A chemical approach for cell-specific targeting of nanomaterials: Small-molecule-initiated misfolding of nanoparticle corona proteins

Article abstract of DOI:10.1021/ja300537u

A major challenge in nanomaterial science is to develop approaches that ensure that when administered in vivo, nanoparticles can be targeted to their requisite site of action. Herein we report the first approach that allows for cell-specific uptake of nan

Full text of DOI:10.1021/ja300537u

Communication
pubs.acs.org/JACS
A Chemical Approach for Cell-Specific Targeting of Nanomaterials:
Small-Molecule-Initiated Misfolding of Nanoparticle Corona Proteins
Kanlaya Prapainop,† Daniel P. Witter,† and Paul Wentworth, Jr.*,†,‡
†Department of Biochemistry, The University of Oxford, South Parks Road, Oxford OX1 3QU, U.K.
‡Department of Chemistry and the Skaggs Institute for Chemical Biology, The Scripps Research Institute, 10550 North Torrey Pines
Road, La Jolla, California 92037, United States
S
* Supporting Information
In this work, we sought to exploit the nanoparticle/protein
ABSTRACT: A major challenge in nanomaterial science
is to develop approaches that ensure that when
administered in vivo, nanoparticles can be targeted to
their requisite site of action. Herein we report the first
approach that allows for cell-specific uptake of nanoma-
terials by a process involving reprogramming of the
behavior of the ubiquitous protein corona of nanomateri-
als. Specifically, judicious surface modification of quantum
dots with a small molecule that induces a protein-
misfolding event in a component of the nanoparticle-
associated protein corona renders the associated nanoma-
terials susceptible to cell-specific, receptor-mediated
endocytosis. We see this chemical approach as a new
and general method for exploiting the inescapable protein
corona to target nanomaterials to specific cells.
corona association in a positive way and use it to target specific
cell types in a controlled manner. Thus, we chemically
decorated the surface of CdSe/ZnS quantum dots (QDs)
with the inflammatory metabolite cholesterol 5,6-secosterol
atheronal-B (1a) and showed that the resulting nanoparticles
(QD585-ath-B, 2a) bind to and induce the misfolding/
aggregation of apolipoprotein B (apo-B100) in low density
lipoprotein (LDL) and are taken up into cultured macrophages
in a cell- and receptor-specific manner (Figure 1).
he past decade has witnessed an explosion in the research
Ton inorganic nanoscale materials for in vivo and in vitro
applications as diagnostics and therapeutics. With this research,
there is a growing appreciation that the nature of the
interactions between components of the biological milieu and
the nanoparticle depends upon the structural composition of
the xenobiotic and is mediated by surface-bound proteins and
lipids, the so-called “protein corona”. Recent work has shown
that all nanoparticles in biological fluids are coated with a
protein/lipid mel
́
ange termed the protein corona.1 Careful
studies have revealed that the composition of this corona is
dynamic,2 reflects the size, shape, and surface properties of the
nanoparticles, and is a major determinant of the localization of
the nanomaterials in vivo.3 Therefore, a current challenge in
nanomedicine is to develop approaches to ensure that
nanoparticles arrive at their site of action in biological systems
with their contents intact while at the same time circumventing
the effects of the protein corona. The sobering implication of
the protein corona is that the in vivo localization of the
nanoparticle at the whole-body, organ-system, tissue, cell, and
organelle levels ultimately depends not on the nanoparticles’
bulk composition or payload but rather on the identity,
organization, and residence time of the proteins at the particle
surface. A recent study has shown that negatively charged
poly(acrylic acid)-coated gold nanoparticles induce unfolding
of bound fibrinogen, promoting the interaction with the
integrin receptor MAC-1 and triggering the unwanted
inflammatory response common to certain nanomaterials.4
Figure 1. Oxysterols 1a and 1b and nanomaterials 2a and 2b
synthesized for a general approach to cell targeting of nanomaterials.
The core science that underpins our approach to exploit the
protein corona stems from our recent discovery of a new class
of oxysterol inflammatory metabolites in vivo, termed the
atheronals.5 What makes the atheronals, such as 1a, of
importance in the context of this study is that we have
shown them to perturb the misfolding/aggregation of a number
of biologically relevant proteins, such as apo-B100,5 β-amyloid,6
α-synuclein,7 antibody light chains,8 a murine prion protein9
and myelin basic protein. Furthermore, the adduction of 1a to
Received: January 17, 2012
Published: February 17, 2012
© 2012 American Chemical Society
4100
dx.doi.org/10.1021/ja300537u | J. Am. Chem. Soc. 2012, 134, 4100−4103

Journal of the American Chemical Society
Communication
apo-B100, the protein component of LDL, causes a conforma-
tional change in the protein that exposes epitopes for
macrophage cell-surface receptors that ultimately leads to
uptake of atheronal-modified LDL particles into cells and foam
cell formation.5,10
The nanoparticles selected as a model for this study were
fluorescent inorganic CdSe/ZnS QDs possessing an amino-
functionalized poly(ethylene glycol) hydrophilic surface,
designated as QD585-NH2 (2b). Q585-ath-B (2a) was
prepared by coupling atheronal-B analogue 1b (synthesized
in three steps from 5-hydroxycholenic acid) to 2b (∼100 amino
groups per particle) using EDC and sulfo-NHS in borate buffer
as previously described.11 The photophysical properties
(absorbance and emission spectra) of 2a in borate buffer (pH
8.0) were essentially identical to those of 2b (Figure S1a in the
Supporting Information). Fluorescamine analysis revealed that
∼50 amino groups remained on each 2a particle after coupling
with 1b, suggesting that ∼50 equiv of 1b were present on the
surface of each nanoparticle. Dynamic light scattering (DLS)
confirmed the hydrodynamic radius of 2a (8.6 0.7 nm) to be
slightly greater than that of the starting nanomaterial 2b (8.0
0.5 nm) and importantly, along with transmission electron
microscopy (TEM), also showed that the more hydrophobic
nanomaterial 2a did not measurably aggregate in aqueous
buffers when stored for 3 months at 4 °C (Figures S1b−d).
The increase in surface hydrophobicity of the atheronal-coated
QDs was further confirmed by measurements of the ζ potential,
which was considerably less positive for 2a than 2b (ζ = +2.66
0.35 vs ζ = +3.51 0.28, respectively). A recent shotgun
proteomic study by our group established that when incubated
in human plasma, QD585-ath-B binds apo-B100 and other
proteins of the LDL proteome in the so-called “hard corona”.11
In that study, the biological milieu was a cell culture medium,
and therefore, the QD585-ath-B nanoparticles were incubated
in cell culture medium (RPMI) plus 1% fetal calf serum (FCS),
and “hard corona” protein components were resolved on a Tris
acetate gel (3−8%) (Figure S2, lane 4). In-gel trypsin digestion
and mass spectrometry analysis confirmed that apo-B100 (∼550
kDa) from bovine LDL in culture medium was indeed bound
to nanoparticles 2a.
Figure 2. QD585-ath-B (2a) nanoparticles induce aggregation of apo-
B100 in LDL, as represented by the scheme showing several of the
plausible equilibria (a−f) established between native apo-B100 (two
green domains), misfolded apoB100 (one green and one red domain),
and nanoparticle 2a. The key equilibria for turbidity are c and f, which
release misfolded apo-B100 from the nanoparticle into bulk solution.
The inset shows the time course of LDL (40 μg/mL) misfolding, as
measured by turbidity (OD at 400 nm), during incubation with 2a
(200 nM) in PBS (pH 7.4) at 37 °C.
either diffuse into the bulk solvent (equilibria c and f) to seed
aggregation in solution or stay bound in the corona for either
further binding to LDL and/or seeding aggregation on the
nanoparticle.
We first studied the uptake of QD585-ath-B (2a) into
cultured macrophage cells (raw 264.7) using both fluorescence
microscopy and flow cytometry (FCM) (Figure 3a,d,g and
Figures S3 and S4). In these studies, we used QD585-NH2
(2b) as a control nanomaterial to allow a fair assessment of the
effect of surface modification with atheronal 1b. These initial
cell studies revealed concentration- and time-dependent
macrophage uptake, with the first QDs appearing in cells
after ∼30 min and continuing to be taken up even after 4 h of
incubation and with measurable uptake of 2a occurring at 10
nM (Figure 3a,d,g and Figure S4A,B). To confirm that binding
of LDL-coated particles alone was not a sufficient trigger for
uptake into macrophages, we performed cell-uptake studies
with 2b nanoparticles (100 nM), which also bind LDL in its
protein corona (Figure S2, lane 2). This control study showed
no macrophage uptake of 2b (100 nM) after 2 h, whereas
under identical conditions 2a (100 nM) was clearly observed
within the cells (Figure S5). Cell viability was not affected by
treatment with either 2a or 2b in these studies (Figure S6).
A comparison between uptake of 2a (100 nM) into
macrophages incubated for 2 h at 37 °C in RPMI medium
supplemented with either FCS (1%, contains lipoproteins),
lipoprotein-deficient serum (LPDS, 1%), or delipidated LPDS
(1%) revealed a clear requirement of LDL for cell uptake
(Figure 3a−c and Figure S7). FCM analysis quantified the
difference (measured as the mean fluorescence intensity) as
being 2-fold more QDs taken up for the medium supplemented
with 1% FCS versus that with 1% LPDS (Figure S7).
To investigate apo-B100 misfolding and aggregation, LDL
(freshly isolated and purified, TBARS-negative, 40 μg/mL) was
incubated quiescently with oxysterol-functionalized nano-
particles 2a (200 nM) in aqueous phosphate-buffered saline
(PBS, pH 7.4) at 37 °C, and the solution turbidity was
measured (Figure 2 inset). In the presence of 2a, the misfolding
and aggregation of apo-B100 was accelerated and proceeded
with a sigmoidal time dependence, with a lag phase of ∼3 h and
a time to 50% maximal aggregation (t50) of 6.0 0.2 h (LDL
alone has a t50 of 32.4
0.4 h under the same conditions).
These are indicative of a classical nucleation-dependent
polymerization model of aggregation,12 which typically involves
a lag phase where nucleation occurs but turbidity is minimal,
followed by a steep rise in turbidity in which amyloid sequence
formation is being seeded by the protein nuclei, and finally a
plateau phase during which no further amyloid is formed. The
way we conceptualize this accelerated aggregation process is as
a series of equilibria between apo-B100 molecules in LDL and
the nanoparticle 2a (equilibria a−f in Figure 2 are
representative examples). Thus, normal apo-B100 (represented
as two green domains) binds to the oxysterol on 2a (equilibria
a and d) and becomes misfolded (represented as one green and
one red domain) (equilibria b and e); the misfolded form may
Since the above data pointed to an LDL-dependent
mechanism of uptake of 2a into cells, we then studied whether
either of the two major receptors responsible for uptake of
modified LDL and oxysterols, SR-A and CD36, were
4101
dx.doi.org/10.1021/ja300537u | J. Am. Chem. Soc. 2012, 134, 4100−4103

Journal of the American Chemical Society
Communication
these conditions, 2a was taken up only into macrophages
(Figures 3g−i).
Preliminary trafficking studies of 2a in cultured macrophages
revealed that after 2 h of incubation there was perinuclear
localization of 2a that shared a good degree of colocalization
with ERp19 (Pearson’s coefficient 0.698), a vesicular marker of
the endoplasmic reticulum (ER) (Figure 3j−l and Figure S10).
It is known that upon being taken up into cells, LDL travels on
the endocytic pathway to late endosomes, after which the lipid
components are transferred to the ER for reprocessing and the
lipoprotein component to the lysosomes for destruction.15
While further in-depth trafficking studies are necessary to fully
resolve the localization of 2a, these preliminary data suggest
that the ultimate fate of a high percentage of the QD585-ath-B
in these cells maybe a result of its sterol surface composition.
In conclusion, we have shown that chemical surface
modification of nanoparticles with small molecules that can
target proteins of the associated corona of the nanoparticles in
biological milieu can induce protein misfolding of the
associated proteins. This misfolding event can then be used
as a trigger for cell-specific uptake of nanomaterials into cells to
which they otherwise may not be able to gain access. Given the
ubiquitous presence of protein corona on nanomaterials in vivo
and the broad interplay between small molecules and protein
misfolding, we see this approach as being a potentially wide-
ranging approach to the problem of cell-specific targeting of
nanomaterials. Looking forward to potential in vivo applica-
tions of atheronal-B surface-modified nanomaterials, we
envision that the ability either to image or to deliver therapeutic
agents selectively to CD36+ cells such as adipocytes, cardiac
and skeletal muscle, retinal pigment epithelial cells, and
dendritic cells would be a major step forward in nano-
technology.16 In more general terms, the ability to program the
nanoparticle protein corona with small molecules to expose
epitopes for a range of receptors could be developed to direct
nanoparticles into cell types that they may not have been able
to reach before.
Figure 3. Cell uptake studies of nanomaterial QD585-ath-B (2a).
Shown are confocal microscopy images of (unless otherwise stated)
raw 264.7 macrophage cells [red, nanoparticle 2a; blue, 4′6-diamidino-
2-phenylindole (DAPI); green, phalloidin]. (a−c) Lipoprotein
dependence studies. Cells were incubated with 2a in RPMI medium
supplemented with either (a) 1% FCS, (b) 1% LPDS, or (c) 1%
delipidated LPDS. (d−f) Receptor dependence studies. Cells were
incubated with 2a pretreated for 30 min with either (d) PBS, (e) anti-
SR-A antibody in PBS, or (f) anti-CD36 antibody in PBS. (g−i) Cell-
specific uptake studies. Images of the following cells treated with 2a
are shown: (g) murine macrophage (raw 264.7); (h) human dermal
fibroblast cells (0038); (i) primary murine glial cells. (j−l)
Colocalization studies. Images of a macrophage cell stained with (j)
anti-ERp19 antibody (green) and (k) 2a (red) are shown; (l) is a
merge of (j) and (k), with colocalization highlighted by the red arrow
(Pearson’s coefficient 0.698).
ASSOCIATED CONTENT
* Supporting Information
Methods and supporting figures. This material is available free
of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
paulw@scripps.edu
■
Notes
involved.13 Thus, we examined whether both antibodies and
ligands known to interact with these receptors affected the
uptake of 2a. This study revealed that neither anti-SR-A
antibodies nor polyinosinic acid (poly I), which is a known
ligand for the SR-A receptor, affected the uptake of 2a into raw
264.7 cells (Figure 3d,e and Figure S8).14 In stark contrast,
preincubation of cells with an anti-CD36 antibody resulted in
significantly reduced uptake of 2a into cells (Figure 3f).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors are grateful to Drs. Frederik Karpe and Leanne
Hodson (OCDEM) for supplying pure hLDL for this study.
Financial support for this work was provided by TSRI, The
Skaggs Institute for Chemical Biology, and The Royal Thai
Government (predoctoral fellowship for K.P.).
The scope of this CD36-receptor-mediated uptake of 2a was
explored by studying the uptake of these QDs into cells that do
not express either SR-A or CD36 (as validated by FCM; Figure
S9). Thus, CD36 null-cultured fibroblast cells and primary
astrocyte cells and CD36+ macrophages were incubated with
2a (100 nM) under identical conditions (RPMI, 1% FCS, 37
°C for 2 h). Fluorescence microscopy confirmed that under
REFERENCES
■
(1) (a) Colvin, V. L. Nat. Biotechnol. 2003, 21, 1166−1170. (b) Nel,
A.; Xia, T.; Madler, L.; Li, N. Science 2006, 311, 622−627. (c) Warheit,
̈
D. B.; Webb, T. R.; Sayes, C. M.; Colvin, V. L.; Reed, K. L. Toxicol. Sci.
2006, 91, 227−236.
(2) (a) Cedervall, T.; Lynch, I.; Foy, M.; Berggard, T.; Donnelly, S.
̊
C.; Cagney, G.; Linse, S.; Dawson, K. A. Angew. Chem., Int. Ed. 2007,
4102
dx.doi.org/10.1021/ja300537u | J. Am. Chem. Soc. 2012, 134, 4100−4103

Journal of the American Chemical Society
Communication
46, 5754−5756. (b) Cedervall, T.; Lynch, I.; Lindman, S.; Berggard,
̊
T.; Thulin, E.; Nilsson, H.; Dawson, K. A. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 2050−2055.
(3) (a) Lundqvist, M.; Stigler, J.; Elia, G.; Lynch, I.; Cedervall, T.;
Dawson, K. A. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 14265−14270.
(b) Rocker, C.; Potzl, M.; Zhang, F.; Parak, W. J.; Nienhaus, G. U. Nat.
Nanotechnol. 2009, 4, 577−580.
(4) Deng, Z. J.; Liang, M.; Monteiro, M.; Toth, I.; Minchin, R. F. Nat.
Nanotechnol. 2011, 6, 39−44.
(5) Wentworth, P. Jr.; Nieva, J.; Takeuchi, C.; Galve, R.; Wentworth,
A. D.; Dilley, R. B.; DeLaria, G. A.; Saven, A.; Babior, B. M.; Janda, K.
D.; Eschenmoser, A.; Lerner, R. A. Science 2003, 302, 1053−1056.
(6) (a) Zhang, Q.; Powers, E. T.; Nieva, J.; Huff, M. E.; Dendle, M.
A.; Bieschke, J.; Glabe, C. G.; Eschenmoser, A.; Wentworth, P. Jr.;
Lerner, R. A.; Kelly, J. W. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
4752−4757. (b) Scheinost, J. C.; Wang, H.; Boldt, G.; Offer, J.;
Wentworth, P. Jr. Angew. Chem., Int. Ed. 2008, 47, 3919−3922.
(c) Bieschke, J.; Zhang, Q.; Powers, E. T.; Lerner, R. A.; Kelly, J. W.
Biochemistry 2005, 44, 4977−4983.
(7) Bieschke, J.; Zhang, Q.; Bosco, D. A.; Lerner, R. A.; Powers, E. T.;
Wentworth, P. Jr.; Kelly, J. W. Acc. Chem. Res. 2006, 39, 611−619.
(8) Nieva, J.; Shafton, A.; Altobell, L. J.; Tripuraneni, S.; Rogel, J. K.;
Wentworth, A. D.; Lerner, R. A.; Wentworth, P. Jr. Biochemistry 2008,
47, 7695−7705.
(9) Scheinost, J. C.; Witter, D. P.; Boldt, G. E.; Offer, J.; Wentworth,
P. Jr. Angew. Chem., Int. Ed. 2009, 48, 9469−9472.
(10) Takeuchi, C.; Galve, R.; Nieva, J.; Witter, D. P.; Wentworth, A.
D.; Troseth, R. P.; Lerner, R. A.; Wentworth, P. Jr. Biochemistry 2006,
45, 7162−7170.
(11) Prapainop, K.; Wentworth, P. Jr. Eur. J. Pharm. Biopharm. 2011,
77, 353−359.
(12) Harper, J. D.; Lansbury, P. T. Annu. Rev. Biochem. 1997, 66,
385−407.
(13) (a) Kunjathoor, V. V.; Febbraio, M.; Podrez, E. A.; Moore, K. J.;
Andersson, L.; Koehn, S.; Rhee, J. S.; Silverstein, R.; Hoff, H. F.;
Freeman, M. W. J. Biol. Chem. 2002, 277, 49982−49988.
(b) Nicholson, A. C.; Han, J.; Febbraio, M.; Silversterin, R. L.;
Hajjar, D. P. Ann. N.Y. Acad. Sci. 2001, 947, 224−228.
(14) van Oosten, M.; van Amersfoort, E. S.; van Berkel, T. J. C.;
Kuiper, J. J. Endotoxin Res. 2001, 7, 381−384.
(15) Chang, T.-Y.; Chang, C. C. Y.; Ohgami, N.; Yamauchi, Y. Annu.
Rev. Cell Dev. Biol. 2006, 22, 129−157.
(16) Febbraio, M.; Silverstein, R. L. Int. J. Biochem. Cell Biol. 2007, 39,
2012−2030.
4103
dx.doi.org/10.1021/ja300537u | J. Am. Chem. Soc. 2012, 134, 4100−4103