Nanobubbles from gas-generating polymeric nanoparticles: Ultrasound imaging of living subjects

Article abstract of DOI:10.1002/anie.200903841

(Figure Presented) A bubbly personality: Coalescence and fusion of carbon dioxide nanobubbles in a tumor leads to microbubbles that can be imaged by ultrasound (see picture). The bubbles are released from gas-generating polymeric nanoparticles that are coated with hydrophobically modified chitosan. Hydrolysis of the polymer poly(BL-PO), which consists of a biodegradable main backbone and a carbonate side chain, results in the release of carbon dioxide.

Full text of DOI:10.1002/anie.200903841

Communications
DOI: 10.1002/anie.200903841
Imaging Agents
Nanobubbles from Gas-Generating Polymeric Nanoparticles:
Ultrasound Imaging of Living Subjects**
Eunah Kang, Hyun Su Min, Jaeyoung Lee, Moon Hee Han, Hyung Jun Ahn, In-Chan Yoon,
Kuiwon Choi, Kwangmeyoung Kim, Kinam Park, and Ick Chan Kwon*
Ultrasound (US) is a noninvasive biomedical imaging modal-
ity that is widely available, inexpensive, safe, and provides
real-time imaging and diagnosis. US contrast enhancement
improves the image quality because of acoustic impedance
mismatch, which can occur when a liquid or gas is encapsu-
lated within shell materials or when liquid or solid particles
with a high density difference are brought together.[1–3]
Microbubbles were primarily used as US contrast agents for
blood flow imaging, however, development of metabolic or
molecular US imaging that employs microbubbles has been
limited because sub-microbubbles or nanoparticles that are
appropriate for tissue penetration have poor echogenic
sensitivity, while nanoparticles can not be visualized at the
resolution of US imaging instruments (typically 50 to
100 mm).[4]
Recently, nanoparticles have been employed in numerous
biomedical applications.[5] Nanoparticles bypass the reticu-
loendothelial system, and allow prolonged circulation and
passive localization within tumor vasculature by tissue
extravasation and enhanced permeation and retention
(EPR).[6] Moreover, nanoparticles can be modified by addi-
tion of ligands that increase nanoparticle affinity toward
specific target sites, such as tumors.[7,8] In addition, the
insertion of protease cleavage sites can provide activatable
imaging probes,[9] thus allowing simultaneous molecular
imaging and therapeutics. The need to use nanosized carriers
for in vivo imaging and therapy has led to the development of
laser-induced photoacoustic US[10,11] and to the exploration of
materials-based approaches that employ perfluorocarbon-
encapsulated lipid-based nanoparticles[12] and polymeric
micelles.[13]
Herein, we describe the generation of nanobubbles, which
can be imaged by US and are derived from gas-generating
polymeric nanoparticles (GGPNPs). The proposed mecha-
nism involves localization of echogenic GGPNPs in a tumor
and coalescence of generated nanobubbles, followed by
fusion of nanobubbles into microbubbles. Significantly, we
produced GGPNP nanobubbles without encapsulation of a
gas precursor (perfluorocarbon). Instead, in our system, the
GGPNP carbonate side chain is degraded to form carbon
dioxide. The mechanism appears to involve diffusion of water
into the GGPNPs, cleavage of the carbonate side chains by
hydrolysis, and formation of carbon dioxide nanobubbles on
the GGPNP surface, followed by expansion or coalescence of
nanobubbles into microbubbles. The resulting microbubbles
exhibit resonance under a US field (Figure 1a).
Polyesters with carbonate side chains (poly(BL-PO))
were synthesized by ring-opening copolymerization of g-
butyrolactone and propylene oxide using samarium diiodide
as initiator (Figure S1b in the Supporting Information).[14,15]
The carbonate side chains were created by conjugation with
hydroxy groups of hydroxy g-butyrolactone and cholesteryl
chloroformate or ethyl chloroformate (Figure S1–S3 in the
Supporting Information). The copolymerization ratios of
cholesteryl carbonate g-butyrolactone and ethyl carbonate
g-butyrolactone were varied with respect to that of propylene
oxide. The resulting chemical structures are shown in
Figure 1b. The initial feed ratios of three monomers were
1:2:1 for poly(BL-PO). The actual ratio of three monomer
units in the poly(BL-PO) was confirmed as 1:1.74:0.81 by
1H NMR spectroscopy (see Figure S4 in the Supporting
Information). This result shows that the compositions of the
copolymers were well-matched to the initial feed ratio. The
presence of the cholesteryl group means that the relative
proportions of cholesteryl carbonate and ethyl carbonate
determine the physical stability of the polymer. The hydro-
lyzed particles produce biocompatible water, carbon dioxide,
cholesterol, and ethanol. To prove the biocompatibility of our
particles, we examined their effect on cell viability. At a
concentration of up to 1 mgmLÀ1, our particles were not
cytotoxic (see Figure S8 in the Supporting Information).
Nanoparticles generated from poly(BL-PO) (1:2:1) were
termed GGPNPs. Dynamic light scattering (DLS) indicated
that the average GGPNP diameter was 581 nm (Figure S6 in
the Supporting Information), and transmission electron
microscopy (TEM) indicated that the GGPNP morphology
was spherical and the diameter was less than 500 nm
(Figure 1c). The size appeared larger by DLS probably
[*] Dr. E. Kang,[+] H. S. Min,[+] Dr. H. J. Ahn, Dr. I. C. Yoon, Dr. K. Choi,
Dr. K. Kim, Dr. I. C. Kwon
Biomedical Research Center
Korea Institute of Science and Technology
39-1 Hawolgok-dong, Seongbuk-gu, Seoul 136 791 (Korea)
Fax: (+82)2-958-5909
E-mail: ikwon@kist.re.kr
Dr. J. Lee, Dr. M. H. Han
Department of Radiology, Seoul National University Hospital
101 Daehangno, Jongno-gu, Seoul 110-744 (Korea)
Dr. K. Park
Department of Biomedical Engineering and Pharmaceutics
Purdue University, West Lafayette, IN 47906 (USA)
[+] These authors contributed equally to this work.
[**] This work was financially supported by the Real-Time Molecular
Imaging Project, 2009K001594, the GRL Program of MEST, and by
grants from the Intramural Research Program of KIST (Theragnosis)
and the Ministry of Health Welfare and Family Affairs (A062254).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200903841.
524
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 524 –528

Angewandte
Chemie
stability of GGPNP at 378C were
observed using the transmitted
ultrasound intensity (Figure 1e),
which reflects the lifetime of the
microbubbles under US irradia-
tion.[17,18] The curve for GGPNPs
presented an initially rapid inten-
sity increase up to 12.5 min, which
indicated that initially generated
gas bubbles from GGPNPs rapidly
burst under US irradiation. How-
ever, the intensity then decreased
slightly until 41 min, thus implying
that GGPNPs may continuously and
rapidly generate a large amount of
bubbles over this time period. Inter-
estingly, GGPNPs showed a pro-
longed lifetime of the gas bubbles
(>1 h) compared to microbubbles
Figure 1. a) Illustration of gas-generating polymeric nanoparticles. b) Chemical structure of poly(BL-
PO). c) Spherical HGC-coated gas-generating polymeric nanoparticles of poly(BL-PO) (1:2:1). d) The
amount of carbon dioxide from poly(BL-PO) nanoparticles generated after 5 min exposure. e) The
transmitted ultrasound intensity of PLGA NP and GGPNP (Io is the ultrasound intensity of PBS) as a
function of time at 378C.
with
a
phospholipid
shell
(ca. 5 min).[17] As expected, the con-
trol poly(lactide-co-glycolic acid)
(PLGA) NPs do not present any
change of the transmitted ultrasound
intensity under US irradiation.
because bubbles were produced on the GGPNP surfaces.
GGPNPs of the appropriate diameter were prepared so that
they might accumulate in tumors by EPR. The hydrophobic
surface was rendered passive using hydrophobically modified
glycol chitosan (HGC) with a 17% substitution ratio of
cholanic acid as a surfactant. It is presumed that the positive
charges on the HGC units provided a tumor affinity, so that
the nanoparticles could accumulate therein.[15] We used mass
spectrometry to measure the generation of carbon dioxide
from GGPNPs following their dispersal in saline for 5 min (to
allow hydrolysis). Carbon dioxide release was then measured
by monitoring the peak at 44 Da (Figure S7 in the Supporting
Information). Based on calibrations of volume and pressure,
the amount of residual carbon dioxide present in saline
(0.025 ccatm) and the amount of GGPNP-generated carbon
dioxide in saline were quan-
To investigate the hydrolysis mechanism of GGPNPs in
PBS at 378C, we measured the changes of particle size and
morphology by using DLS and TEM. From the DLS
measurements, GGPNPs showed an increased particle size
of about 640 nm at 1 min post-incubation (Figure 2a) com-
pared to the size of 200–500 nm for freshly prepared GGPNPs
(Figure 1c), thus indicating that bubbles were rapidly pro-
duced on the GGPNP surfaces (Figure 2c). At 30 min post-
incubation, a large number of microbubbles (> 1 mm) were
observed, thus indicating that the hydrolyzed GGPNPs
produced microsized bubbles. As expected, the TEM micro-
graphs showed some surface erosion at 1 min post-incubation
of GGPNPs in PBS at 378C, and this surface erosion
increased with incubation times up to 30 min. Importantly,
after 2 h post-incubation, most of the polymers in the core
tified by using mass spec-
trometry (Figure 1d). The
latter levels were 0.048,
0.052, and 0.074 ccatm for
copolymerization ratios of
cholesteryl:ethyl carbonate
of 0.5:2.5, 1:2, and 3:1,
respectively.
Thus,
the
amount of GGPNP-gener-
ated carbon dioxide was
1.92, 2.08, and 2.96 times
greater than that of intrinsic
carbon dioxide. In vitro and
in vivo characterization of
GGPNPs as a US contrast
agent employed poly(BL-
PO) (1:2:1). The acoustic
Figure 2. a) Size distribution of GGPNPs incubated in PBS at 378C for 30 min measured using DLS. b) TEM
micrographs of GGPNPs hydrolyzed in PBS for 2 h at 378C. c) Optical micrograms of gas-generating profiles
of microsized GGPNPs (6–8 mm) incubated in PBS at 378C for 1 h (arrows indicate gas-generating bubbles).
characteristic and bubble Original magnification is ꢀ400.
Angew. Chem. Int. Ed. 2010, 49, 524 –528
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
525

Communications
were degraded: the hydrolysis reaction might be auto-
accelerated because of the higher concentration of CO2
(Figure 2b). Finally, we directly visualized the gas generation
profile of hydrolyzed GGPNPs by using optical microscopy.
To visualize the direct gas formation, we used gas-generating
microparticles with an average size of about 6–8 mm, as the
nanosized GGPNPs were not observed using optical micros-
copy, but the microparticles had the same chemical compo-
sition and molecular weight of the GGPNPs. Interestingly,
within 1 min post-incubation in PBS, many small bubbles
(< micrometer) were observed on the particle surface. As the
incubation time increased up to 1 h, the small bubbles formed
larger bubbles (> micrometer), because of the expansion or
coalescence of nanobubbles into microbubbles. These gas-
generation profile data were closely related to the hydrolysis
pattern of gas-generating polymeric particles.
We performed in vitro comparisons of US contrast agents
consisting of GGPNPs and (control) PLGA nanoparticles
employing US imaging equipment with 2D and cadence pulse
sequence imaging modes, and a 15 MHz linear transducer
(Figure 3a). The PLGA nanoparticles (negative control)
Figure 3. In vitro characterization of GGPNP in aqueous conditions.
a) US images of GGPNP dispersion at 2D and cadence (CA) mode.
b) Histogram of US intensity from PLGA and GGPNP dispersions. The
image was taken at under the same condition of gain=10/MI=0.5. *
indicates significance level of p<0.05. MI=mechanical index.
Figure 4. Ultrasound images aided by contrast agent administered by
subcutaneous injection. All images were taken at cadence mode under
the same condition of MI=0.5/gain=10. After subcutaneous injec-
tion, US images taken as a function of a) concentration and b) time at
a concentration of 3 w/v%. Intensity profiles were presented as a
function of c) concentration and d) time.
showed no acoustic reflectivity contrast under either mode
because the resolution of US imaging is approximately 50–
100 mm and the PLGA nanoparticles, which do not form
bubbles, were not echogenic. In contrast, US contrast images
of GGPNP dispersions were significantly enhanced in both
modes (p < 0.05; Figure 3b). The contrast enhancement that
arises from GGPNP dispersion in the cadence mode indicated
that the presence of microbubbles or the coalescence of
carbon dioxide nanobubbles was responsible for US resona-
tion.
We subsequently investigated the feasibility of using
GGPNPs as an in vivo US contrast agent. We first adminis-
tered GGPNP dispersions (0.5–5.0% w/v) subcutaneously
into the lower backs of BALB/c mice. US images were taken
immediately after injection and the change in US contrast was
observed over time by cadence pulse sequencing and 2D
modes (Figure 4a). The results clearly showed that the
intensity increased as a function of GGPNP concentration
(Figure 4c). Under the same conditions of gain, mechanical
index, and power, the US intensity of the region of interest
was normalized to the background intensity of ultrasound gel
on the skin. The local area where the saline control was
injected showed a clear margin with no contrast signal in the
cadence mode, thus indicating the absence of acoustically
reflective bubbles. However, a very small US contrast
enhancement by GGPNPs could be seen at a GGPNP
concentration of 0.5% w/v, which is presumably near the
threshold of signal detection. As the GGPNP concentration
increased, the nanobubbles coalesced or merged into micro-
bubbles and the signal became stronger. We also monitored
the change in the US signal over time (Figure 4b). The US
signal initially increased, maintained a plateau for approx-
imately 10 min, and then gradually decreased (Figure 4d).
The contrast enhancement of the cadence and 2D modes was
526
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 524 –528

Angewandte
Chemie
propylene oxide were purchased from Sigma Chemical Co. (St. Louis,
MO).
Polymerization of poly(butyrolactone-co-propylene oxide) with a
visible for 30 min, which should be sufficient to allow for
in vivo real-time imaging. Notably, our results show that the
cadence-mode US signal was caused by bubble resonance,
and not by enhancement of 2D back-scattering. The presence
of a signal in the cadence mode clearly indicates that the
dispersion of polymeric nanoparticles generated coalesced
nanobubbles of carbon dioxide and that the resulting micro-
bubbles vibrated under the applied US field.
To demonstrate US imaging of a tumor, we administered
an intratumoral injection of 3% w/v GGPNP dispersion to
tumor-xenograft-bearing BALB/c mice and compared US
images pre- and post-injection (Figure 5). The US intensity at
pendant cholesteryl/ethyl carbonate group: Detailed procedures are
described in the Supporting Information. Briefly, a-hydroxy-g-
butyrolactone and either cholesteryl chloroformate or ethyl chloro-
formate were dissolved in dichloromethane. The mixture was placed
in an ice bath and pyridine was added dropwise under stirring. After
20 h at room temperature, the solution was washed with 1m HCl,
saturated NaHCO3, and distilled water, and precipitated by addition
of cold ethyl ether. The resulting white crude powder was isolated and
dried under vacuum. The yield was 85.4%.
Polymerization of poly(butyrolactone-co-propylene oxide) with a
pendant cholesteryl/ethyl carbonate group was performed by initia-
tion with samarium diiodide/samarium (SmI2/Sm). Different mono-
mer ratios of g-butyrolactone with a cholesteryl carbonate or ethyl
carbonate pendant group and propylene oxide (PO) were dissolved in
toluene. Following addition of SmI2/Sm, the mixture was stirred at
1008C under reduced pressure (0.1Torr). After 48 h, the product was
precipitated by addition of cold ethyl ether and filtered. The resulting
crude powder was dried under vacuum. The yield of poly(BL-PO)
was 68.23%.
Particle preparation and morphology: Nanoparticles were syn-
thesized by emulsification of the oil-in-water phase using a probe
sonicator. An HGC solution (0.2% w/v) was used as the water phase
(10 mL of fluid). A polymer solution (6% w/v) in dichloromethane
(1 mL) was added dropwise into the water phase. After sonication,
the particle dispersion was stirred for 2 h to remove residual solvent.
After centrifugation, particles were redispersed in Tween 20 (0.5%
w/v) and maltose (5% w/v; disintegration agent in reconstitution) and
freeze-dried. The dried powder was resuspended at the desired
concentration prior to use. The particle morphology and diameter
were determined using DLS (Spectra Physics, Mountain View, CA)
and TEM (CM-200, Philips, CA).
The carbon dioxide content of GGPNPs were measured using a
quadrupole mass spectrometer (Prisma QME 200, Germany)
equipped with a Faraday cup detector. The vapor was analyzed with
an emission current of 1.5 mA, an electron energy of 10 eV, and a
resolution of 50. Saline (3 mL) and GGPNPs (20 mg) were placed in a
gas-tight syringe. As a control, dissolved gas species in saline alone
were measured. The gas generated from GGPNPs after hydrolysis
was quantified after 5 min exposure to saline.
In vitro acoustic measurements were performed by detecting
attenuation signals of GGPNPs exposed to a focused ultrasound pulse
(10 MHz), as described previously.[18] Briefly, the dispersed GGPNPs
in phosphate-buffered saline (PBS; 3%, w/v) were placed in a 50 mL
acryl chamber located between an immersing type 10 MHz trans-
ducer (12.7 mm diameter, V311, Panametrics Inc., Waltham, MA)
and a home-made metallic reflector in a 378C water bath with gentle
stirring. This experiment was repeated 5 times. A pulse of ultrasound
was sent to the sample chamber and then the reflected pulse was
observed using a pulser/receiver. The reflected pulse was converted to
radiofrequency (RF) signals using a digital oscilloscope (Ultrawave-
2020, MKC Korea Inc., Korea). The RF signals of GGPNPs were
processed by fast Fourier transform (FFT) and filtered using the
LabView program (National Instruments Corp., TX). Attenuation
spectra were obtained by dividing the GGPNPs values in the vicinity
of the FFT maxima by the values of pure PBS under the same
conditions.
Figure 5. Contrast enhancement of a tumor before (pre) and after
(post) intratumoral injection. a) US images taken at both 2D and
cadence mode. b) Histogram of normalized intensities derived from
the images. * indicate a significance level of p<0.05.
the tumor site was weak compared with that seen after
subcutaneous injection because of the rapid leakage of
GGPNPs into blood vessels after injection. However, imme-
diately after injection and for the following 1 min, we
observed localized spots of enhanced contrast in the tumor
by the cadence mode. A comparison of contrast enhancement
in the 2D and cadence modes indicated a significant intensity
difference of in the cadence mode pre- and post-injection (p <
0.05). The US intensity in the cadence mode increased by
58% after intratumoral administration, whereas the intensity
in the 2D mode was not significantly enhanced. The
appearance of local spots of enhanced contrast suggests that
adequate accumulation of GGPNPs occurred at the tumor
site, thus allowing US imaging. Therefore, the GGPNPs
penetrated the tumor interstitium, coalesced to form nano-
bubbles, and provided a vibrating reactivity visible under the
cadence mode of US. As the resolution of our US instrument
is approximately 100 mm, we could not resolve nanobubbles.
The localized spots clearly visible in the tumor indicated that
nanobubbles generated from gas-generating nanoparticles
had coalesced into microbubbles. Our results indicate that
GGPNPs can generate nanobubbles that coalesce into micro-
bubbles, which resonate under a US field. GGPNPs have
great potential as contrast agents that can facilitate in vivo US
imaging.
Hydrolysis test: GGPNPs (1 mgmLÀ1) were dissolved in PBS and
incubated 2 h at 378C. Under these hydrolysis conditions, the changes
in particle size and morphology were characterized using DLS and
TEM according to the incubation time. To directly visualize the gas-
bubbling profile of GGPNPs, we made gas-generating microparticles
with an average size of about 6–8 mm using a homogenizer with
15000 rpm. The microparticles have the same chemical composition
and molecular weight of GGPNPs. After incubating microparticles in
Experimental Section
Materials: Samarium, 1, 2-diiodoethane, sodium thiosulfate solution,
pyridine, a-hydroxy-g-butyrolactone, cholesteryl chloroformate, and
Angew. Chem. Int. Ed. 2010, 49, 524 –528
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
527

Communications
PBS, the time-dependent gas-generating profile was monitored in situ
using optical microscope equipped with 40 ꢀ focal lens (BX51;
Olympus Co. Ltd., Japan).
[2] K. Ferrara, R. Pollard, M. Borden, Annu. Rev. Biomed. Eng.
2007, 9, 415 – 447.
[3] E. G. Schutt, D. H. Klein, R. M. Mattrey, J. G. Riess, Angew.
Chem. 2003, 115, 3336 – 3355; Angew. Chem. Int. Ed. 2003, 42,
3218 – 3235.
[4] J. R. Lindner, Nat. Rev. Drug Discovery 2004, 3, 527 – 532.
[5] G. S. Kwon, M. Naito, M. Yokoyama, T. Okano, Y. Sakurai, K.
Kataoka, Pharm. Res. 1995, 12, 192 – 195.
[6] Y. Takakura, R. I. Mahato, M. Hashida, Adv. Drug Delivery Rev.
1998, 34, 93 – 108.
[7] S. Keren, C. Zavaleta, Z. Cheng, A. de La Gheysens, S. S.
Gambhir, Proc. Natl. Acad. Sci. USA 2008, 105, 5844 – 5849.
[8] G. E. Weller, M. K. Wong, R. A. Modzelewski, E . Lu, A. L.
Klibanov, W. R. Wagner, F. S. Villanueva, Cancer Res. 2005, 65,
533 – 539.
[9] R. Weissleder, C. H. Tung, U. Mahmood, A. Jr Bogdanov, Nat.
Biotechnol. 1999, 17, 375 – 378.
[10] K. H. Song, C. Kim, C. M. Cobley, Y. Xia, L. V. Wang, Nano Lett.
2009, 9, 183 – 188.
[11] X. Wang, Y. Pang, G. Ku, X. Xie, G. Stoica, L. V. Wang, Nat.
Biotechnol. 2003, 21, 803 – 806.
[12] F. S. Villanueva, W. R. Wagner, Nat. Clin. Pract. Cardiovasc.
Med. 2008, 5 Suppl 2, S26 – 32.
[13] N. Rapoport, Z. G. Gao, A. Kennedy, J. Natl. Cancer Inst. 2007,
99, 1095 – 1106.
[14] S. Agarwal, X. Xie, Macromolecules 2003, 36, 3545 – 3549.
[15] M. Nishiura, Z. Hou, T. A. Koizumi, T. Imamoto, Y. Wakatsuki,
Macromolecules 1999, 32, 8245 – 8251.
In vitro US imaging was performed in the static state. Eppendorf
tubes were embedded in agarose gel (3%, w/v) and GGPNP
dispersions (0–5%, w/v) were placed into the tubes. As negative
controls, US contrast agents consisting of PLGA nanoparticle
dispersions at the same concentrations were employed. US was
performed in the 2D and cadence pulse sequencing (CPS) modes. US
imaging (Acuson Sequoia 512 system; Siemens Medical Solutions,
Malvern, PA) was achieved using a 15 MHz probe (Acuson model
15 L8). Statistical analysis employed the Tukey method of one-way
analysis of variance (ANOVA). A p value less than 0.05 was
considered significant.
In vivo ultrasound imaging: All animal experiments were per-
formed according to the Guidelines for Care and Use of Research
Animals developed by the Seoul National University Animal Study
Committee. Male BALB/c mice (n = 3, 25–30 g body weight) less than
6 weeks old were anesthetized using isofluorane gas (2–3%, v/v).
Lower back hair was removed with depilatory cream and each animal
received a subcutaneous inoculation of 1 ꢀ 106 squamous cell carci-
noma (SCC7) cells in the lower back. Cells were obtained from the
American Type Culture Collection (ATCC) and suspended in 50 mL
of medium. Xenograft tumor-bearing mice with tumor volumes of
approximately 1500 mm3 (calculated by [ab2]/2) were given intra-
tumoral injections of a GGPNP dispersion (3%, w/v) at four sites
around the tumor. The total received dose was 100 mL.
[16] K. Park, J. H. Kim, Y. S. Nam, S. Lee, H. Y. Nam, K. Kim, J. H.
Park, I. S. Kim, K. Choi, S. Y. Kim, I. C. Kwon, J. Controlled
Release 2007, 122, 305 – 314.
Received: July 14, 2009
Revised: November 3, 2009
Published online: December 9, 2009
[17] S. Rossi, G. Waton, M. P. Krafft, ChemPhysChem 2008, 9, 1982 –
1985.
[18] F. Gerber, M. P. Krafft, G. Waton, T. F. Vandamme, New J.
Chem. 2006, 30, 524 – 527.
Keywords: carbon dioxide · imaging agents · nanostructures ·
polymers · ultrasound
.
[1] S. P. Qin, C. F. Caskey, K. W. Ferrara, Phys. Med. Biol. 2009, 54,
R27 – R57.
528
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 524 –528