Leading neuroblastoma cells to die by multiple premeditated attacks from a multifunctionalized nanoconstruct

Article abstract of DOI:10.1021/ja206118a

To conquer complex and devastating diseases such as cancer, more coordinated and combined attack strategies are needed. We suggest that these can be beautifully achieved by using nanoconstruct design. We present an example showing that neuroblastoma cells are selectively killed by a nanoconstruct that specifically targets neuroblastoma cells, pushes cells to the vulnerable phase of the cell cycle, and greatly enhances radiation-induced cell death. The success of this multipronged attack approach launched by cell-embedded nanoconstructs demonstrates the power and flexibility of nanotechnology in treating cancer, a difficult task for a small molecule.

Full text of DOI:10.1021/ja206118a

COMMUNICATION
pubs.acs.org/JACS
Leading Neuroblastoma Cells To Die by Multiple Premeditated
Attacks from a Multifunctionalized Nanoconstruct
||
||
||
||
Peifu Jiao,†, Hongyu Zhou,|| Mario Otto,§ Qingxin Mu,|| Liwen Li,‡, Gaoxing Su,‡, Yi Zhang,†,
Elizabeth R. Butch,# Scott E. Snyder,# Guibin Jiang,^ and Bing Yan*,‡,
||
†School of Pharmaceutical Sciences and ‡School of Chemistry and Chemical Engineering, Shandong University, Jinan,
Shandong 250012, China
Departments of Chemical Biology and Therapeutics, §Oncology, and #Radiological Sciences, St. Jude Children’s Research Hospital,
Memphis, Tennessee 38105, United States
^Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China
S Supporting Information
b
specifically and causes antibody-dependent cytotoxicity but much
ABSTRACT: To conquer complex and devastating dis-
eases such as cancer, more coordinated and combined attack
strategies are needed. We suggest that these can be beauti-
fully achieved by using nanoconstruct design. We present an
example showing that neuroblastoma cells are selectively
killed by a nanoconstruct that specifically targets neuroblas-
toma cells, pushes cells to the vulnerable phase of the cell
cycle, and greatly enhances radiation-induced cell death.
The success of this multipronged attack approach launched
by cell-embedded nanoconstructs demonstrates the power
and flexibility of nanotechnology in treating cancer, a
difficult task for a small molecule.
less complement activation than the related chimeric anti-GD2
antibody, ch14.18,8,9 which is currently the focus of an immunother-
apy trial by the Children’s Oncology Group.10 We incorporated
hu14.18K322A into our nanoconstruct as a targeting moiety to
recognize neuroblastoma cells while avoiding normal cells.
We and others have previously shown that gold nanoparticles
(GNPs), like other elements with high atomic number (Z),
enhance radiation-induced cell death.11À17 All cells are not
equally sensitive to radiation during all stages of the cell cycle.18
For example, cells are radiation-resistant in the S phase, whereas they
become very sensitive to radiation in the G2/M phase.19,20 An agent
that arrests cells at the G2/M phase of the cell cycle, such as
paclitaxel (PTX), would thus be an effective chemical weapon to
make cells vulnerable to radiation. To carry out a specific lethal
assault on cancer cells while sparing normal cells, we designed a
nanoconstruct-initiated three-strike approach. First, neuroblastoma
cells are specifically targeted. Next, cells are artificially arrested in the
G2/M phase of the cell cycle. Finally, the cancer cells are killed by
the enhanced radiation cytotoxicity.
To assemble a nanoconstruct with these multiple arsenals, we
first prepared thiol-linker-containing hu14.18K322A antibody along
with β-cyclodextrin (β-CD) for paclitaxel loading. Through AuÀS
bonding, they were all incorporated into GNPs along with PEG
polymers to form hu14.18K322A-targeted GNPs (HGNPs) and
hu14.18K322A-targeted, PTX-loaded GNPs (HPGNPs) [see the
Supporting Information (SI) for details].
The number of hu14.18K322A molecules on each GNP was
determined by elemental analysis of nitrogen content (%) in a
given amount of nanoconstructs. To determine the number of
PTX molecules on each HPGNP, LCÀMS analysis was used to
quantify the amount of PTX remaining after the drug was loaded
onto the HPGNPs. There were 16 hu14.18K322A molecules and
∼11 PTX molecules on each HPGNP (Table 1 and Figure S3
in the SI). The diameter of a bare GNP was ∼15 nm on the basis
of transmission electron microscopy (TEM) measurements
(Table 1, Figure 1a, and Figure S2). Phosphotungstic acid
(PTA) was used to stain hu14.18K322A on the surface of the
GNPs, enabling the visualization of GNP-bound hu14.18K322A
ancer is a deadly disease with high complexity. Multipronged
C
and coordinated approaches are needed to conquer this
devastating disease. Because of intrinsic limitations, it is difficult to
build multiple capabilities into a single small molecule; however, it is
possible to equip a nanoconstruct with multiple arsenals. Among
these arsenals, chemical and biological agents can be launched one-
by-one to stage multiple attacks on cancer cells in a coordinated
effort to kill cells. Here we report a cancer-targeting, cell-modulating,
radiation-enhancing, biocompatible nanoconstruct that strategically
kills neuroblastoma cells with high specificity.
Neuroblastoma is the most common extracranial solid tumor
in childhood.1 More than 50% of neuroblastoma cases occur in
children younger than 2 years old. As a disease of the sympaticoa-
drenal lineage of the neural crest, tumors can develop anywhere in
the sympathetic nervous system.2 Current treatments, including
surgery, chemotherapy, radiation, and stem cell transplantation,
induce significant side effects such as growth delays, hearing loss,
and learning disabilities. Despite multimodality therapy, many
patients with advanced-stage disease ultimately succumb to relapse.
Therefore, more effective and cancer-specific therapy is urgently
needed, especially for this fragile patient population.
GD2 disialoganglioside is an antigen expressed on tumors of
neuroectodermal origin, including neuroblastoma,3À5 with a
highly restricted expression on normal tissues.6 Hu14.18K322A
is a humanized anti-GD2 antibody currently being investigated in
a phase-I immunotherapy study in neuroblastoma patients at St.
Jude Children’s Research Hospital.7 This antibody binds to GD2
Received: July 1, 2011
Published: August 09, 2011
r
2011 American Chemical Society
13918
dx.doi.org/10.1021/ja206118a J. Am. Chem. Soc. 2011, 133, 13918–13921
|

Journal of the American Chemical Society
COMMUNICATION
Table 1. Characteristics of GNPs, Hu14.18K322A, HGNPs,
and HPGNPs
Figure S4). PC-3, a human prostate cancer cell line, did not have any
detectable GD2 (MFI = 120), as previously reported.21,22 It is
therefore a good representative of normal cells and was used as
negative control.
number per GNP
With nanoconstructs made and cell lines selected, our first test
was to determine whether HPGNPs specifically target neuro-
blastoma cells by avoiding cells that do not express the GD2
antigen, such as normal cells. Dark-field microscopy images
showed that HPGNPs accumulated abundantly in GD2-positive
cells but not in GD2-negative cells (Figure S7). Additional
ultrastructural details of the time-dependent cellular uptake of
HPGNPs were provided by TEM. After a 4 h incubation of cells
with HPGNP, individual particles were selectively bound to
IMR32 and CHLA-20 plasma membranes (Figure 2b,c) but
not to a PC-3 membrane, indicating specific GD2 recognition
and binding. At 12 h, bound HPGNPs were internalized by cells.
The internalized HPGNPs were localized in endosomal- or
lysosomal-like organelles and also in the cytoplasm, indicating
that they might enter cells by penetrating the plasma membrane
or by breaking endosomal membranes. Membranes of some
endosomes or lysosomes were actually ruptured (arrows in
Figure 2e,f). This is very interesting, although we currently have
no explanation for this observation. Furthermore, we also found
particles in the cytoplasm at 4 h, when endosomal membranes
were not ruptured (Figure 2c inset).
sample
GNPs
avg size (nm)a
hu14.18K322Ab
PTXc
15 ( 5
À
À
À
À
hu14.18K322A
HGNPs
12 ( 3
50 ( 10
50 ( 15
16
À
HPGNPs
16
11
a Determined by DLS analysis (Figure S2). b Determined by elemental
analysis (Figure S3). c Determined by HPLCÀMS (Figure S3).
We next analyzed cellular gold content by using ICP-MS to
evaluate quantitatively the cells’ recognition and internalization
of HPGNPs. CHLA-20 and IMR32 cells had 60- and 50-fold
more HPGNPs, respectively, than did GD2-negative PC-3 cells
(Figure 2g,h), indicating extremely high selectivity. Additionally,
the average uptake rate of CHLA-20 cells (kCHLA-20 = 1.4820)
was 1.4-fold higher than that of IMR32 cells (kIMR32 = 1.0881)
(Figure S5), suggesting that the amount of GD2 expression on
the cell surface played a role in accelerating the cell uptake of
HPGNPs. The amounts of internalized HPGNPs reached plateaus
at 24 h when the HPGNP concentration was 2.5 nM, indicating that
the cellular concentrations of HPGNPs finally reached a steady state
(Figure 2h). At equilibrium, the total internalized HPGNP contents
of CHLA-20 and IMR32 cells were 62- and 48-fold higher,
respectively, than those of PEGÀGNPs in the corresponding cells
(nonspecific uptake; Figure S6). This result suggested that the
hu14.18K322AÀGD2 interaction was responsible for the specific
uptake of HPGNPs by GD2-positive cells. To provide further
confirmation that the recognition of HPGNPs by GD2-positive
neuroblastoma cells was a result of GD2Àhu14.18K322A interac-
tions, an antibody competition experiment was performed. CHLA-
20 cells were pretreated with free hu14.18K322A for 4 h and then
treated with HPGNPs. The cellular uptake of HPGNPs was
quantitatively evaluated using ICP-MS (Figure 2i). The HPGNP
uptake by CHLA-20 was nearly blocked by 25À100 nM free
hu14.18K322A. Such a blockade was also confirmed by TEM analysis
(Figure 2j,k). In the presence of 25 nM hu14.18K322A, fewer
HPGNPs were bound to cells or internalized. These results demon-
strate that the enhanced cell recognition in GD2-positive neuroblas-
toma cells was a direct result of GD2Àhu14.18K322A interactions.
After HPGNPs were specifically internalized into neuroblas-
toma cells through hu14.18K322AÀGD2 interactions, they were
ready to push cells into the G2/M phase by delivering PTX mole-
cules to intracellular targets (i.e., tubulins). As shown in Figure 3, the
HPGNPs arrested 68% of the CHLA-20 cells in the G2/M phase at
12 h, whereas PTX caused only a 44% G2/M arrest rate. Similar
results were seen for IMR32 cells. In PC-3 cells, free PTX caused
Figure 1. Characterization of HPGNPs. (a, b) TEM images of (a) bare
GNPs and (b) HPGNPs (stained with 2% PTA, pH 6.0, scale bar
100 nm). (c) Structure of an HPGNP. (d) SECÀHPLC traces (mobile
phase: 3 mM sodium citrate). (e) Stability of HPGNPs in phosphate-
buffered saline at 4 ꢀC.
molecules (Figure 1b). The dark clouds around the GNPs represent
molecular backbones of hu14.18K322A. The little white dots closer
to the GNP surface are molecules that could not be stained by PTA,
such as bound CD, loaded PTX, PEG units, and polysaccharides in
the Fc domain of hu14.18K322A. Furthermore, the distances
between adjacent HPGNPs were larger than those between adjacent
GNPs, suggesting that hu14.18K322A molecules were conjugated to
the GNPs. Dynamic light scattering (DLS) was used to measure the
hydrodynamic size of the particles in an aqueous solution (Figure
S2). The diameter of a bare GNP was 15 ( 5 nm, in agreement with
the sizes obtained using TEM. After conjugation with hu14.18K322A
(12 ( 3 nm in diameter), PEG thiol (HS-PEG, MW 5000 Da), and
thiol-appended β-CD (HS-β-CD), the hydrodynamic diameter of
the GNPs increased to 50 ( 15 nm. Size-exclusion chromatogra-
phyÀHPLC (SECÀHPLC) analysis also showed the size increase
(evidenced by early elution; Figure 1d). Because free hu14.18K322A
molecules in the HPGNP solution might hinder the targeting
efficiency of HPGNP in GD2-positive tumor cells, excess antibodies
were completely removed by six rounds of washing and centrifuga-
tion. The presence of pure HPGNPs without any free antibody was
confirmed by the SECÀHPLC results (Figure 1d).
Cells may express various levels of GD2 antigens on their
surface. Using a fluorescent secondary antibody and flow cytometry,
we measured the amount of GD2 expression on the surface of
CHLA-20, IMR32, and PC-3 cells. We found that the neuroblastoma
cell lines, CHLA-20 and IMR32, were both GD2-positive, although
the expression of GD2 in IMR32 [mean fluorescence intensity
(MFI) = 7304] was only 77% of that in CHLA-20 (MFI = 9554;
13919
dx.doi.org/10.1021/ja206118a |J. Am. Chem. Soc. 2011, 133, 13918–13921

Journal of the American Chemical Society
COMMUNICATION
Figure 3. Induction of G2/M arrest by HPGNPs (PTX content 11 nM,
12 h) and PTX (11 nM, 12 h) in GD2-negative cells and neuroblastoma cells.
the up-regulation of ABC transporter genes, which leads to the rapid
elimination of various cytotoxic agents, including taxols.23 IMR32
and CHLA-20 cells are multi-drug-resistant cell lines that utilize this
mechanism.24,25 As an alternative explanation for the increased G2/M
arrest rate conferred by HPGNPs, one could thus speculate that
intracellular HGNP-bound PTX may at least partially evade rapid
elimination mediated by ABC transporters and increase its cytotoxic
effects, as we observed in our experiments. In sum, these results
demonstrate that the GD2-targeting HPGNPs can specifically deliver
PTX to neuroblastoma cells to induce cell-cycle arrest without affect-
ing GD2-negative cells (i.e., normal cells).
Once HPGNPs specifically brought cells to the vulnerable
G2/M phase, we aimed to kill neuroblastoma cells using radia-
tion, taking advantage of the enhanced radiation effects caused by
HPGNPs and the enhanced vulnerability of dividing cells.
HGNPs alone showed no cytotoxicity to either GD2-positive
or -negative cells, indicating a high degree of biocompatibility
(Figure 4aÀc). As expected, HGNPs enhanced the radiation-
induced cell death in GD2-positive neuroblastoma cells (Figure 4b,c
and Figure S10). Both CHLA-20 and IMR32 cells have been
reported to be drug-resistant,26À28 especially to PTX. According to
our results, PTX (11 nM) did not cause much toxicity to these cell
lines after 12 h (Figure 4e,f). However, incubation of cells with
HPGNPs without radiation induced more cell death under the same
conditions (Figure 4h,i and Figure S9), suggesting that cancer-
targeting HPGNPs transported many more PTX molecules into
neuroblastoma cells and possibly evaded efflux generated by ABC
transporters. HPGNPs were not toxic to PC-3 cells (Figure 4g),
although the PTX was highly toxic under the same conditions
(Figure 4d). This result strongly demonstrates that the HPGNPs
effectively reduced the toxicity of PTX to GD2-negative cells by
evading them. Because HPGNP caused neuroblastoma cells to pause
attheG2/Mphase, radiationtreatmentbecamemuchmoreeffective,
thus doubling the cell death (Figure 4h,i and Figure S9).
Figure 2. HPGNP recognition by and internalization into neuroblastoma
cells. (aÀf) TEM images revealing the targeting process and cellular distribu-
tion of HPGNPs (2.5 nM) in neuroblastoma cells and GD2-negative PC-3
cells (scale bar 500 nm). (g) Dose-dependent cellular uptake of HPGNPs in
neuroblastoma cells and GD2-negative PC-3 cells (incubation time 12 h).
(h) Time-dependent cellular uptake of HPGNPs in neuroblastoma cells
and GD2-negative PC-3 cells (2.5 nM). (i) Cellular uptake of HPGNPs
(2.5 nM) in CHLA-20 cells pretreated with hu14.18K322A for 4 h and then
with HPGNPs for 12 h. (j, k) TEM images of CHLA-20 cells pretreated
with free antibody (0 and 25 nM) for 4 h and then treated with HPGNPs
(2.5 nM) for 12 h (scale bar 100 nm).
In summary, we have designed and assembled a novel nano-
construct, HPGNPs, with multipronged attack capabilities spe-
cifically targeted to neuroblastoma cells. HPGNPs accomplish
three key strikes: they specifically target neuroblastoma cells;
significant cell-cycle arrest at the G2/M phase (43%). However,
HPGNPs did not induce G2/M arrest, suggesting that HPGNPs
effectively blocked the cellular uptake of PTX by GD2-negative cells.
A major and critical mechanism of drug resistance in cancer cells is
13920
dx.doi.org/10.1021/ja206118a |J. Am. Chem. Soc. 2011, 133, 13918–13921

Journal of the American Chemical Society
COMMUNICATION
’ REFERENCES
(1) Baker, D. L.; Schmidt, M. L.; Cohn, S. L.; Maris, J. M.; London,
W. B.; Buxton, A.; Stram, D.; Castleberry, R. P.; Shimada, H.; Sandler, A.;
Shamberger, R. C.; Look, A. T.; Reynolds, C. P.; Seeger, R. C.; Matthay,
K. K. New Engl. J. Med. 2010, 363, 1313.
(2) Maris, J. M.; Hogarty, M. D.; Bagatell, R.; Cohn, S. L. Lancet
2007, 369, 2106.
(3) Mujoo, K.; Cheresh, D. A.; Yang, H. M.; Reisfeld, R. A. Cancer
Res. 1987, 47, 1098.
(4) Cheung, N. K. V.; Saarinen, U. M.; Neely, J. E.; Landmeier, B.;
Donovan, D.; Coccia, P. F. Cancer Res. 1985, 45, 2642.
(5) Kramer, K.; Gerald, W. L.; Kushner, B. H.; Larson, S. M.;
Hameed, M.; Cheung, N. K. V. Med. Pediatr. Oncol. 2001, 36, 194.
(6) Svennerholm, L.; Bostrom, K.; Fredman, P.; Jungbjer, B.; Lekman,
A.; Mansson, J. E.; Rynmark, B. M. Biochem. Biophys. Acta 1994, 1214, 115.
(7) Navid, F.; Barfield, R. C.; Handgretinger, R.; Sondel, P. M.; Shulkin,
B. L.; Kaufman, R.; Billups, C.; Wu, J.; Furman, W. L.; McGregor, L. M.;
Otto, M.; Gillies, S.; Santana, V. M. J. Clin. Oncol. 2011, 29 (Suppl.), 9523.
(8) Navid, F.; Santana, V. M.; Barfield, R. C. Curr. Cancer Drug
Targets 2010, 10, 200.
(9) Sorkin, L. S.; Otto, M.; Baldwin, W. M.; Vail, E.; Gillies, S. D.;
Handgretinger, R.; Barfield, R. C.; Yu, H. M.; Yu, A. L. Pain 2010, 149, 135.
(10) Yu, A. L.; et al. New Engl. J. Med. 2010, 363, 1324.
(11) Li, X.; Zhou, H.; Yang, L.; Du, G.; Pai-Panandiker, A. S.; Huang,
X.; Yan, B. Biomaterials 2011, 32, 2540 .
Figure 4. Enhancement of radiation cytotoxicity to neuroblastoma cells
by HPGNPs. All cells were treated with or without HGNP, PTX, and
HPGNP for 12 h. After exposure to X-rays (3 Gy dose), all particles or drugs
were removed, and fresh medium was added. After another 48 h, cell viability
was measured using the WST-1 method. The concentrations of HPGNPs in
(gÀi) were kept the same as that of HGNPs, so the PTX concentrations on
HPGNPs were identical to those in (dÀf).
(12) Zhou, H. Y.; Jiao, P. F.; Yang, L.; Li, X.; Yan, B. J. Am. Chem. Soc.
2011, 133, 680.
(13) Chattopadhyay, N.; Cai, Z. L.; Pignol, J. P.; Keller, B.; Lechtman,
E.; Bendayan, R.; Reilly, R. M. Mol. Pharmaceutics 2010, 7, 2194.
(14) Hainfeld, J. F.; Dilmanian, F. A.; Zhong, Z.; Slatkin, D. N.;
Kalef-Ezra, J. A.; Smilowitz, H. M. Phys. Med. Biol. 2010, 55, 3045.
(15) Jiao, P. F.; Zhou, H. Y.; Chen, L. X.; Yan, B. Curr. Med. Chem.
2011, 18, 2086.
(16) Jelveh, S.; Chithrani, D. B. Cancers 2011, 3, 1081.
(17) Chithrani, D. B.; Jelveh, S.; Jalali, F.; van Prooijen, M.; Allen, C.;
Bristow, R. G.; Hill, R. P.; Jaffray, D. A. Radiat. Res. 2010, 173, 719.
(18) Terasima, T.; Tolmach, L. J. Nature 1961, 190, 1210.
(19) Dunne, A. L.; Mothersill, C.; Robson, T.; Wilson, G. D.; Hirst,
D. G. Oncol. Res. 2004, 14, 447.
drive them to the G2/M phase of the cell cycle, thus making them
highly vulnerable to radiation; and significantly enhance radia-
tion-induced cell death. HPGNPs did not bind GD2-negative
cells and caused little toxicity or radiation-induced cell death in
these cells. Our data demonstrate the power and flexibility of
nanotechnology in treating complex diseases such as cancer by
using multiple strategies to attack cancer cells from multifunc-
tional nanoconstructs, a task that is difficult to achieve by using a
small molecule alone.
(20) Hennequin, C. Cancer Radiother. 2004, 8 (Suppl. 1), S95.
(21) Voss, S. D.; Smith, S. V.; DiBartolo, N.; McIntosh, L. J.; Cyr, E. M.;
Bonab, A. A.; Dearling, J. L.; Carter, E. A.; Fischman, A. J.; Treves, S. T.;
Gillies, S. D.; Sargeson, A. M.; Huston, J. S.; Packard, A. B. Proc. Natl. Acad.
Sci. U.S.A. 2007, 104, 17489.
(22) Vavere, A.; Butch, E.; Dearling, J.; Packard, A.; Snyder, S. J. Nucl.
Med. 2010, 51 (Suppl. 2), 480.
(23) Goda, K.; Bacso, Z.; Szabo, G. Curr. Cancer Drug Targets 2009,
9, 281.
(24) Inge, T. H.; Harris, N. L.; Wu, J. Q.; Azizkhan, R. G.; Priebe, W.
’ ASSOCIATED CONTENT
S
Supporting Information. Preparation and characteriza-
b
tion of GNPs, HGNPs, HPGNPs, and β-CD derivatives; cell
culture; GD2 expression measurement; cellular uptake of
HPGNPs; competition experiment with free hu14.18K322A;
dark-field microscopy; TEM; release of PTX; cell cycle assay;
cytotoxicity assay; and complete ref 10. This material is available
free of charge via the Internet at http://pubs.acs.org.
J. Surg. Res. 2004, 121, 187.
(25) Kuss, B. J.; Corbo, M.; Lau, W. M.; Fennell, D. A.; Dean, N. M.;
Cotter, F. E. Int. J. Cancer 2002, 98, 128.
’ AUTHOR INFORMATION
(26) Goto, H.; Yang, B.; Petersen, D.; Pepper, K. A.; Alfaro, P. A.; Kohn,
D. B.; Reynolds, C. P. Mol. Cancer Ther. 2003, 2, 911.
(27) Keshelava, N.; Groshen, S.; Reynolds, C. P. Cancer Chemother.
Pharmacol. 2000, 45, 1.
Corresponding Author
dr.bingyan@gmail.com
(28) Tweddle, D. A.; Pearson, A. D. J.; Haber, M.; Norris, M. D.; Xue,
C.; Flemming, C.; Lunec, J. Cancer Lett. 2003, 197, 93.
’ ACKNOWLEDGMENT
We thank Sharon Frase, Linda Mann, Jennifer Peters, Tom
Mohaupt, and Richard Ashmun for technical assistance and
Merck KGaA for the generous gift of hu14.18K322A antibody.
We also thank Cherise M. Guess for editing and checking the text
of the manuscript. This work was supported by the National
Basic Research Program of China (2010CB933504), the Na-
tional Natural Science Foundation of China (90913006 and
21077068), the National Cancer Institute (P30CA027165), and
the American Lebanese Syrian Associated Charities (ALSAC).
13921
dx.doi.org/10.1021/ja206118a |J. Am. Chem. Soc. 2011, 133, 13918–13921