SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
nature materials
| www.nature.com/naturematerials
1
Supplementary Information for "Gold Nanocages Covered by Smart
Polymers for Controlled Release
with Near-Infrared Light"
Mustafa S. Yavuz, Yiyun Cheng, Jingyi Chen,
Claire, M. Cobley, Qiang, Zhang, Matthew
Rycenga, Jingwei Xie, Chulhong Kim, Andrea
G. Schwartz, Lihong V. Wang & Younan Xia
Department of Biomedical Engi
neering, Washington University,
St. Louis, Missouri 63130, USA
Correspondence should be addressed to Y.X. (e-mail: xia@biomed.wustl.edu)
Chemicals & Materials
The monomer,
N
-isopropylacrylamide (NIPAAm, 99%) was obtained from Sigma-Aldrich and
re-crystallized in hexane prior to use. Acrylamide (AAm, 99%) was purchased from Sigma-
Aldrich and recrystallized in methanol
before use. Copper(I) bromide (CuBr),
N,N
,
N
’,
N’
,
N’
-
pentamethyldiethylenetriamine (PMDETA), po
ly(ethylene glycol) methyl ether (MPEG,
M
W
≈
5,000),
N
,
N
’-dicyclohexylcarbodiimide (DCC), 4
-(dimethylamino)pyridine (DMAP),
doxorubicin hydrochloride (Dox), 5
-(3-nitrophenylazo) sa
licylic acid sodium salt (alizarin
yellow), and diethyl ether we
re obtained from Sigma-Aldr
ich and used as received.
Tetrahydrofuran (THF), methanol, acetic acid, and
cellulose dialysis tubes were obtained from
Fisher Scientific and used as received. The di
sulfide-containing initia
tor, bis(2-hydroxyethyl)
disulfidebis(2-bromo propionate) (BHEDS(BP)
2
) was synthesized according to the literature (see
Ref. 13 of the main text). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT
assay) and Dulbecco's Phosphate Buffered Saline
(DPBS) were purchased from Invitrogen and
used as received. Micro BCA Prot
ein Assay kit was obtained from Thermo Scientific and used as
received. Lysozyme (from chicken egg white) and
Micrococcus Lysodeikticus
(lyophilized,
washed with 0.01 M EDTA at pH=7.0 and then 0.15 M NaCl) were obtained from Sigma-
Aldrich and used as received.
Instrumentation
A tunable Ti:sapphire laser (LT-2211A, LOTI
S TII) pumped by a Q-switched Nd:YAG laser
S1
2
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
(LS-2137/2, LOTIS TII) was used as the light so
urce for the drug-release experiments. The laser
beam had a ~6.5 ns pulse duration and a 10 Hz
pulse-repetition rate. The output wavelengths
range from 730 to 820 nm, with a peak at 790 nm.
The average power of th
e laser could be tuned
in the range of 0.2-0.4 W. The energy was atte
nuated by several opti
cal components for the
laser-triggered release. In
addition, a ground glass with 660
grit number (from Thorlabs) was
used to expand the laser beam homogeneously. In all experiments, the average power was kept
constant, and the power density was adjusted
by controlling the beam diameter. The power
density was controlled in
the range of 10-40 mW/cm
2
, corresponding to an en
ergy density of 1-4
mJ/cm
2
, which are well below the safety limit (20 mJ/cm
2
) of pulse energy for skin at 800 nm in
the absence of any contrast agent (see American
National Standard for Safe Use of Lasers: ANSI
Z136.1-2000). A UV-Vis-NIR spectrometer (Cary 50
Bio, Varian Inc.) was used to obtain the
absorbance spectra.
For gel permeation chromatography (GPC) measurements, the polymer sample (10 mg) was
dissolved in 5 mL of tetrahydrofuran (THF),
followed by filtering over
a nonsterile PTFE filter
with a pore size of 0.45 μm. The me
asurements were performed at 34
o
C with a Waters 150-C
Plus GPC equipped with a Waters 410 different
ial refractometer, a Waters 2487 dual wavelength
absorbance UV-vis detector set to 254 nm,
a Polymer Laboratories PL-ELS 1000 evaporative
light scattering detector, and a
Jordi Gel DVB 105 Å, a PL Ge
l 104 Å, a Jordi Gel DVB 100 Å,
and a Waters Ultrastyragel 500 Å column setup. THF
was used as an eluent
at a flow rate of 3
mL/min. Number-average (M
n
) and weight-average (M
w
) molecular weights were determined
from calibration plots constructed
with polystyrene standards.
We characterized the polymer-covered nanocages
by thermal gravimetric analysis (TGA) using
Hi-Res TGA 2950. The analysis measures weight lo
ss as a function of temperature or time. The
sample was added to a platinum pan and then
the pan was placed on the sample holder of the
TGA instrument. Water in the sample was removed by ramping at 10
o
C/min to 120
o
C and
holding isothermally for 20 min under argon. After
the sample had been cooled down to 50
o
C,
the TGA measurement was recorded by ramping at 10
o
C/min to 525
o
C under argon. The
decomposition temperature of the copolymer was measured to be around 415
o
C. The
S2
nature materials
| www.nature.com/naturematerials
3
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
hydrodynamic diameters of the polymer-covered nano
cages were measured using dynamic light
scattering (DSL) with a Zetasizer Nano-ZS from Malvern Instruments.
Cell Culture Study
The cultured SK-BR-3 breast cancer cells were tr
ansferred to 12 wells of
a 24-well plate and
kept at 37
o
C before use. Each well contained 4×10
4
cells in 1.5 mL
of DPBS. Half an hour
before the addition of Dox-loaded Au nanocages to
the wells and after the releasing experiments,
the plate was kept at 37
o
C, below the LCST (39
o
C) of the copolymer. Each experiment was
repeated three times. A total of 6 wells we
re used for the controls (C-1 and C-2).
C-1
: with laser
irradiation in the absen
ce of Au nanocages at a power density of 20 mW/cm
2
for 2 min;
C-2
:
with laser irradiation in the presence of empty Au nanocages at a pow
er density of 20 mW/cm
2
for 2 min. Dox-loaded Au nanocages were added to
the next 6 wells. The cells were exposed to
laser for 2 min and 5 min at a fixed power density of 20 mW/cm
2
. After exposure to the laser, the
samples were incubated for 24 h. Subsequently
, 30 μL of MTT (3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide) assay and 270 μL
of the medium were added to each well.
After incubation for 3 h, the supernatant in each
well was aspirated. 300 μL of isopropanol was
added to each well before the plate was examined
using a microplate reader
(Infinite F200 series,
Tecan Group Ltd.). Cell viability was determined
from the absorbance at 560 nm relative to the
controls. In this case, the absorbance represents the amount of formazan formed due to the
reaction between the mitochondria of
viable cells and the MTT assay
.
Bioactivity Test for the Released Lysozyme
The bioactivity of lysozyme was determined from the rate of lysis for
Micrococcus lysodeikticus
.
0.1 mL of DPBS buffe
r containing 0.05 mg
Micrococcus Lysodeikticus
was added to a 96-well
plate. 0.1 mL of native and rel
eased lysozymes (at the same c
oncentration) was added to the
wells containing
Micrococcus Lysodeikticus
, and the samples were immediately examined using
a microplate reader. The absorbance at 450 nm wa
s recorded every minute. We then plotted the
absorbance at 450 nm against th
e concentration of lysozyme
for both native and released
lyszoymes. The ratio of the slopes for these two
linear fits provides the
percentage of enzyme
bioactivity after going through the lo
ading and releasing processes.
S3
4
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Potential Rupture of the Au-Thiolate Bond and
thus Desorption of the Polymer from the
Surface of Gold Nanocages during Exposure to a Pulsed Laser
It has been reported by
a number of studies (
J. Am. Chem. Soc.
,
2006
,
128
, 2426;
J. Am. Chem.
Soc.
,
2006
,
128
, 3709; and
ACS Nano
,
2009
,
3
, 80) that the Au-thiolate bond could be ruptured
to cause desorption for the thiolate molecules du
ring exposure to a pulsed laser. In the present
work, the polymer on Au nanocages did not come of
f during laser irradiation due to the use of a
much lower power density. In our study, we typi
cally used an energy density of 1 mJ/cm
2
in the
nanosecond regime and the peak power dens
ity of each pulse was on the order of 10
5
W/cm
2
. For
all other studies mentioned above, the energy density in the sub-picosecond regime that caused
the Au-thiolate bond to dissociate was 38, 0.8, and 1.68 mJ/cm
2
, respectively. In addition, their
peak power density of each
pulse was in the range of 10
9
to 10
11
W/cm
2
. Essentially, their peak
power densities were 4-6 orders of magnitude hi
gher than the nanosecond
pulse we used in the
present work. We suspect that the high peak
power density generated by the sub-picosecond
pulsed laser is responsible for the observed di
ssociation of the Au-thiolate bond on an electron-
phonon coupling time scale (typically within ~1
ps for Au, as reported in another study,
J. Am.
Chem. Soc.
,
2006
,
128
, 2426). To our knowledge, that is no re
port on the dissociation of the Au-
thiolate by exposure to a nanosecond pulsed laser. The melting of our Au nanocages at an energy
density of 4 mJ/cm
2
may cause some of the thiolate molecu
les to desorb from the surface of the
Au nanocages but this energy density is much higher th
an the range of 1-2 mJ/cm
2
we typically
used for the laser-triggered release.
A Plausible Mechanism for the Laser Trigged Release
For our system, we assume that each Au nano
cage has essentially a
“monolayer” of pNIPAAm
on its surface. Since each polymer chain is teth
ered to the nanocage vi
a a Au-thiolate bond, the
time scales associated with the
phase transition should be consiste
nt with the phase transition of
a polymer chain rather than a network of polymer
chains like in a smart “gel” system (where a
phase separation is involved). That said we can estimate the time constant (
τ
) for the polymer
folding, or collapse, according to the diffusion-collision model [1, 2]:
τ
-1
~ 3
D
R
MIN
/ (
R
MAX
)
3
S4
nature materials
| www.nature.com/naturematerials
5
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
where R
MAX
and R
MIN
are the hydrodynamic radi
i of the coiled and globular states of the
pNIPAAm chain on the nanocage (and these we de
termined from DLS measurements to be 24
and 17 nm, respectively), and
D
is the diffusion constant associated with the polymer chain.
From the parameters of the laser system we used, we estimate that the nanocages will be above
the LSCT for about 6.5 ns during the irradiation of each pulse. Solving for
D
we find that the
pNIPAAm chain would need
a diffusion constant of 3
×
10
-7
m
2
s
-1
. We note that typical values
of
D
for pNIPAAm in solution are 10
-8
-10
-10
m
2
s
-1
[3]. This difference between
D
is likely due
to the fact our polymers are tethered to a solid su
rface. This has two important effects: (i) we can
assume that each polymer chain is more or less isolated from each other as we have formed a
“monolayer” on the nanocage, and (ii) the surface of the nanocage will
stabilize the hydrophobic
globular conformation of the polymer as it limits
its contact with the surrounding water. These
could reduce the folding time, or increase th
e polymer’s diffusion, as the polymers are not
overlapped and the energy barrier to the globul
ar state may be lowered. Recent experiments
show that the coil-globular transition of pNIP
AAm occurs in solution on timescales near 30
μ
s
[3]. We suspect that surface confinement of th
e pNIPAAm chains will result in faster phase
transition, although we admit that transitions
on the nanosecond timescale are not well-known.
However, nanoscale heat transport is a very ne
w concept and a recent report suggests that the
heat propagated through the m
onolayer coating on a nanoparticle
via molecular vibrations on the
picosecond timescale, forcing the molecules to
change conformation [4]. Once the polymer chain
is in the globular state, it might
take a longer tim
escale to go back to the co
il state so the pores on
the Au nanocage can remain open
for a certain period of time. Th
ere might be other mechanisms
responsible for the observed release under irradiati
on by a pulsed laser. More fundamental
studies are needed in order to
completely resolve this issue.
References
[1] M. Karplus and D. L. Weaver,
Protein Sci
.
1994
, 4, 650.
[2] Y. Tsuboi, Y. Yoshida, N. Kitamura, K. Iwai,
Chem. Phys. Lett
.,
2009
, 468, 42.
[3] Y. Tsuboi, Y. Yoshida, K. Okada, N. Kitamura,
J. Phys. Chem. B
,
2008
, 112, 2562.
[4] J. A. Carter, Z.
Wang, D. D. Dlott,
J. Phys. Chem. A
,
2008
, 112, 3523.
S5
6
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S1
The LCST (or cloud point) measured spectro
sphotometrically with
the solution being
heated at a rate of 1
o
C/min. The temperature at 90% light transmittance (at 450 nm) of the
original polymer solution was defined as the LCST.
S6
nature materials
| www.nature.com/naturematerials
7
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S2
TGA analysis of the copolymer-covered Au nanocages. The 11% weight loss up till
450
o
C corresponds to the desorption and decomposition of the copolymer.
S7
8
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S3
Plots of dynamic light scattering data
showing the mean hydrodynamic diameters of
Au nanocages as a funciton of
solution temperature. We tested
both Au nanocages covered by
PVP and the copolymer. Note that only the copol
ymer-covered Au nanocages showed reversible
size changes in response to temperature variatio
ns. The lines were added for aiding visulization
purpose only.
S8
nature materials
| www.nature.com/naturematerials
9
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S4
UV-vis absorbance spectra of alizarin-PEG
released from the copolymer-covered Au
nanocages at two different temperatures: 42
o
C (red) and 25
o
C (black). These results indicate
that the release of alizarin-PEG was very negl
igible under ambient cond
itions and was increased
drastically as the sample was heated to a temperature above the LCST (39
o
C) of the copolymer.
S9
10
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S5
TEM image of melted Au nanocages upon expos
ure to the laser at a power density of
40 mW/cm
2
. In this case, the nanocages have been tran
sformed into more or less solid particles.
S10
nature materials
| www.nature.com/naturematerials
11
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Figure S6
Absorbance spectra of alizarin-PEG rel
eased from copolymer-covered Au nanocages.
The Au nanocages were used first for dye loadin
g and laser-triggered rele
ase (Figure 2b), loaded
again and then released by heating.
S11
12
nature materials
| www.nature.com/naturematerials
SUPPLEMENTARY INFORMATION
doi: 10.1038/nmat2564
Table S1
Characterization of th
e pNIPAAm and pNIPAAm-
co
-pAAm.
NIPAAm/AAm
(molar ratio)
LCST
a
(
o
C)
M
n
(×10
4
)
M
w
(×10
4
)
Polydispersity index
(PDI=M
w
/M
n
)
100 : 0
32.3
6.87
12.4
1.80
95 : 5
34.8
6.77
11.9
1.76
90 : 10
39.0
6.93
12.1
1.75
87.5 : 12.5
40.8
7.24
12.8
1.77
75 : 25
48.9
7.89
14.1
1.79
a
The LCST was determined from Figure S1.
S12