Cellular uptake mediated off/on responsive near-infrared fluorescent nanoparticles

Article abstract of DOI:10.1021/ja208086e

Fluorescence imaging, utilizing molecular fluorophores, often acts as a central tool for the investigation of fundamental biological processes and offers huge future potential for human imaging coupled to therapeutic procedures. An often encountered limit

Full text of DOI:10.1021/ja208086e

COMMUNICATION
pubs.acs.org/JACS
Cellular Uptake Mediated Off/On Responsive Near-Infrared
Fluorescent Nanoparticles
Aniello Palma,† Luis A. Alvarez,‡ Dimitri Scholz,‡ Daniel O. Frimannsson,† Marco Grossi,† Susan J. Quinn,†
and Donal F. O’Shea*,†
†Centre for Synthesis and Chemical Biology, School of Chemistry and Chemical Biology, University College Dublin,
Belfield, Dublin 4, Ireland
‡UCD Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland
S Supporting Information
b
ABSTRACT: Fluorescence imaging, utilizing molecular
fluorophores, often acts as a central tool for the investigation
of fundamental biological processes and offers huge future
potential for human imaging coupled to therapeutic proce-
dures. An often encountered limitation with fluorescence
imaging is the difficulty in discriminating nonspecific back-
ground fluorophore emission from a fluorophore local-
ized at a specific region of interest. This limits imaging to
individual time points at which background fluorescence has
been minimized. It would be of significant advantage if the
fluorescence output could be modulated from off to on in
response to specific biological events as this would permit
imaging of such events in real time without background
interference. Here we report our approach to achieve this for
Figure 1. Switching mechanism design for off/on fluorescent nanoconstructs.
the most fundamental of cellular processes, i.e. endocytosis.
We describe a new near-infrared off to on fluorescence
If fluorophores could be engineered to remain fluorescent silent
and only be activated in recognition of a specific biological event
this would provide a huge signal-to-noise ratio in the targeted
area and minimize background fluorescence. This would open
the way to continuous real-time imaging rather than acquiring
single time point static images following clearance of a non-
specific fluorophore (either by washing for in vitro cells or natural
clearance in vivo).
Fluorophore immobilization on or within particles can yield
significant photophysical benefits and is being increasingly
investigated for roles in nanomedicine, nanobiotechnology,
and fluorescence imaging.11,12 Recently, the advent of particle-
based fluorophores has focused on noncovalently linked dye-
doped polymer particles. In contrast, the development of parti-
cles with surface covalently bound fluorophores is far more
limited. A potential advantage of covalently linking the fluor-
ophores at the surface of a particle is that they may be induced
to respond to stimuli in the surrounding microenvironment as
they are not shielded by a surrounding particle. Therefore the
engineering of surface modified fluorescent nanoparticles which
can act as a responsive fluorescent imaging agent represents a
worthwhile pursuit (Figure 1).
switchable nanoparticle construct that is capable of switch-
ing its fluorescence on following cellular uptake but remains
switched off in extracellular environments. This permits
continuous real-time imaging of the uptake process as
extracellular particles are nonfluorescent. The principles
behind the fluorescence off/on switch can be understood
by encapsulation of particles in cellular organelles which
effect a microenvironmental change establishing a fluores-
cence signal.
he use of organic fluorophores as in vitro reporters of
T
dynamic biological and environmental processes is ubiqui-
tous in molecular and cell biology research programs. The recent
technological advances in noninvasive small animal fluorescence
imaging and the future prospect of routine human fluorescence
imaging have given rise to a resurgent interest in suitable
molecular fluorophores which have emission in the near-infrared
(>700 nm) spectral region.1 Their simplicity of use, coupled with
high sensitivity, ensures a continual expansion of their applica-
tions for in vitro and in vivo imaging.2À5 While several new classes
of molecular NIR probes are currently emerging6À8 the oppor-
tunity exists to develop a superseding next generation of smart
fluorescent probes which have both the optimal photophysical
characteristics and the ability of switching their fluorescence
signal from off to on in response to specific biological stimuli.9,10
In this report we outline the design, synthesis, characteriza-
tion, and in vitro illustration of a cellular uptake activated off to on
switchable fluorescent nanoparticles. The switching mechanism
Received: August 26, 2011
Published: November 10, 2011
r
2011 American Chemical Society
19618
dx.doi.org/10.1021/ja208086e J. Am. Chem. Soc. 2011, 133, 19618–19621
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Synthesis of NP1a and NP1b
Figure 2. Schematic of cellular uptake controlled off/on fluorescent
nanoparticle switching.
is based on the control of molecular fluorophore aggregates at the
particle surface.13 The particle off state (nonfluorescent) is
engineered in aqueous environments by excited state quenching
due to hydrophobic fluorophore aggregation (Figure 1, off state).
Yet in the presence of a micelle forming deaggregating agent the
individual fluorophore molecules would be able to relax away
from the surface of the particle and be shielded from the sur-
rounding water molecules thereby giving rise to a large enhance-
ment of their fluorescence intensity (Figure 1, on state).
Our first anticipated biological application was to use these
nanoparticle constructs for continuous live imaging of their
cellular uptake. Effective visualization of this process in real time
requires extracellular nanoparticles to have low fluorescence
intensity, due to the quenching effects described above, but upon
uptake into cells the particles would be surrounded by cellular
vesiclessuchasendosomeswhichwouldturnontheirfluorescence
(Figure 2). While many different cellular uptake mechanisms
for particles exist, a common feature is their encapsulation
within a cellular vesicle upon uptake which could effect a
microenvironmental change and cause the fluorescence to be
switched on. This approach would eliminate background fluor-
escence and permit continuous imaging of the internalization
of particles in the presence of nonfluorescent extracellular
particles.
In order to build such a responsive nanoconstruct, our par-
ticles of choice were poly(styrene-co-methacrylic acid) polymers
which have a defined high loading of surface carboxylic acid
functional groups available for synthetic transformations. For
ease of handling a 200 nm diameter particle size was selected for
this initial study. As it is strongly preferential to image in the near-
infrared spectral region, the BF2 chelated tetraarylazadipyrro-
methenes were selected as fluorophores as they have a strong
emission at ∼725 nm.8,10 Hydrophobic fluorophores were
chosen as they would be effectively quenched in an aqueous
environment especially at high loadings on the particle surface.
Two different BF2 chelated azadipyrromethenes 1a, b were
used differing only by the inclusion of a short polyether group
attached to one aryl ring in 1b instead of a methoxy group as in 1a
(Scheme 1). It was anticipated that the polyether substituent may
impart subtle differences in the cellular uptake of the particles.14
Our synthetic approach started with particles NPCOOH which
were converted into their corresponding activated esters NPEster
by reaction with N-hydroxysuccinimide (NHS) and 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDCI) at rt (Scheme 1).
Reaction of NPEster with fluorophores 1a and 1b yielded the cor-
responding nanoparticles NP1a and NP1b respectively (Table 1).
Particles were purified by passing through C-18 reversed phase
silica eluting with water which effectively removed any noncova-
lently linked fluorophore.
Table 1. Characterization of NPCOOH, NP1a, and NP1ba
nano-
particlea
zeta potential
(mV)
particle size/
H2O
particle size/
entry
H2O + TX-100
1
2
3
NPCOOH
NP1a
À62.9
9.4
205 ( 0.5
198 ( 3
199 ( 1
À
217 ( 6
272 ( 19
NP1b
À11.2
a Concentration 0.001% w/w in HPLC grade water.
Table 2. Characterization of NP1a and NP1b
λmax emission (nm)a
entry
NP
toluene
PBS
PBS+TX-100b
FEFc
1
2
NP1a
NP1b
717
718
717
720
723
725
16
11
a Concentration of 0.01 mg/mL. b Solution of TX-100 in PBS
(0.34 mM). c Emission peak area in PBS+TX-100 (0.34 mM)/emission
peak area in PBS.
functionalization with the neutral fluorophores the surface
charge was dramatically altered. NP1a and NP1b gave measured
zeta potentials of +9.4 and À11.2 respectively which indicates
that the majority of surface carboxylic acids had been reacted and
the particle surface had become relatively uncharged (Table 1,
entries 2, 3). Analysis of the nanoparticles emission profiles in
organic solvents such as toluene showed them to be highly
fluorescent with λmax values close to that of their correspond-
ing molecular fluorophores (Table 2, Supporting Informa-
tion (SI)).8 In aqueous media such as phosphate buffered saline
(PBS), as predicted, the fluorescence was dramatically quenched
but with little change in λmax of emission (Figure 3, Table 2).
The zeta potential of the starting nanoparticles NPCOOH was
measured at À62.9 mV (Table 1, entry 1) whereas following
19619
dx.doi.org/10.1021/ja208086e |J. Am. Chem. Soc. 2011, 133, 19618–19621

Journal of the American Chemical Society
COMMUNICATION
Figure 4. Left: Static fluorescence images of MDA-MB-231 cells with
NP1b following 2 h of incubation, washing, and mounting of cells (scale
bar 10 μm). Right: Single cell image; see SI for Z-stack images and
emission spectra taken from points within the cell.
Figure 3. Top: Incremental titration of NP1a, 1b with TX-100 in PBS.
Bottom: Normalized fluorescence intensity difference of NP1a and
NP1b in Dulbecco’s media + 10% FCS (gray), Dulbecco’s media/
TX-100 (red) at particle concentrations of 48, 24, 12, 6, 3, 1.5, 0.75, and
0.375 μg/well (each well contains 300 μL of media).
In contrast, when the nonionic surfactant Triton X-100 (TX-
100) was added the fluorescence intensities were dramatically
increased with a fluorescence enhancement factor (FEF) of 16
and 11 recorded for NP1a and NP1b, respectively (Table 2,
Figure 3, SI). Such high FEF values indicate a very effective
switching mechanism. In contrast, particle equilibration with
bovine serum albumin (0.1 mM) for 2 h gave rise to a relatively
small fluorescence increase (SI).
A similar increased fluorescence response was observed follow-
ing the addition of the anionic surfactant sodium dodecylsulphate
(SDS) and the phospholipid L-α-phosphatidylcholine (SI). The
latter was particularly encouraging as phosphatidylcholines are a
class of phospholipids that are a major component of biological
membranes. Most interestingly, when TX-100 was added incre-
mentally to a PBS solution of NP1a or NP1b the fluorescence
intensity remained relatively unchanged until the critical mi-
celle concentration (CMC) range of TX-100 (0.19À0.30 mM)
(Figure 3).15 This illustrates that it is not a linear fluorescence
response to the TX-100 but rather a response to a particleÀ
surfactant interaction initiating at the CMC.16
In order to probe this further additional characterization of
these systems was undertaken using dynamic light scattering
(DLS). DLS analysis showed that the starting and fluorophore
functionalized particles NPCOOH, NP1a ,and NP1b were mono-
dispersed as individual particles in water giving a measured
diameter size of 205, 198, and 199 nm, respectively (Table 1).
Revealingly, upon the addition of TX-100 to the three particle
types the measured diameter size of NPCOOH did not change, yet
NP1a and NP1b increased to 217 and 272 nm, respectively. This
increase of diameter for the hydrophobic fluorophore function-
alized particles is indicative of a close association of TX-100
around the particle.
Figure 5. Real-time fluorescence imaging of NP1b uptake into HEK293T
cells (scale bar 10 μm) (see SI for movie).
nanoparticles from 48 to 0.375 μg of nanoparticle per well in
both their off and on states were recorded. Encouragingly, it was
possible to obtain plate readable differences at each concentra-
tion level (Figure 3, bottom).
Based on these findings, NP1b was investigated for its
potential as a real time in vitro imaging agent using MDA-
MB-231 (breast cancer), HEK293T (kidney), and CAKI-1 (renal
cancer) cell lines. Our aspiration was that the process of cellular
endocytosis of NP1b would induce the off to on switching of the
particles in a manner observed with surfactants. Initially, a single
time point confocal imaging of NP1b using MDA-MB-231 cells
was investigated. NP1b and cells were coincubated for 2 h, and
the cells were washed, mounted on a glass slide, and imaged.
Encouragingly the cells were highly fluorescent illustrating that
the particles were effectively uptaken and highly fluorescent
within the cells (Figure 4).
Real time imaging was next attempted using HEK293T and
CAKI-1 cells as examples. The fact that extracellular NP1b
particles are virtually fluorescent silent in cell growth media
allowed the particles and cells to be coincubated together without
a masking fluorescent signal.
As it could be readily envisaged that this off to on switching
parameter could be exploited in a diagnostic assay format, the
detection limit of the switching NP1a and NP1b in Dulbecco’s
cell growth media supplemented with 10% FCS (fetal calf serum)
(with and without TX-100) was determined using a 96-well plate
reader. The fluorescence intensities from a serial dilution of
Following addition of particles to the cell culture chamber,
fluorescence images were acquired at short time intervals (to suit
the rate of uptake) and compiled into a real-time movie of the
uptake process. For HEK293T cells, images were acquired every
60 s for 120 min (see SI for movie). In Figure 5 six static frames of
19620
dx.doi.org/10.1021/ja208086e |J. Am. Chem. Soc. 2011, 133, 19618–19621

Journal of the American Chemical Society
COMMUNICATION
’ ACKNOWLEDGMENT
The authors acknowledge financial support from Science
Foundation Ireland and the donation of nanoparticles from Varian
Inc. M.G. thanks IRCSET for funding. Thanks are extended to the
CSCB Mass Spectrometry and NMR Centres and the Conway
Institutes imaging facilities.
’ REFERENCES
(1) Escobedo, J. O.; Rusin, O.; Lim, S.; Strongin, R. M. Curr. Opin.
Chem. Biol. 2010, 14, 64–70.
(2) Ntziachristos, V.; Ripoll, J.; Wang, L. V.; Weissleder, R. Nat.
Biotechnol. 2005, 23, 313–320.
(3) Weissleder, R.; Pittet, M. J. Nature 2008, 452, 580–589.
(4) Frangioni, J. V. J. Clin. Oncol. 2008, 26, 4012–4021.
(5) Razansky, D.; Distel, M.; Vinegoni, C.; Ma, R.; Perrimon, N.;
K€oster, R. W.; Ntziachristos, V. Nature Phot. 2009, 3, 412–417.
(6) (a) White, A. G.; Fu, N.; Leevy, W. M.; Lee, J.-L.; Blasco, M. A.;
Smith, B. D. Bioconjugate Chem. 2010, 21, 1297–1304. (b) Yang, Y.;
Lowry, M.; Xu, X.; Escobedo, J. O.; Sibrian-Vazquez, M.; Wong, L.;
Schowalter, C. M.; Jensen, T. J.; Fronczek, F. R.; Warner, I. M.; Strongin,
R. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 8829–8834.
(7) Trivedi, E. R.; Harney, A. S.; Olive, M. B.; Podgorski, I.; Moin, K.;
Sloane, B. F.; Barrett, A. G. M.; Meade, T. J.; Hoffman, B. M. Proc. Natl.
Acad. Sci. U.S.A. 2010, 107, 1284–1288.
(8) (a) Tasior, M.; O’Shea, D. F. Bioconjugate Chem. 2010,
21, 1130–1133. (b) Gorman, A.; Killoran, J.; O’Shea, C.; Kenna, T.;
Gallagher, W. M.; O’Shea, D. F. J. Am. Chem. Soc. 2004, 126, 10619–10631.
(9) Nguyen, Q. T.; Olson, E.; Aguilera, T. A.; Jiamg, T.; Scadeng, M.;
Ellies, L. G.; Tsien, R. Y. Proc. Natl. Acad. Sci. U.S.A. 2010,
107, 4317–4322.
Figure 6. Real-time fluorescence imaging of NP1b uptake into CAKI-1
cells (scale bar 10 μm) (see SI for movie).
the uptake imaging of HEK293T cells with NP1b are shown in
which it is clear that the fluorescence intensity increases over time
as the particles are uptaken but also that the extracellular particles
are relatively nonfluorescent (dark background). Over the 2 h time
period the particle fluorescence can be clearly seen to pass through
an initial diffuse pattern to concentrate in the perinuclear region.
In the case of CAKI-1 cells, uptake occurred more rapidly so
images were acquired every 20 s for 30 min. In this case a rapid
cellular uptake could again be observed culminating in the
dramatic cell death of individual cells (Figure 6; see SI for movie).
The real-time observation of these cellular events clearly
illustrates the potential of this imaging approach.
(10) Murtagh, J.; Frimannsson, D. O.; O’Shea, D. F. Org. Lett. 2009,
11, 5386–5389.
(11) Stark, W. J. Angew. Chem., Int. Ed. 2011, 50, 1242–1258.
(12) (a) He, X.; Wang, K.; Zhen Cheng, Z. WIREs Nanomed.
Nanobiotechnol. 2010, 2, 349–366. (b) Rungta, P.; Bandera, Y. P.;
Roeder, R. D.; Li, Y.; Baldwin, W. S.; Sharma, D.; Sehorn, M. G.;
Luzinov, I.; Foulger, S. H. Macromol. Biosci. 2011, 11, 927–937.
(13) Solution aggregates of molecular fluorophores are known to pre-
sent different photophysical properties to their deaggregrated molecules
with fluorescence quantum yield often being one of the most pronounced
differences; for examples, see: Chen, Z.; Lohr, A.; Saha-M€oller, C. R.;
W€urthner, F. Chem. Soc. Rev. 2009, 38, 564–584.
In conclusion, a synthetically robust and efficient method for
the surface functionalization of nanoparticles with NIR hydro-
phobic fluorophores has been described. Surface loading of the
hydrophobic fluorophores results in highly efficient fluorescence
quenching in aqueous media which defines a particle off state.
Particle fluorescence can be established in aqueous media in
response to various surfactants. This off/on fluorescence respon-
sive behavior has been extrapolated into a real-time biological
imaging domain in which the cellular uptake of the particles with
inclusion within organelles mimics the off/on switching obtained
with surfactants. The ability of these fluorescence responsive
nanoconstructs to allow real-time continuous imaging of their
uptake into cells has been demonstrated in a movie format.
Further targeted cellular imaging and diagnostic applications are
currently being developed as is the expansion of our concepts
to real-time in vivo imaging. As illustrated above different cell
types show different responses to the same particles. As such a
systematic response study of particle size, concentration, and
surface fluorophore for different cell types is ongoing. The results
from these studies will be reported in due course.
(14) Maldiney, T.; Richard, C.; Seguin, J.; Wattier, N.; Bessodes, M.;
Scherman, D. ACS Nano 2011, 5, 854–862.
(15) Li, M.; Rharbi, Y.; Huang, X.; Winnik, M. A. J. Colloid Interface
Sci. 2000, 230, 135–139.
(16) This does not preclude TX-100Àparticle interactions at con-
centrations lower than the CMC though they have a less significant
impact on particle fluorescence.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental methods, sup-
b
porting figures, and real-time cellular uptake movies. This material
is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
donal.f.oshea@ucd.ie
19621
dx.doi.org/10.1021/ja208086e |J. Am. Chem. Soc. 2011, 133, 19618–19621