Targeted Quantum Dot Conjugates for siRNA Delivery
Austin M. Derfus,†,‡Alice A. Chen,§,3Dal-Hee Min,§Erkki Ruoslahti,‡,#and Sangeeta N. Bhatia*,†,§
Department of Bioengineering, University of California at San Diego, La Jolla, California 92093, Burnham Institute for Medical
Research, La Jolla, California 92037, Health Sciences and Technology/Electrical Engineering and Computer Science,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, and School of Engineering and Applied Sciences,
Harvard University, Cambridge, MA 02138. Received November 25, 2006; Revised Manuscript Received April 3, 2007
Treatment of human diseases such as cancer generally involves the sequential use of diagnostic tools and therapeutic
modalities. Multifunctional platforms combining therapeutic and diagnostic imaging functions in a single vehicle
promise to change this paradigm. in particular, nanoparticle-based multifunctional platforms offer the potential to
improve the pharmacokinetics of drug formulations, while providing attachment sites for diagnostic imaging and
disease targeting features. We have applied these principles to the delivery of small interfering RNA (siRNA)
therapeutics, where systemic delivery is hampered by rapid excretion and nontargeted tissue distribution. Using
a PEGlyated quantum dot (QD) core as a scaffold, siRNA and tumor-homing peptides (F3) were conjugated to
functional groups on the particle’s surface. We found that the homing peptide was required for targeted
internalization by tumor cells, and that siRNA cargo could be coattached without affecting the function of the
peptide. Using an EGFP model system, the role of conjugation chemistry was investigated, with siRNA attached
to the particle by disulfide cross-linkers showing greater silencing efficiency than when attached by a nonreducible
thioether linkage. Since each particle contains a limited number of attachment sites, we further explored the
tradeoff between number of F3 peptides and the number of siRNA per particle, leading to an optimized formulation.
Delivery of these F3/siRNA-QDs to EGFP-transfected HeLa cells and release from their endosomal entrapment
led to significant knockdown of EGFP signal. By designing the siRNA sequence against a therapeutic target
(e.g., oncogene) instead of EGFP, this technology may be ultimately adapted to simultaneously treat and image
The development of multifunctional nanoparticles for treat-
ment of focal disease is attractive for several reasons: they ex-
hibit unique pharmacokinetics including minimal renal filtration,
they have high surface to volume ratios enabling modification
with surface functional groups that can be used to specifically
target the delivery of therapeutic agents to sites of disease, and
they can serve as vehicles for integration of diagnostic imaging
and therapeutic drug delivery, a potentially transformative
clinical paradigm. Use of a nanoparticle imaging core that is
decorated with functional moieties provides a strategy that is
particularly amenable to modular design of a multifunctional
nanoparticle where features may be interchanged or combined
to tailor formulations for a plethora of applications. In an attempt
to move toward this goal, we have previously combined peptides
derived from phage display- a powerful biological screening
technique- with fluorescent semiconductor quantum dots to
target multivalent nanoparticles to tumors (1). In this report,
we further explore the feasibility of incorporating an oligo-
nucleotide-based therapeutic cargo, siRNA. Short, double-
stranded small-interfering-RNAs (siRNA) are one manifestation
of a phenomenon known as RNA interference whereby transla-
tion of a target protein is inhibited. This type of therapeutic
cargo is of particular interest in recent years because it has the
potential to modulate so-called ‘nondruggable’ targets (2, 3).
The untargeted, systemic delivery of siRNA has been
explored by conglomeration of duplexes into nanosized com-
plexes that reduce their renal filtration rates, extending the
circulation half-life well beyond the ∼6 min observed for
unmodified siRNA (4). For example, cholesterol-siRNA con-
jugates bind serum albumin after intravenous injection, forming
long-circulating “natural” nanoparticles (4). Similarly, siRNA-
carrier complexes can be formed ex vivo, prior to injection, by
condensing the nucleic acid with a cationic protein (e.g.,
protamine (5)) or polymer (e.g., poly(ethylene imine) (PEI) (6),
cyclodextrin-containing polycations (7), or PEG-based block
catiomer (8)). In addition, targeted delivery of such agents has
the potential to limit collateral toxicity and ‘off-target’ effects.
Targeting has been explored through use of through the
attachment of antibodies (5), small molecules (e.g., transferrin
(7)), aptamers (9, 10), or well-established peptide ligands (6));
however, these approaches are typically not modular nor
multifunctional (i.e., do not incorporate imaging moieties).
Addition of a nanoparticle-based imaging agent to siRNA
delivery strategies may be particularly advantageous as protein
knockdown by RNAi is delayed (>48 h or more after
administration), and many fluorescent dyes are not stable for
monitoring delivery over extended periods of time in vivo. In
vitro, co-delivery of fluorescent reporter plasmid along with the
siRNA is often utilized; however, this is unlikely to be used
clinically given the potential risks associated with integration
into host DNA.
Quantum dots offer the potential to serve as photostable
beacons to track siRNA delivery. We have previously explored
their utility for monitoring siRNA delivery in vitro by co-
complexing QDs with cationic liposomes and siRNA (11, 12);
however, this approach is not amenable to either systemic
delivery, because of their relatively large size and rapid uptake
* Author to whom correspondence should be addressed. Phone:
(617) 324-0221, fax: (617) 324-0740, e-mail: firstname.lastname@example.org.
†University of California at San Diego.
‡Burnham Institute for Medical Research.
#Burnham Institute for Medical Research, University of California,
Santa Barbara, Santa Barbara, CA 93106-9610.
Bioconjugate Chem. 2007, 18, 1391−1396
10.1021/bc060367e CCC: $37.00© 2007 American Chemical Society
Published on Web 07/14/2007
by macrophages, or targeted delivery, as they are ubiquitously
internalized. In this report, we sought to generate multifunctional
conjugates of a size compatible with eventual systemic delivery
(5< d < 200 nm) upon a quantum dot core that presented both
targeting peptides and siRNA (Figure 1). A QD core with
emission in the near-infrared (NIR) was chosen to allow greater
imaging depth in tissue than that achieved with visible wave-
lengths. The siRNA was attached to the surface with a chemical
cross-linker, rather than the electrostatic interactions used in
many schemes. This strategy grants a smaller overall size than
polymer-based particles, which may carry advantages in evading
uptake by the reticuloendothelial system (13). As a proof-of-
concept, the attached siRNA was designed against the enhanced
green fluorescent protein (EGFP) gene, and complexes were
delivered to EGFP-transfected HeLa cells in vitro. A tumor-
homing peptide (F3), targeting cell-surface nucleolin (14), was
attached to achieve cell uptake. F3 peptide has been shown to
target tumor cells in vitro and in vivo when administered as a
free peptide and also when coconjugated with poly(ethylene
glycol) to quantum dots delivered systemically in mice bearing
Combining these components onto a single particle brings
several engineering requirements and constraints that affect their
design and fabrication. Accordingly, in this report we first
investigate whether the attachment of the peptide is necessary
and sufficient to achieve cell internalization. As particle surface
area is limited (∼100 surface amines on these QDs), the actual
copy numbers of targeting peptide and siRNA required are also
quantified. Increasing the number of peptides on the QDs lowers
the siRNA dose per particle and vice versa. Next, because
conjugation of siRNA to the particles must be stable in cell
media yet also allow the duplex to interact with the RNA-
induced silencing complex (RISC) in the cytosol, we test the
efficacy of two different cross-linkers for this task. While RISC
binding may be possible with the duplex attached to the particle,
release of the siRNA from the particle’s surface upon internal-
ization proved advantageous to knockdown efficacy. Finally,
cytoplasmic delivery, free of endosomes, is essential to allow
siRNA to interact with cellular silencing machinery. While
fluorophore-labeled F3 peptide has been shown to reach the
nuclei of tumor cells (15), internalization of particles generally
follows the endolysosomal pathway. In this report, the addition
of endosome-disrupting agents was required to aid escape and
increase siRNA-mediated knockdown.
Materials. Quantum dots with emission maxima of 655 or
705 nm and modified with PEG and amino groups were obtained
from Quantum Dot Corporation (ITK amino). QD concentrations
were measured by optical absorbance at 595 nm, using extinc-
tion coefficients provided by the supplier. Cross-linkers used
were sulfo-LC-SPDP (sulfosuccinimidyl 6-(3′-[2-pyridyldithio]-
propionamido)hexanoate) (Pierce) and sulfo-SMCC (sulfosuc-
(Sigma). Synthetic RNA duplexes directed against the EGFP
mRNA were synthesized, with the sense strand modified to
contain a 5′ thiol group (Dharmacon) (sense: 5′-Th-(CH2)6-
GGC UAC GUC CAG GAG CGC ACC; antisense: 5′-UGC
GCU CCU GGA CGU AGC CUU). The F3 peptide was
synthesized with an aminohexanoic acid (Ahx) spacer and
cysteine residue added for conjugation (final sequence: C[Ahx]-
AKVK DEPQR RSARL SAKPA PPKPE PKPKK APAKK).
A FITC-labeled F3 peptide was also synthesized, along with
KAREC (Lys-Ala-Arg-Glu-Cys), a five amino acid control
peptide. All peptides were synthesized by N-(9-fluorenyl-
methoxycarbonyl)-L-amino acid chemistry with a solid-phase
synthesizer and purified by HPLC. The composition of the
peptides was confirmed by MS.
Conjugation of Peptides and Nucleic Acid to QDs. Amino-
modified QDs were conjugated to thiol-containing siRNA and
peptides using sulfo-LC-SPDP and sulfo-SMCC cross-linkers.
QDs were resuspended in 50 mM sodium phosphate, 150 mM
sodium chloride, pH 7.2, using Amicon Ultra-4 (100 kDa cutoff)
filters. Cross-linker (1000-fold excess) was added to QDs and
allowed to react for 1 h. Samples were filtered on a NAP-5
gravity column (to remove excess cross-linker) into similar
buffer supplemented with 10 mM EDTA. siRNA was treated
with 0.1 M DTT for 1 h and filtered on a NAP-5 column into
EDTA-containing buffer. Peptides were typically used from
lyophilized powder. Peptide and/or siRNA was added to filtered
QDs and allowed to react overnight at 4 °C. Using three Amicon
filters, product was filtered twice with Dulbecco’s phosphate-
buffered saline (PBS), twice with a high salt buffer (1.0 M
sodium chloride, 100 mM sodium citrate, pH 7.2), and twice
again with PBS. High salt washes were required to remove
electrostatically bound siRNA and peptide, which was not
removed with PBS washes alone.
For siRNA-QDs, a 10-fold excess of siRNA was typically
used for both cross-linkers. In the case of sulfo-LC-SPDP, the
amount of conjugated siRNA was assayed using gel electro-
Figure 1. Design of a multifunctional nanoparticle for siRNA delivery. Because of their photostable fluorescence and multivalency, QDs are
suitable vehicles for ferrying siRNA into live cells in vitro and in vivo. Conjugation of homing peptides (along with the siRNA cargo) to the QD
surface allows targeted internalization in tumor cells. Once internalized, these particles must escape the endolysomal pathway and reach the cytoplasm
to interact with the RNA-induced silencing complex (RISC), which leads to degradation of mRNA homologous to the siRNA sequence.
Bioconjugate Chem., Vol. 18, No. 5, 2007 Derfus et al.
phoresis (20% TBE gel, Invitrogen), staining with SYBR Gold
(Invitrogen). To confirm that similar amounts of siRNA
(approximately two per QD) were conjugated to QDs using
sulfo-SMCC, particles were stained with SYBR Gold and
measured with a fluorimeter (SpectraMax Gemini XS, Molecular
For F3/siRNA-QDs and KAREC/siRNA-QDs, a molar ratio
of 15:70:1 (siRNA:peptide:QDs) was found to be optimum,
though a variety of ratios were attempted (Figure 4A). These
conditions yielded approximately 20 F3 peptides and 1 siRNA
duplex per particle.
Cell Culture. Internalization and knockdown experiments
were performed using a HeLa cell line transfected with
destabilized EGFP with a half-life of 1 h (courtesy of Phillip
Sharp, MIT). Growth media was Dulbecco’s modified Eagle’s
medium (DMEM) containing 4.5 g/L glucose and supplemented
with 10% FBS, 100 units/mL penicillin, 100 ug/mL strepto-
mycin, and 292 ug/mL L-glutamine. Cells were passaged into
24-well plates and used at 50-80% confluency for internaliza-
tion experiments and 20-40% confluency for knockdown
For internalization experiments (Figure 2), QDs were added
to cell monolayers in media without serum at a final concentra-
tion of 50 nM. After 4 h, cells were washed with media, treated
with trypsin (0.25%) and EDTA, and resuspended in 1% BSA
(in PBS) for flow cytometry (BD FACSort, FL1 for EGFP signal
and FL3 for QD signal). Fluorescence data on 10 000 cells was
collected for each sample, and the geometric mean of intensity
For knockdown experiments in Figure 3, siRNA-QDs (in
50 uL serum/antibiotic-free media) were added to Lipofectamine
2000 (1 µL in 50 µL media, Invitrogen) and allowed to complex
for 20 min. Cell media was changed to 400 µL of serum/
antibiotic-free per well, and QD solutions (100 µL) were added
dropwise. Complete media was added 12-18 h later, and 48 h
after the QD were added, cells were trypsinized and assayed
for fluorescence by flow cytometry.
To assess EGFP knockdown, 50 nM or 10 nM concentrations
of F3/siRNA-QDs or KAREC/siRNA-QDs were added to cell
monolayers (20-40% confluent) in media with serum/antibiot-
ics. To demonstrate specificity, irrelevant siRNA (designed
against the Lamin A/C gene, as described in ref 12) was
conjugated to QDs along with F3 and also delivered to cells.
Four hours later, cells were washed with similar media. Some
samples were then treated with 1 µL of Lipofectamine per well
(added dropwise in 100 µL media) either immediately after
washing or after a 90 min incubation at 37 °C (to allow
membrane recycling). For all samples, media was changed to
complete DMEM with serum/antibiotics ∼16 h after the addition
of QDs and assayed by flow cytometry 48 h from the start of
the experiment. For imaging, cells were initially seeded on glass-
bottom dishes (Mat-Tek) and observed 48 h after the addition
of QDs using a 60× oil immersion objective. Images were
captured with a SPOT camera mounted on a Nikon TE200
inverted epifluorescence microscope.
RESULTS AND DISCUSSION
Taking a modular approach, particle internalization and
siRNA attachment were investigated separately before these
functions were combined in a single particle. First, peptides were
conjugated to QDs to improve tumor cell uptake. Addition of
as-purchased PEGlyated QDs to HeLa cell monolayers led to
minimal cell uptake, as quantified with flow cytometry (Figure
2A). Conjugation of siRNA or a control pentapeptide (KAREC)
did not increase QD internalization, but addition of F3 peptide
to the QDs improved the uptake significantly (2 orders of
magnitude). To confirm the specificity of F3 uptake, free F3
peptide was added to cells along with 50 nM F3-QDs (Figure
2B). Dose-dependent inhibition of uptake was observed with
F3 peptide concentrations from 1 µM to 1 mM. Inhibition of
uptake by an irrelevant peptide, free KAREC, was minimal by
comparison. The large excess of free peptide required for
inhibition may be due to multiple copies of the F3 peptide on
each QD and improved receptor binding as a result of multi-
To quantify the number of peptides added per particle, FITC-
labeled F3 peptide was synthesized and attached to QDs using
Figure 2. Attachment of F3 peptide leads to QD internalization in
HeLa cells. Thiolated peptides (F3 and KAREC control) and siRNA
were conjugated to PEG-amino QD705 particles using sulfo-SMCC.
Particles were filtered to remove excess peptide or siRNA and incubated
with HeLa cell monolayers for 4 h. Flow cytometry indicated the F3
peptide is required for cell entry (A). The addition of free F3 peptide
inhibits F3-QD uptake, while KAREC peptide does not, suggesting
the F3 peptide and F3-labeled particles target the same receptor (B).
In part C, the relationship between the number of F3 peptides per QD
and cell uptake was examined. In these experiments, FITC-labeled
peptide was conjugated to QDs using sulfo-LC-SPDP. For each
formulation (black circles), peptide:QD ratio was determined by
measuring the QD concentration by absorbance and then treating the
conjugate with 2-mercaptoethanol, filtering out the QDs, and measuring
the FITC fluorescence. Cell uptake increases dramatically with peptide
number but appears to saturate around 10-15 F3s per QD.
Targeted Quantum Dot Conjugates for siRNA DeliveryBioconjugate Chem., Vol. 18, No. 5, 2007
a cleavable cross-linker (sulfo-LC-SPDP). After unreacted
peptide was removed by filtration, 2-mercaptoethanol (2-ME)
was added to reduce the disulfide bond between peptide and
QD. Using a 100 kDa cutoff filter, F3-FITC peptide was
separated from the QDs and quantified by fluorescence. Several
reactions were performed with various amounts of FITC-F3 and
siRNA as reactants. For each formulation, the cellular uptake
was quantified by flow cytometry and F3 number measured
(Figure 2C, each point indicates a separate formulation).
Attachment of a small number of peptides (0-5) did not lead
to significant uptake (less than 10% of maximum). Uptake
increases with peptide number but begins to saturate around 15
copies per QD.
The use of cleavable (sulfo-LC-SPDP) or noncleavable (sulfo-
SMCC) cross-linkers for the attachment of F3 peptide did not
significantly affect cell uptake. The choice of cross-linker,
however, may affect the ability of the siRNA cargo to interact
with RISC. The interior of the cell is a reducing environment,
which would lead to cleavage of the disulfide bond generated
by sulfo-LC-SPDP, freeing the siRNA. On the other hand, the
amide bond produced by sulfo-SMCC is unaffected by reducing
conditions (confirmed by treating the conjugates with 2.5%
2-ME for 30 min), leaving the intracellular QD/siRNA conjugate
intact. We compared the efficiency of QD/siRNA conjugates
prepared with both cross-linkers using an EGFP model system.
Delivery of the conjugates to EGFP-labeled HeLa cells was
performed by first complexing the particles with a cationic
liposome transfection reagent (Lipofectamine 2000), to satisfy
the functions of cell internalization and endosome escape, and
knockdown efficiency was quantified by a reduction in EGFP
fluorescence over controls (Lipofectamine only).
Using gel electrophoresis, the amount of siRNA conjugated
per particle was quantified relative to double-stranded RNA
standards. Particles conjugated using sulfo-LC-SPDP were first
introduced under native (nonreduced) conditions (Figure 3B).
The absence of a siRNA band in the QD/siRNA lanes indicates
that no siRNA is noncovalently bound to the particles. Exposing
the particles to 2-ME for 30 min led to the appearance of a
siRNA band in the SPDP lane, which was quantified with RNA
standards and ImageQuant software. Using this approach,
approximately two siRNA duplexes were conjugated per QD
under these conditions (Figure 3C). Cellular fluorescence was
quantified 48 h after incubation with HeLa cells using flow
cytometry. As hypothesized, the QD/siRNA formulation pro-
duced with the disulfide bond (using sulfo-LC-SPDP) led to
greater EGFP knockdown (Figure 3D). The level of knockdown
attained with disulfide-linked QD/siRNA, however, was less
than observed when an equal concentration of free siRNA was
delivered with Lipofectamine. On the basis of previous observa-
tions(12), one potential cause may be the limited surface area
of the cationic liposomes, which is shared by QDs and siRNA.
Additionally, the presence of the QDs in the endosome may
reduce the efficiency of escape, or reduction of the disulfide
bond may be incomplete, resulting in less efficient complexation
The F3:siRNA reaction ratio was varied with the goal of
generating a formulation capable of high cell uptake as well as
the ability to carry a significant payload of siRNA. The cleavable
cross-linker allowed the removal and quantification of both
species after F3 peptide and siRNA coattachment. The results
indicate a tradeoff between one siRNA per particle with high
uptake (>15 peptides) and two duplexes but lower uptake (<10
peptides) (Figure 4A). Negatively charged siRNA may be
electrostatically adsorbing to the surface of the aminated QDs,
preventing the attachment of additional F3 peptides. Potentially,
performing the reaction in high salt conditions, or in the presence
of a surfactant, may allow higher loading. Since both high uptake
efficiency and siRNA number are required for knockdown,
particles with ∼20 F3s and a single siRNA duplex were further
investigated in cell studies.
When incubated with cells, these 20-F3/1-siRNA-QDs were
shown to internalize significantly but did not lead to reduction
in EGFP fluorescence 48 h later. Fluorescence microscopy
revealed that the particles were intracellular, but they colocalized
with an endosomal marker (LysoSensor, Molecular Probes).
Figure 3. Conjugation of siRNA to QDs with cleavable or noncleav-
able cross-linkers. Thiol-modified siRNA was attached to PEG-amino
QDs using the water-soluble heterobifunctional cross-linkers sulfo-
SMCC and sulfo-LC-SPDP (A). The cross-link produced by SPDP is
cleavable with 2-mercaptoethanol (2-ME), while SMCC yields a
nonreducible linkage. Gel electrophoresis of the disulfide-linked
conjugates indicated that no siRNA are electrostatically bound to the
conjugate (B). Upon treatment with 2-ME, the QD/siRNA cross-link
is reduced and the siRNA migrated down the gel alongside siRNA
standards (C). QD/siRNA conjugates (or siRNA alone) were delivered
to EGFP-expressing HeLa cells using Lipofectamine 2000 (cationic
liposome reagent). Cells were trypsinized and assayed by flow
cytometry 48 h later. Comparison with control cells (treated with
Lipofectamine alone) indicated the disulfide bond leads to superior
EGFP knockdown (% reduction in geometric mean fluorescence) (D).
Comparing a dot-plot of cells treated with Lipofectamine alone (E) or
disulfide-linked QD/siRNA (F) revealed a negative correlation between
QD uptake and EGFP signal. Thus, the QD label can serve as a means
of quantifying siRNA delivery and thus knockdown.
Bioconjugate Chem., Vol. 18, No. 5, 2007 Derfus et al.
Addition of an endosome escape agent, therefore, was required
to achieve knockdown. Specifically, after incubation of cells
with F3/siRNA-QDs and washing, cationic liposomes were
added for 12 h. Although cationic liposomes and polymers are
typically used to form complexes with nucleic acids or particles,
thereby ferrying the payload inside cells, in this case the reagent
leads to endosomal escape of previously internalized QDs. We
hypothesize that the cationic liposomes are internalized into new
endosomes, which fuse with the endosomes carrying the QDs.
As the pH of the vesicle is lowered by the cell, osmotic lysis
leads to the release of both species into the cytoplasm. To assess
the importance of the targeting ligand, particles carrying siRNA
and a control peptide (KAREC) were used. These KAREC/
siRNA particles were not internalized, and no EGFP knockdown
was observed, despite endosome disruption. The specificity of
knockdown was verified by delivery of F3/siRNA QDs contain-
ing an irrelevant siRNA (directed against Lamin A/C). Ad-
ditionally, a time lag of 90 min between washing the cells free
of QDs and cationic liposome addition did not lead to significant
reduction in efficiency, indicating that endosomal degradation
of the siRNA is not an issue on this time scale.
In addition to cationic liposomes, some chemotherapeutics,
such as chloroquine have been shown to be capable of
endosomal escape (16). While an endosome escape step could
be a realistic part of a treatment regimen, there is also potential
that this function could be built into each particle. Addition of
fusogenic moieties to the QD surface, for example, may further
improve delivery of the multifunctional particles described (17).
Decorating the surface of a fluorescent quantum dot with both
a targeting ligand and siRNA duplex requires a tradeoff in the
number of each species but can be used to generate a conjugate
capable of knockdown in vitro. We found that multiple copies
of the F3 targeting peptide were required for QD uptake, but
that siRNA cargo could be co-attached without affecting the
function of the peptide. Disulfide (sulfo-LC-SPDP) and covalent
(sulfo-SMCC) cross-linkers were investigated for the attachment
of siRNA to the particle, with the disulfide bond showing greater
silencing efficiency. Finally, after delivery to cells and release
from their endosomal entrapment, F3/siRNA-QDs produced
significant knockdown of EGFP signal. By designing the siRNA
sequence against a therapeutic target (e.g., oncogene) instead
of EGFP, the technology explored in this study may be adapted
to treat and image diseases such as cancer. This technology
could also be readily adapted to other nanoparticle platforms,
such as iron oxide or gold cores, which allow image contrast
on magnetic resonance or X-ray imaging, respectively, and may
therefore mitigate concerns over QD cytotoxicity and the limited
tissue penetration of light. QDs, however, remain an attractive
tool for in vitro and animal testing, where fluorescence is the
most accessible and common imaging modality.
We thank Fernando Ferrer for peptide synthesis, Phillip Sharp
for HeLa cell line and siRNA sequence, Joel Nielson for
assistance with siRNA reagents and procedures, and we
acknowledge financial support of the David and Lucile Packard
Foundation, the National Cancer Institute of the National
Institutes of Health (contract no. N01-C0-37117), the Centers
and MIT (U54 CA119349-01), and a Bioengineering Research
Partnership grant (1 R01 CA124427-01). A.M.D. thanks the UC
Biotechnology Research and Education Program for a G.R.E.A.T.
fellowship (no. 2004-16).
Figure 4. Co-attachment of F3 peptide and siRNA cargo allows EGFP
knockdown upon delivery and endosome escape. Because of a limited
number of attachment sites on the QDs, the goal of coattachment was
to maximize siRNA loading while conjugating sufficient F3 peptides
to allow internalization (>15). Varying the F3:siRNA ratio resulted in
a number of QD formulations (black circles, A). We chose a formulation
containing ∼20 F3 peptides and ∼1 siRNA per QD for knockdown
studies in cell monolayers (B-D). EGFP-expressing HeLa cells were
treated with 50 nM F3/siRNA-QDs for 4 h and then washed with cell
media. When assayed for green fluorescence 48 h later, no knockdown
was observed (“control”, B). When these cells were treated with cationic
liposomes (Lipofectamine 2000) immediately after removing the QDs
and washing, a ∼29% reduction in EGFP was observed. A lower
concentration of QDs (10 nM) is less effective (21% knockdown).
Incubation with KAREC-labeled particles followed by cationic lipo-
somes leads to minimal particle internalization and thus no knockdown.
Attachment of an irrelevant siRNA (directed against Lamin A/C) to
QDs along with F3 peptide also results in no knockdown when delivered
to cells followed by lipofectamine. Fluorescence imaging of cells
incubated with F3/siRNA QDs (red dots) showed a reduced green
fluorescence (D), compared with control cells incubated with Lipo-
fectamine alone (C).
Targeted Quantum Dot Conjugates for siRNA Delivery Bioconjugate Chem., Vol. 18, No. 5, 2007
(1) Akerman, M. E., Chan, W. C., Laakkonen, P., Bhatia, S. N., and
Ruoslahti, E. (2002) Nanocrystal targeting in vivo. Proc. Natl. Acad.
Sci. U.S.A. 99, 12617-21.
(2) Dykxhoorn, D. M., Palliser, D., and Lieberman, J. (2006) The silent
treatment: siRNAs as small molecule drugs. Gene Ther. 13, 541-
(3) Uprichard, S. L. (2005) The therapeutic potential of RNA interfer-
ence. FEBS Lett. 579, 5996-6007.
(4) Soutschek, J., Akinc, A., Bramlage, B., Charisse, K., Constien, R.,
Donoghue, M., Elbashir, S., Geick, A., Hadwiger, P., Harborth, J.,
John, M., Kesavan, V., Lavine, G., Pandey, R. K., Racie, T., Rajeev,
K. G., Rohl, I., Toudjarska, I., Wang, G., Wuschko, S., Bumcrot,
D., Koteliansky, V., Limmer, S., Manoharan, M., and Vornlocher,
H. P. (2004) Therapeutic silencing of an endogenous gene by
systemic administration of modified siRNAs. Nature 432, 173-8.
(5) Song, E., Zhu, P., Lee, S. K., Chowdhury, D., Kussman, S.,
Dykxhoorn, D. M., Feng, Y., Palliser, D., Weiner, D. B., Shankar,
P., Marasco, W. A., and Lieberman, J. (2005) Antibody mediated
in vivo delivery of small interfering RNAs via cell-surface receptors.
Nat. Biotechnol. 23, 709-17.
(6) Schiffelers, R. M., Ansari, A., Xu, J., Zhou, Q., Tang, Q., Storm,
G., Molema, G., Lu, P. Y., Scaria, P. V., and Woodle, M. C. (2004)
Cancer siRNA therapy by tumor selective delivery with ligand-
targeted sterically stabilized nanoparticle. Nucleic Acids Res. 32,
(7) Hu-Lieskovan, S., Heidel, J. D., Bartlett, D. W., Davis, M. E., and
Triche, T. J. (2005) Sequence-specific knockdown of EWS-FLI1
by targeted, nonviral delivery of small interfering RNA inhibits tumor
growth in a murine model of metastatic Ewing’s sarcoma. Cancer
Res. 65, 8984-92.
(8) Itaka, K., Kanayama, N., Nishiyama, N., Jang, W. D., Yamasaki,
Y., Nakamura, K., Kawaguchi, H., and Kataoka, K. (2004) Su-
pramolecular nanocarrier of siRNA from PEG-based block catiomer
carrying diamine side chain with distinctive pKadirected to enhance
intracellular gene silencing. J. Am. Chem. Soc. 126, 13612-3.
(9) Chu, T. C., Twu, K. Y., Ellington, A. D., and Levy, M. (2006)
Aptamer mediated siRNA delivery. Nucleic Acids Res. 34, e73.
(10) McNamara, J. O., 2nd, Andrechek, E. R., Wang, Y., Viles, K.
D., Rempel, R. E., Gilboa, E., Sullenger, B. A., and Giangrande, P.
H. (2006) Cell type-specific delivery of siRNAs with aptamer-siRNA
chimeras. Nat. Biotechnol. 24, 1005-15.
(11) Derfus, A. M., Chan, W. C. W., and Bhatia, S. N. (2004)
Intracellular Delivery of Quantum Dots for Live Cell Labeling and
Organelle Tracking. AdV. Mater 16, 961-966.
(12) Chen, A. A., Derfus, A. M., Khetani, S. R., and Bhatia, S. N.
(2005) Quantum dots to monitor RNAi delivery and improve gene
silencing. Nucleic Acids Res. 33, e190.
(13) Moghimi, S. M., Hunter, A. C., and Murray, J. C. (2001) Long-
circulating and target-specific nanoparticles: theory to practice.
Pharmacol. ReV. 53, 283-318.
(14) Christian, S., Pilch, J., Akerman, M. E., Porkka, K., Laakkonen,
P., and Ruoslahti, E. (2003) Nucleolin expressed at the cell surface
is a marker of endothelial cells in angiogenic blood vessels. J. Cell
Biol. 163, 871-8.
(15) Porkka, K., Laakkonen, P., Hoffman, J. A., Bernasconi, M., and
Ruoslahti, E. (2002) A fragment of the HMGN2 protein homes to
the nuclei of tumor cells and tumor endothelial cells in vivo. Proc.
Natl. Acad. Sci. U.S.A. 99, 7444-9.
(16) Won, J., Kim, M., Yi, Y. W., Kim, Y. H., Jung, N., and Kim, T.
K. (2005) A magnetic nanoprobe technology for detecting molecular
interactions in live cells. Science 309, 121-5.
(17) Plank, C., Oberhauser, B., Mechtler, K., Koch, C., and Wagner,
E. (1994) The influence of endosome-disruptive peptides on gene
transfer using synthetic virus-like gene transfer systems. J. Biol.
Chem. 269, 12918-12924.
Bioconjugate Chem., Vol. 18, No. 5, 2007Derfus et al.