan, F. M. Raymo, J. Am. Chem. Soc.
2011, 133, 871–879.
[4] Coating nanoparticles with a biocompatible material reduces the
toxicity, but does not eliminate blinking; a) M. Bruchez, M. Mo-
ronne, P. Gin, S. Weiss, A. P. Alivisatos, Science 1998, 281, 2013–
2016; b) W. C. W. Chan, S. Nie, Science 1998, 281, 2016–2018; c) X.
Michalet, F. F. Pinaud, L. A. Bentolila, J. M. Tsay, J. J. Li, G. Sundar-
esan, A. M. Wu, S. S. Gambhir, S. Weiss, Science 2005, 307, 538–544.
[5] a) Z. Q. Li, Y. Zhang, S. Jiang, Adv. Mater. 2008, 20, 4765–4769;
b) F. Wang, Y. Han, C. S. Lim, Y. Lu, J. Wang, J. Xu, H. Chen, C.
[9] See the Supporting Information for details.
[10] A. K. Tucker-Schwartz, R. L. Garrell, Chem. Eur. J. 2010, 16, 12718–
12726.
[11] H. S. Mader, M. Link, D. E. Achatz, K. Uhlmann, X. Li, O. S. Wolf-
beis, Chem. Eur. J. 2010, 16, 5416–5424.
[12] FTIR spectroscopy shows that not all the azide groups have reacted.
[13] The relative amounts of emission were measured at a single wave-
length for each emission (green and red). The areas under the peaks
were not compared because the relative amount of quenching at
each wavelength stays constant throughout the emission bands (see
Figure S7 in the Supporting Information for details).
[14] This amount of quenching corresponds to the maximum possible, as-
suming a 82% photostationary state. If the open-isomer could be
completely converted to the ring-closed counterpart, the emission
would be quenched to only 16% of the original intensity.
[15] We attribute the low FRET efficiency to the relatively small number
of molecular switches on the surface of the nanoparticles. The life-
times for the red emission were not measured.
[16] S. Brenner, Genetics 1974, 77, 71–94.
Received: November 30, 2011
Published online: February 16, 2012
3126
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3122 – 3126