Ultrasound-enhanced Chemotherapy and Gene Delivery for Glioma

Document Sample
Ultrasound-enhanced Chemotherapy and Gene Delivery for Glioma Powered By Docstoc
					                                                                                                      Technology in Cancer Research and Treatment
                                                                                                      ISSN 1533-0346
                                                                                                      Volume 6, Number 5, October 2007
                                                                                                      ©Adenine Press (2007)

                      Ultrasound-enhanced Chemotherapy and                                            Vladimir G. Zarnitsyn, Ph.D.
                               Gene Delivery for Glioma Cells                                         Pavel P. Kamaev, Ph.D.
                                                                                                      Mark R. Prausnitz, Ph.D.*

Treatment of brain cancer is limited in part by inefficient intracellular delivery of drugs and       School of Chemical and
DNA for chemotherapy and gene therapy, respectively. This study tested the hypothesis                 Biomolecular Engineering
that ultrasound may be used to enhance intracellular delivery and efficacy of chemothera-
                                                                                                      Georgia Institute of Technology
peutics and genes in glioma cells in vitro. First, suitable ultrasound conditions were identi-
fied by measuring intracellular uptake of calcein and viability of GS 9L rat gliosarcoma cells        315 Ferst Drive
after a range of different ultrasound exposures. We selected sonication at 10 J/cm2, which            Atlanta, GA 30332-0100, USA
achieved intracellular delivery of ~106 molecules/cell. Next, glial cells were sonicated with
varying concentrations of model chemotherapeutics: BCNU and bleomycin. For both drugs,
cytotoxicity was increased in a synergistic manner when accompanied by ultrasound expo-
sure. Finally, expression of a plasmid DNA encoding a GFP reporter was increased up to
30-fold when exposed to ultrasound. Altogether, these findings suggest that ultrasound may
be useful to increase the efficacy of chemotherapy and gene therapy of glioma cells.

Key words: Chemotherapy; Cytotoxicity; DNA transfection; Gene therapy; Gliosarcoma
brain cancer; and Ultrasound.


Malignant brain cancer afflicts more than 80,000 Americans and many more
worldwide. In the United States, close to 18,000 new cases are diagnosed and
13,000 patients die each year (1-3). Treatment by surgery is preferred if the
tumor is accessable. If not, radiation is often used. Chemotherapy has gener-
ally been found to be less effective against brain cancer, because gliomas are
relatively slow-dividing tumors and the blood brain barrier hampers delivery of
chemotherapeutics to the brain. Gene therapy may provide a new way to treat
cancer, including suicide gene therapy (4), downregulation of oncogenes (5),
and antiangiogenic therapy (6).

In all of these treatment options, one of the critical limitations of current therapy
is insufficient targeting to cancer cells (7). In recent years, BCNU biodegradable
wafers (Gliadel) implaneted locally into brain have been introduced as a targeted
therapy, which has been shown to increase life expectancy (8, 9). Despite mod-
erate clinical success of Gliadel wafers, the prognosis for brain cancer patients
generally remains poor and overall treatment of brain cancer, especially for high-
grade gliomas, needs significant improvement (10).

This study addresses the hypothesis that ultrasound may be used to enhance
intracellular delivery and efficacy of chemotherapeutics and genes in glioma                      Corresponding Author:

cells. We and others have previously shown that under appropriate conditions                      Mark R. Prausnitz, Ph.D.
ultrasound can drive large numbers of molecules into cells (11). In vitro studies                 Email: prausnitz@gatech.edu

434	                                                                                                            Zarnitsyn	et	al.

have demonstrated that ultrasound’s ability to transport mol-         plasma membrane can be torn open to produce a hole of up
ecules into cells appears to be ubiquitous to many different          to 1 μm in size (24, 25). Molecules can transport through
cell types, including cancer cell lines, primary cells, yeast         this hole to gain access to the cytosol. By a mechanism
(12), and bacteria (13), and a variety of different classes of        involving transiently elevated intracellular calcium levels,
molecules, including drugs, proteins, and plasmid DNA (14).           cells traffic intracellular vesicles to the site of injury and ac-
Of direct relevance to this study, ultrasound has been shown          tively patch the hole on a timescale of a few minutes. Under
to enhance transfection of rat (9L) and canine (J3T) glioma           appropriate conditions, cells survive this process, retaining
cells during exposure to complexes of DNA and cationic lipo-          large numbers of intracellular molecules and appearing to
somes consisting of DOSPA/DOPE (15). It was also shown                regain normal function within hours. In other cases, cellular
that ultrasound in the presence of microbubbles can enhance           damage is too great and cells die by mechanisms that can
gene transfer in a variety of different cells types (16).             involve necrosis or programmed death.

In vivo studies have demonstrated enhanced and targeted               The effects of ultrasound on a population of cells are typi-
cancer chemotherapy and gene delivery/expression with the             cally heterogeneous (26). After exposing a cell suspension to
use of ultrasound in animals in more than a dozen different           “mild” ultrasound conditions, almost all cells usually remain
studies in a variety of different tissues (but not yet in brain tu-   viable and a small fraction (e.g., 10%) show intracellular up-
mors) (17). In related clinical trials, transcranial ultrasound       take of molecules. The remaining cells appear to be gener-
was shown to enhance transport for thrombolysis in acute              ally unaffected. After stronger ultrasound exposures, larger
ischemic stroke, demonstrating the clinical feasibility of ul-        fractions of cells can show uptake (e.g., 10-40%), but cell
trasound-based therapy of the brain (18).                             viability correspondingly drops. At very strong ultrasound
                                                                      conditions, essentially all cells can be killed. This hetero-
Ultrasound-mediated enhancement of chemotherapy has also              geneity can be explained by the observation that cavitation
received attention in a limited number of studies. For ex-            bubble activity occurs at discrete times and locations (27,
ample, sonication of chinese hamster ovary cells (HA1) and            28). Cells that are near a cavitation bubble upon, for ex-
murine tumors in the presence of adriamycin was shown to              ample, its transient collapse, should be impacted more than
produce a synergistic cytotoxic effect (19, 20), which is be-         cells located some distance away. In this way, a gradient of
lieved to be caused by increased intracellular uptake of the          effects can be observed based on the distance that each cell is
chemotherapeutic molecules. In a different study addressing           located from the nearest bubble.
“sonodynamic therapy,” certain drugs were shown to be ac-
tivated by exposure to ultrasound by a mechanism believed             Guided by this understanding of ultrasound mechanisms and
to involve ultrasound-mediated production of highly active            previous successes with ultrasound for enhanced drug ther-
radicals that increase cytotoxicity (21).                             apy, this study sought to determine if ultrasound could en-
                                                                      hance cytotoxicity of chemotherapeutic drugs for brain can-
Although ultrasound is commonly used in medicine for other            cer. Specifically, we looked for synergistic effects between
applications, the ultrasound conditions needed to observe             ultrasound and exposure to BCNU (carmustine), which is
these therapeutic effects are different from those used, for          an established chemotherapeutic for local treatment of brain
example, in diagnostic imaging or tissue heating for physi-           cancer (8), and bleomycin, which is not typically used for
cal therapy (22). Enhanced drug and gene delivery usually             brain cancer, but its efficacy is known to be limited by poor
requires application of ultrasound at a high mechanical index         intracellular uptake (29), which could be improved by ul-
(MI) and/or with the combined use of contrast agents, which           trasound. We addressed these questions using a 9L rat glio-
serve as nucleation sites for cavitation. Conventional clinical       sarcoma model in vitro. To our knowledge, this is the first
applications of ultrasound seek to avoid cavitation, which is         study to examine ultrasound’s effects on chemotherapy of
the creation, oscillation and, in some cases, collapse of gas         glial cells. A related study showed that ultrasound increased
bubbles in the alternating pressure field generated by ultra-         blood-brain barrier permeability to improve Herceptin de-
sound. Cavitation can have damaging effects on cells and              livery into the brain (30).
tissues, such as the shattering of kidney stones by cavitation
generated during lithotripsy. However, under controlled con-          This study also sought to assess the ability of ultrasound to
ditions, the mechanical effects of cavitation can be harnessed        facilitate gene transfer and expression in glioma cells. Nu-
to drive molecules into living cells, as studied here.                cleic acid therapeutics offer an exciting opportunity to effi-
                                                                      ciently target therapy to cancer cells, but require intracellular
The mechanism of intracellular delivery using ultrasound is           delivery of genetic material, which is notoriously inefficient
believed to be initiated by local fluid dynamics generated            using nonviral methods. In recent studies, it was shown that
by oscillating and/or collapsing cavitation bubbles (23). By          gene transfection of prostate cancer cells may be dramatical-
either shearing or direct impact of fluid with the cell, its          ly increased using optimized ultrasound protocols (31, 32).

Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
Ultrasound-enhanced	Chemotherapy	                                                                                             435

This study builds off those observations to partially optimize       described in detail previously (35, 36). Briefly, 1.1 MHz
transfection of gliosarcoma cells.                                   ultrasound was produced using an immersible, focused,
                                                                     piezoceramic transducer (model H-101, Sonic Concepts,
Materials and Methods                                                Woodinville, WA). The transducer has a 70 mm diameter,
                                                                     a 52 mm focal length, and a 1.5 mm focal width at half
Cell Culture                                                         amplitude (-6 dB).

Rat gliosarcoma cells GS 9L (courtesy of John Hopkins Uni-           A pulsed rectangular signal was produced by a first program-
versity) were cultured as monolayers in a humidified atmo-           mable waveform generator (model DS345, Stanford Re-
sphere of 95% air and 5% CO2 at 37 ºC (Incubator model               search Instruments, Sunnyvale, CA) and modulated by a 1.1
3110, Forma Scientific, Marietta, OH) in RPMI-1640 media             MHz waveform from a second waveform generator (model
supplemented with 10% (v/v) heat-inactivated fetal bovine            33120A, Agilent, Austin, TX). This approach let us control
serum (FBS) and 100 μg/ml penicillin-streptomycin (Invitro-          the number of pulses, pulse length, frequency and peak-to-
gen, Carlsbad, CA) (33). After reaching 80-90% confluence,           peak voltage. The waveform was then amplified by an RF
cells were harvested under sterile conditions prior to each ex-      broadband power amplifier (model 3100LA, Electronic Nav-
periment by trypsin digestion for 2 min (0.25% trypsin, 0.1%         igation Industries, Rochester, NY) before passing through
EDTA, Cellgro, Herndon, VA). Trypsin was inactivated by              an impedance matching network (model H-101, Sonic Con-
addition of five parts full media to one part trypsin. Cells were    cepts) and applied to the transducer.
then centrifuged (1000 × g, 6 min, model GS-15R, Beckman
Coulter, Fullerton, CA) and resuspended in RPMI-1640 me-             The transducer was housed in a polycarbonate tank (30.5 ×
dia at a concentration of 106 cell/ml, as determined by a Coul-      29 × 37 cm) containing approximately 26 L of deionized, dis-
ter Multisizer II (Beckman Coulter), unless otherwise noted.         tilled, and partially degassed water at room temperature (21-
                                                                     23 ºC). A 5 cm thick acoustic absorber (acoustic rubber SC-
BCNU, Bleomycin, DNA, and Calcein                                    501, Sonic Concepts) was mounted opposite the transducer
                                                                     to minimize standing wave formation. A three-axis position-
Concentrated stock solutions of each chemotherapeutic were           ing system (10 μm resolution, Velmex, Bloomfield, NY) was
first created. 1,3-Bis(2-chloroethyl)-1-nitrosourea (also            mounted on top of the tank to position samples and a hydro-
known as BCNU or carmustine, Sigma Chemical, St. Louis,              phone, discussed below, at desired locations in the tank.
MO) was dissolved in 95% ethanol immediately prior to use.
Bleomycin (containing a mixture of bleomycin sulphate B2             To map and calibrate the acoustic field produced by the trans-
and A2, Gensia Sicor Pharmaceuticals, Irvine, CA) was dis-           ducer versus the peak-to-peak voltage signal provided by the
solved in sterilized DI water prior to use. The stock solu-          function generator, a PVDF membrane hydrophone (model
tions were diluted into the culture media to concentrations of       MHA200A, NTR Systems, Seattle, WA) was used to mea-
0-1 mM BCNU or 0-100 μM bleomycin immediately before                 sure spatial-peak-temporal-peak negative pressure.
application of ultrasound. After addition of BCNU, the final
concentration of ethanol in the cell suspension culture was          Ultrasound Treatment
kept below 1:1000 (v/v), which has been shown to be non-
toxic for cells (34).                                                Before each experiment, the hydrophone was used to position
                                                                     the cell sample approximately 1 cm out of the ultrasound’s
For transfection experiments, a concentrated stock solution          focus toward the transducer on the main axis of the transduc-
of DNA plasmid coding for green fluorescent protein (GFP)            er. The acoustic pressure was calibrated versus the peak-to-
(gwiz pGFP (5,757 bp), Aldevron, Fargo, ND) was added to             peak voltage of the signal created by the function generator
cell suspensions immediately prior to application of ultra-          using the hydrophone at the desired location. This out-of-fo-
sound at a final concentration of 0-30 μg/ml.                        cus location was used because it had a broader acoustic beam
                                                                     than at the focus, which was approximately 10 mm wide at
Calcein, which is a green fluorescent molecule that cannot           half amplitude (-6 dB). This broader acoustic beam enabled
cross intact cell membranes, was used to quantify molecular          a more uniform acoustic exposure across the sample. Vigor-
uptake into viable cells. Calcein (Molecular Probes, Eugene,         ous mixing caused by cavitation during ultrasound exposure
OR) was added to cell suspensions at a concentration of 10           further facilitated a uniform average exposure of the sample.
μM before application of ultrasound.
                                                                     To promote cavitation, albumin-stabilized perfluoropro-
Ultrasound Apparatus                                                 pane-gas bubbles (Optison, Mallinckrodt, St. Louis, MO)
                                                                     were slowly added to the cell suspension in all experiments
The ultrasound apparatus and its characterization have been          using a 22-gauge flat needle (17 μl/ml, approximately 1.1

                                                         Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
436	                                                                                                        Zarnitsyn	et	al.

× 107 bubbles/ml, with average diameter of 2.0-4.5 μm, ac-         We also added 2.5 μm fluorescent beads (LinearFlow Green
cording to the manufacturer).                                      Flow Cytometry Intensity Calibration Kit L-14821, Molecular
                                                                   Probes) to cell samples at a concentration of 105 beads/ml for
The sample chamber was a 0.4 ml polyethylene transfer pi-          volumetric calibration, as previously described (37). Briefly,
pette (catalog no. 293, Samco Scientific, San Fernando, CA),       cells that were destroyed or otherwise lost during the protocol
from which the bottom tab was removed and filled with sili-        were identified by comparing the ratio of cells-to-fluorescent
cone glue (Silicone II, GE Sealants & Adhesives, Hunters-          beads in each sample to that of control samples. Especially
ville, NC) several days prior to the experiment. The upper         after extended incubation (i.e., for five days in the context of
stem was cut to 4 mm length. Chambers were sterilized by           chemotherapeutic experiments), this measure of cell viability
placing under UV light in a biological safety cabinet for 30       accounted for possible cell death due to the treatment, as well
min prior to each experiment. Each chamber was filled with         as decreased cell concentration due to impaired cell division.
cell suspension up to the top of the pipette stem and an alu-
minum rod of 3.5 mm diameter was inserted 2 mm into the            For each sample, 20,000 viable cells were analyzed by flow
stem. Overflow of solution was allowed to ensure no visible        cytometry (BD LSR Flow Cytometer, Becton Dickinson, San
air bubbles were trapped in the sample.                            Jose, CA), as described previously (33), which determined
                                                                   the concentration of viable and dead cells in each sample.
Ultrasound exposures at 1.1 MHz were performed at a burst          For DNA transfection and calcein uptake experiments, the
length of 1 ms, duty cycle of 1% (i.e., 10 pulses per sec-         intensity of green fluorescence (530/28 nm bandpass filter)
ond), peak negative pressures of 1.2, 1.7, 2.0, 2.5, 3.0, and      was measured in viable cells. Cells were counted as GFP-
3.5 MPa, and exposure times of 3, 10, 30, 100, 300, 1000,          expressing cells if they had a green fluorescence greater than
and 3000 ms. Note that exposure time refers to the amount          that of 99.9% of cells in control samples. Transfection data
of time ultrasound was actively applied. Given the 1% duty         are reported as the fraction of cells expressing GFP among
cycle, the length of each experiment ranged from 300 ms to         all cells exposed to ultrasound, which provides the most rig-
300 s. This low duty cycle was used to minimize heating,           orous assessment of transfection efficiency. This contrasts
which was < 1 ºC for all conditions studied.                       with some other methods used in the literature, in which
                                                                   transfection efficiencies have been reported as a fraction of
For each experiment, at least three samples were not exposed       cells remaining viable at the time of analysis (38, 39).
to ultrasound, but were subjected to all other procedures.
These "sham" samples were used as negative controls for vi-        Intracellular uptake of calcein was quantified by converting
ability and transfection measurements.                             fluorescence measurements into absolute numbers of calcein
                                                                   molecules delivered into each cell by calibration relative to
After ultrasound application, cells were incubated at 37 ºC        a panel of fluorescent microspheres of known fluorescence
for 5 min to allow cell recovery. Cell samples were then           (Quantum MESF FITC high level, Bangs Laboratories,
washed once by centrifugation (1000 × g, 6 min, Eppendorf          Fishers, IN) (37).
5415 C, Brinkmann Instruments, Westbury, NY), resuspend-
ed in full growth medium (2 ml for each sample) and placed         Data Analysis
into six-well cell culture plates (Corning Inc., Corning, NY)
maintained in the cell culture incubator under cell growth         Intracellular calcein uptake is reported as the average number
conditions for five days when assessing BCNU or bleomy-            of molecules per cell by calibrating fluorescence measure-
cin cytotoxicity and 24 h when assessing DNA transfection.         ments relative to the calibration microspheres mentioned
When calcein uptake and cell viability were determined, sam-       above. Cell viability is calculated as the percent of viable
ples were not place in the incubator, but were washed three        cells compared to sham-exposed cells in the absence of
times and maintained on ice for up to 1 h before analysis.         chemotherapeutic. However, for samples exposed to both
                                                                   ultrasound and chemotherapeutic, it is of special interest to
Transfection, Uptake, and Viability Measurements                   identify synergistic effects of ultrasound and chemotherapeu-
                                                                   tics. For these dual exposures, viability measurements were
Expression of GFP, uptake of calcein, and cell viability were      corrected to subtract out the additive effects of ultrasound
quantified by flow cytometry. Cells were harvested from six-       alone and report only the synergistically increased chemo-
well plates by trypsinization and centrifugation, as described     therapeutic cytotoxicity induced by exposure to ultrasound.
above, and then suspended in phosphate-buffered saline             Thus, under the conditions used during experiments to assess
(PBS, Cellgro). Propidium iodide (Molecular Probes) was            BCNU cytotoxicity, ultrasound alone caused a 30% loss of
added to cell suspensions at a final concentration of 1 μg/ml      viability (i.e., viability was 70%). To subtract out the effects
to label non-viable cells. After 10 min incubation at room         of ultrasound alone, the reported viability was corrected by
temperature, cells were placed on ice until analysis.              multiplying by the factor 1.0/0.7 = 1.43. Similarly, under

Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
Ultrasound-enhanced	Chemotherapy	                                                                                                                       437

the conditions used during experiments to assess bleomycin          uptake exceeded 106 molecules per cell. Dividing this num-
cytotoxicity, ultrasound alone caused a 60% loss of viability       ber of molecules by the average cell volume of 1600 μm3
(i.e., viability was 40%). To subtract out the effects of ultra-    (determined by a Beckman-Coulter Multisizer II) yields an
sound alone, the viability was corrected by multiplying by          intracellular calcein concentration on the order of 10% of
the factor 1.0/0.4 = 2.5. Although the same ultrasound con-         the extracellular concentration, which indicates extensive
ditions were used, cell viability from ultrasound alone was         uptake. These results are similar to those reported previ-
higher in the BCNU study than in the bleomycin study. This          ously for other cell types (11).
is probably an artifact created because we used a new batch
of Optison for the bleomycin study, which probably showed           Achieving intracellular uptake in some cells was often as-
more cavitation activity and thereby caused more cell death.        sociated with loss of viability in other cells. As shown in
                                                                    Figure 1B, cell viability decreased with increasing ultra-
After exposure to DNA, transfection efficiency is reported (i)      sound energy dose. At the weakest conditions (< 1.7 J/
on a relative basis as the number of GFP-positive cells after       cm2), cell viability did not change significantly from 100%
ultrasound exposure relative to sham-exposed cells incubat-         (ANOVA, p = 0.11). However, at larger energies, cell vi-
ed in the same concentration of DNA and (ii) on an absolute         ability dropped to 75-80% viability at intermediate energies
basis as the GFP-positive cells as a percent of all exposed         and down to 20-30% viability at the strongest conditions
cells. Other measurements determined the amount of GFP              used (ANOVA, p < 0.0001 for E > 200 J/cm2).
expressed by cells, which is reported as the average fluores-
cence (530/20 nm bandpass filter) among GFP-positive cells          Guided by these findings, we selected ultrasound condi-
relative to sham-exposed cells incubated in the same concen-        tions that applied 10 J/cm2 using a pressure of 2 MPa and
tration of DNA. Finally, total GFP expression is reported as        an exposure time of 10 ms for subsequent experiments to
the product of the relative increase in the number of GFP-
positive cells (i.e., transfection efficiency) and the relative            A                        1.E+07

increase in intracellular GFP among GFP-positive cells.
                                                                      Calcein Molecules/Cell

Experiments at all conditions were performed at least three
times (n ≥ 3), each on a different day. Student’s t-test was
used to identify differences between pairs of data points,                                          1.E+05
                                                                                                                                                1.2 MPa

whereas Analysis of Variance (ANOVA) was employed to                                                                                            1.7 MPa

compare three or more data points using StatView statistical
                                                                                                                                                2 MPa

software (SAS Institutes, Cary, NC). A value of p < 0.05 was
                                                                                                                                                2.5 MPa

interpreted as significant. Data on graphs are reported as the
                                                                                                                                                3 MPa
                                                                                                                                                3.5 MPa
mean ± SEM (standard error of the mean).
                                                                                                          0.1       1         10        100    1000
Results                                                                                                             Energy Exposure (J/cm2)

Intracellular Uptake and Cell Viability                                                                                                          1.2 MPa
After Ultrasound Exposure                                                                                                                        1.7 MPa
                                                                                                           100                                   2 MPa

We hypothesize that ultrasound may be used to enhance intra-                                                                                     2.5 MPa
                                                                                      Cell viability (%)

cellular delivery of chemotherapeutics and genes into glioma                                                75
                                                                                                                                                 3 MPa

cells. As a first test, we examined the effect of ultrasound on                                                                                  3.5 MPa

intracellular uptake of calcein, which is an inert, fluorescent                                             50
marker that is not normally taken up into cells. Gliosarco-
ma cells were exposed to ultrasound in vitro over a range of
acoustic pressures (1.2-3.5 MPa) and exposure times (8-1480                                                 25

ms) that corresponded to acoustic energies between 0.4-600
J/cm2. After exposure to ultrasound, intracellular uptake of                                                 0

calcein and cell viability were quantified.                                                                   0.1   1         10         100   1000

                                                                                                                    Energy Exposure (J/cm2)

As shown in Figure 1A, cells exposed to ultrasound ex-              Figure 1: Effects of ultrasound on (A) intracellular uptake of calcein and
hibited high levels of intracellular uptake. At the weakest         (B) viability of 9L gliosarcoma cells in vitro. Data shown in each graph are
                                                                    for the same cell populations analyzed simultaneously by flow cytometry for
conditions used (i.e., ~1 J/cm2), uptake averaged between           uptake and viability. A range of acoustic energy exposures were achieved by
104 and 105 molecules per cell. At larger energies, average         varying acoustic pressure (1.2-3.5 MPa) and exposure time (8-1480 ms).

                                                        Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
438	                                                                                                                                    Zarnitsyn	et	al.

deliver chemotherapeutic agents. Under these conditions,
there was extensive intracellular uptake and limited cell vi-                                    100
ability loss due to the ultrasound.

                                                                    Cell viability (%)
Synergy of Ultrasound and BCNU Cytotoxicity

To identify possible synergy between cell death caused by

ultrasound and BCNU, we first measured the cytotoxicity of
BCNU alone as a function of BCNU concentration. BCNU                                              25

was selected as a model chemotherapeutic because of its es-
tablished use in brain cancer therapy (40). As shown in Fig-                                       0

ure 2 (white data points), BCNU at concentrations ≤ 30 μM











had no significant effect on cell viability (ANOVA, p = 0.32).

                                                                                                                       BCNU concentration (µM)
At 100 μM BCNU, cell viability dropped to 50%. At larger
                                                                   Figure 2: Synergistic effect of ultrasound and BCNU cytotoxicity. Cell
BCNU concentrations, cell viability dropped still further.         viability is shown as a function of BCNU concentration for BCNU alone
                                                                   (white data points) and BCNU with ultrasound, after a correction to the
Cytotoxicity was significantly greater when cells were addi-       data (black data points). For the corrected data, viability loss due to ul-
tionally exposed to ultrasound. As shown in Figure 2 (black        trasound alone has been removed from the overall viability (see Methods
                                                                   and Materials section).
data points), the synergistic effects of ultrasound to increase
BCNU cytotoxicity were evident at all BCNU concentrations
                                                                   Effect of Ultrasound Conditions on DNA Transfection
studied (two-way ANOVA, p < 0.0001). At small concentra-
tions of BCNU (i.e., 0.1-10 μM), BCNU alone did not kill
                                                                   Gene-based therapies offer exciting prospects to treat brain
significant numbers of cells, but the addition of ultrasound
                                                                   cancer, but have been limited by inefficient gene delivery
increased BCNU cytotoxicity to reduce cell viability to ap-
                                                                   and expression. To test the hypothesis that ultrasound can be
proximately 75%. Larger BCNU concentrations increased
                                                                   used to increase gene transfection of glioma cells, we mea-
cytotoxicity still more (ANOVA, p = 0.003). In this way,
                                                                   sured the effects of ultrasound parameters on the transfection
the initial increase of BCNU cytotoxicity was observed at ≥
                                                                   efficiency of 9L gliosarcoma cells by plasmid DNA encod-
1000-fold lower concentration (i.e., from 100 μM to < 0.1
                                                                   ing green fluorescent protein (GFP). As shown in Figure 4,
μM) and the dose to kill 50% of cells (LD50) was reduced by
                                                                   transfection efficiency increased with ultrasound pressure
more than a factor of 3 (i.e., from 100 μM to 30 μM). We hy-
                                                                   over the range of conditions tested (two-way ANOVA, p = <
pothesize that this increased cytotoxicity is due to increased
                                                                   0.0001). Transfection also increased with ultrasound expo-
intracellular uptake of BCNU due to ultrasound.
                                                                   sure time at short exposures (≤ 15 ms; two-way ANOVA, p =
                                                                   0.01), but showed no significant change at longer exposures
Synergy of Ultrasound and Bleomycin Cytotoxicity
                                                                   (≥ 75 ms; two-way ANOVA, p = 0.24). The best transfection
We next looked for possible synergy between cell death
caused by ultrasound and bleomycin using an approach                                             100
similar to that used for BCNU. Although bleomycin is
not commonly used to treat brain cancer, we selected it as
a model chemotherapeutic because it is known to have dif-                                         75
                                                                            Cell viability (%)

ficulty crossing the cell membrane, but is highly cytotoxic
once it reaches its intracellular target (37). For this reason,                                   50
cytotoxicity of bleomycin has been shown to be synergisti-
cally increased when combined with electroporation, which
is known to transiently increase cell membrane permeability

and thereby increase intracellular delivery (42).
As shown in Figure 3 (white data points), bleomycin alone                                                   0           0.1       1         10         100
was increasingly cytotoxic to gliosarcoma cells at increasing                                                      Bleomycin concentration (µM)
drug concentrations. When combined with ultrasound, cy-            Figure 3: Synergistic effect of ultrasound and bleomycin cytotoxicity. Cell
totoxicity was significantly increased (Figure 3, black data       viability is shown as a function of bleomycin concentration for bleomycin
points, two-way ANOVA, p = 0.015). The difference was              alone (white data points) and bleomycin with ultrasound, after a correction
                                                                   to the data (black data points). For the corrected data, viability loss due to
statistically significant, even when the outlier point at 1 μM     ultrasound alone has been removed from the overall viability (see Methods
concentration is included.                                         and Materials section).

Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
Ultrasound-enhanced	Chemotherapy	                                                                                                                                                                                        439

                                                                                                                       0.0001) but increase in energy exposure above 1 J/cm2 did
                                   P=0.5 MPa
                                                                                                                       not change it significantly (p = 0.52, respectively). At the op-
Transfection efficiency

                                                                                             Transfection efficiency
                                                                                               (% of treated cells)
                                                                                                                       timal conditions, GFP expression was increased by 30-fold
                                   P=1 MPa
   (relative to sham)

                          40                                                             3
                                   P=2 MPa

                                                                                                                       relative to sham-exposed controls.

                          20                                                                                           Discussion
                                                                                                                       Prospects of Ultrasound-enhanced Chemotherapy
                           0                                                             0
                               0       1       4   8       15   75   25    10     30
                                                                                                                       To study the ability of ultrasound to increase chemotherapeu-
                                                                       0     00     00

                                                       On time, ms                                                     tic cytotoxicity, we first needed to identify suitable ultrasound
 Figure 4: Effect of ultrasound pressure and exposure time on transfec-                                                parameters. Initial studies identified a trade off between high
 tion efficiency. Ultrasound was applied at 0.5 MPa (white bars), 1.0 MPa
 (gray bars), or 2.0 MPa (black bars) peak negative pressure for total ex-
 posure times of 0-3000 ms in the presence of 30 μg/ml DNA encoding                                                            A                                                                              4%
 GFP. Transfection efficiency generally increased with increasing strength                                                                           40
                                                                                                                                                              3 ug/ml
 of ultrasound conditions.                                                                                                                                    10 ug/ml

                                                                                                                                                                                                                   Transfection efficiency
                                                                                                                           Transfection efficiency
                                                                                                                                                              30 ug/ml

                                                                                                                                                                                                                     (% of treated cells)

                                                                                                                              (relative to sham)
 efficiency obtained in this set of experiments increased the
 number of cells expressing detectable amounts of GFP by                                                                                             20

 30-fold, which corresponded to 2.5% of treated cells.
                                                                                                                                                     10                                                       1%

 The maximum transfection efficiency is often limited by cell
 death at stronger conditions. In these experiments, cell vi-
                                                                                                                                                      0                                                       0%
                                                                                                                                                              0             1      3      10     33     100
 ability was found to decrease with increasing pressure and                                                                                                                  Energy exposure (J/cm2)
 exposure time (two-way ANOVA, p = 0.008 and p < 0.0001,
 respectively) to values as low as 60% at the strongest condi-                                                                                         3
 tions used (data not shown). Thus, the plateau and possible                                                                   B                                  3 ug/ml

 decrease in transfection efficiency at the strongest ultrasound                                                                                                  10 ug/ml

 conditions used can be explained by the competing effects of
                                                                                                                                                                  30 ug/ml
                                                                                                                         (relative to sham)
                                                                                                                         Intracellular GFP


 ultrasound more effectively driving DNA into cells, but also                                                                                        1.5
 more extensively killing cells, such that the net effect is a
 leveling off or reduction in overall transfection efficiency.                                                                                         1

 Effect of DNA Concentration on DNA Transfection
                                                                                                                                                               0            1      3     10     33     100
 The effect of DNA concentration on transfection efficiency is                                                                                                              Energy exposure (J/cm2)
 shown in Figure 5A. Although the effect was small, increas-
 ing DNA concentration increased the number of transfected                                                                                            40
 cells (two-way ANOVA, p = 0.0005). One might also expect,                                                                          C                         3 ug/ml

 however, that DNA concentration could additionally affect
                                                                                                                                                              10 ug/ml
                                                                                                                              Total GFP expression

 the amount of GFP expressed by each transfected cell. Con-
                                                                                                                                                              30 ug/ml
                                                                                                                                (relative to sham)

 sistent with this expectation, Figure 5B shows that increasing                                                                                       20
 DNA concentration increased the intensity of GFP expression
 by each cell (two-way ANOVA, p < 0.0001). In contrast, in-                                                                                           10

 creasing ultrasound energy exposure did not affect the average
 intensity of GFP expression (two-way ANOVA, p = 0.75).                                                                                                   0
                                                                                                                                                              0             1      3     10     33     100
                                                                                                                                                                            Energy exposure (J/cm2)
 Often, the total amount of protein expression is more im-                                                             Figure 5: Effect of DNA concentration and ultrasound energy on transfec-
 portant that the number of cells expressing the protein. To                                                           tion efficiency. In (A), transfection efficiency is determined on the basis of
 assess this metric, the total amount of GFP expressed was                                                             the number of cells with detectable levels of GFP expression. In (B), the
 determined by multiplying the number of GFP-positive cells                                                            average cellular intensity of GFP expression is reported for each transfected
                                                                                                                       cell. In (C), the values in (A) and (B) are multiplied to determine the total
 (Figure 5A) by the average cellular intensity of GFP expres-                                                          amount of GFP expression among transfected cells. To achieve the range of
 sion (Figure 5B). As shown in Figure 5C, DNA concentra-                                                               ultrasound energies, ultrasound was applied at 2 MPa peak negative pres-
 tion increased total GFP expression (two-way ANOVA, p <                                                               sure for total exposure times of 0-750 ms.

                                                                                               Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
440	                                                                                                            Zarnitsyn	et	al.

levels of intracellular uptake and high levels of cell viability.      cells and increasing the DNA concentration increased the ex-
Because ultrasound is intended to facilitate delivery of chemo-        pression level in each transfected cell. The best conditions
therapeutics that specifically target cancer cells, it is not desir-   tested in this study found a 30-fold increase in total expres-
able for ultrasound to indiscriminately kill cells by itself. As       sion levels over non-sonicated controls.
a compromise, we selected ultrasound conditions that loaded
cells with millions of molecules per cell and maintained 70%           Despite large increases in transfection efficiency, the ob-
viability. If higher viability were important, then weaker ultra-      served increases may not be enough for some applications.
sound conditions could have been selected to deliver on the or-        At partially optimized conditions, just 2.5% of cells showed
der of 105 molecules per cell with minimal cell loss. Moreover,        increased expression above background. For applications
it is likely that cell viability would be higher in vivo, where        in which, for example, transfection needs to be achieved in
cells are supported by their natural extracellular environment,        most, if not all, cells for efficient tumor removal, further en-
rather than suspension in media (43). For this proof-of-prin-          hancement of expression is needed. In contrast, other cancer
ciple study, ultrasound conditions were not fully optimized.           gene therapy approaches require only a subset of cells to be
                                                                       transfected so that the expression products can be distributed
Chemotherapeutic cytotoxicity was first measured using                 among neighboring cells via extracellular or intracellular
BCNU either with or without exposure to ultrasound. As                 pathways. For example, in a previous study, suicide gene
expected, BCNU alone was effective at killing gliosarcoma              plasmids were shown to be effective when transfected into
cells. However, the addition of ultrasound increased cyto-             just 2% of the glioma cell population (46).
toxicity in a synergistic way. Our proposed hypothesis is that
ultrasound facilitated greater intracellular delivery of BCNU          Conclusion
into cells and thereby increased BCNU efficacy. However,
BCNU is a small (214 Da), lipophilic (XlogP = 1.26) mol-               This study assessed the prospects of using ultrasound to en-
ecule that should cross cell membranes fairly well. Thus, an           hance treatments of brain cancer. Guided by initial experi-
alternative hypothesis is that ultrasound otherwise sensitized         ments to identify suitable ultrasound conditions for intra-
cells to BCNU. For example, a previous study showed that               cellular delivery to glioma cells, we found that ultrasound
exposure of cells to ultrasound in combination with another            synergistically increased in vitro cytotoxicity of two chemo-
chemotherapeutic, quercetin, synergistically increased cyto-           therapeutics with very different physicochemical properties:
toxicity by a mechanism believed to involve depletion of a             BCNU and bleomycin. This suggests the possibility that che-
heat shock protein by ultrasound (44). In a separate study             motherapeutics could be administered to the brain in combina-
not involving ultrasound, topoisomerase I inhibitors were              tion with ultrasound focused on the tumor to increase efficacy
shown to sensitize cells to BCNU (45).                                 by a mechanism believed to involve increased intracellular
                                                                       uptake of drug. We also found that ultrasound increased in
We also examined the effects of ultrasound on cytotoxicity of          vitro glioma cell transfection with a GFP reporter DNA plas-
bleomycin. This drug provides a useful contrast with BCNU,             mid by 30-fold. This similarly suggests that ultrasound could
because bleomycin is large (1.4 kDa) and hydrophilic and,              facilitate gene therapy in the brain, especially for indications
therefore, cannot easily cross cell membranes (41). None-              where the dangers of virus-based delivery are undesirable
theless, ultrasound also increased the cytotoxicity of bleomy-         and moderate transfection efficiency is sufficient.
cin in a synergistic way. In this case, the proposed mecha-
nism involving increased intracellular delivery of bleomycin           Acknowledgements
caused by ultrasound is more plausible.
                                                                       We would like to thank Johnafel Crow, Daniel Hallow, and
When considered all together, this study suggests that ul-             Robyn Schlicher for helpful discussions and Henry Brem and
trasound may be a useful adjunct to chemotherapy of brain              Betty Tyler for providing the GS 9L cell line. This work was
cancer. Ultrasound was shown to synergistically increase               supported in part by the National Institutes of Health, Cyt-
cytotoxicity of two chemotherapeutics with very different              oDome, Inc., and the Georgia Research Alliance. M. R. P. is
physicochemical properties. This observation should moti-              a member of the Center for Drug Design, Development and
vate further studies of protocol optimization, in vivo efficacy        Delivery and the Institute for Bioengineering and Bioscience
and safety, and clinical considerations.                               at Georgia Tech. V. G. Z. carried out, designed, and analyzed
                                                                       all transfection experiments, most chemotherapeutic cytotox-
Prospects of Ultrasound-enhanced Gene Therapy                          icity experiments, and some calcein uptake experiments. P.
                                                                       P. K. carried out, designed, and analyzed most calcein uptake
Ultrasound may also be useful to facilitate gene therapy as            experiments and some chemotherapeutic cytotoxicity experi-
a nonviral approach. This study showed that increasing the             ments. M. R. P. helped design and interpret all aspects of the
strength of ultrasound increased the number of transfected             study, and served as the principal investigator.

Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
Ultrasound-enhanced	Chemotherapy	                                                                                                                     441

References                                                                        22. Stewart, H. F., Stratmeyer, M. E. An Overview of Ultrasound: Theo-
                                                                                      ry, Measurement, Medical Applications and Biological Effects. HHS
   1. Ries, L. A. G., Smith, M. A., Burney, J. G., Linet, M., Tamra, T.,              Publication FDA 82-8290, U.S. Government Printing Office, Wash-
      Young, J. L., Bunin, G. R., Editors. Cancer Incidence and Surviv-               ington, DC (1982).
      al among Children and Adolescents: United States SEER Program               23. Prentice, P., Cuschierp, A., Dholakia, K., Prausnitz, M. R., Campbell,
      1975-1995, National Cancer Institute, SEER Program, NIH Publ. No.               P. Membrane disruption by optically controlled microbubble cavita-
      99-4649, Bethesda, MD (1999).                                                   tion. Nature Physics 1, 107-110 (2005).
   2. Jemal, A., Tiwari, R. C., Murray, T., Ghafoor, A., Samuels, A., Ward,       24. McNeil, P. L, Steinhardt, R. A. Plasma membrane disruption: repair,
      E., Feuer, E. J., Thun, M. J. Cancer statistics, 2004. CA Cancer J Clin         prevention, adaptation. Annu Rev Cell Dev Biol 19, 697-731 (2003).
      54, 8-29 (2004).                                                            25. Schlicher, R. K., Radkhakrishna, H., Tolentino, T. P., Apkarian, R. P.,
   3. Davis, F. G., Kupelian, V., Freels, S., McCarthy, B., Surawicz, T. Prev-        Zarnitsyn, V., Prausnitz, M. R. Mechanism of intracellular deivery by
      alence estimates for primary brain tumors in the United States by be-           acoustic cavitation. Ultrasound Med Biol 32, 915-924 (2006).
      havior and major histology groups. Neuro-oncol 3, 152-158 (2001).           26. Guzman, H. R., Nguyen, D. X., Khan, S., Prausnitz, M. R. Ultra-
   4. Prasad, G., Wang, H., Hill, D. L., Zhang, R. Recent advances in ex-             sound-mediated disruption of cell membranes. II. Heterogeneous ef-
      perimental molecular therapeutics for malignant gliomas. Curr Med               fects on cells. J Acoust Soc Am 110, 597-606 (2001).
      Chem Anti-Canc Agents 4, 347-361 (2004).                                    27. Guzman, H. R., Prausnitz, M. R. Bioeffects caused by changes in
   5. Surmacz, E. Growth factor receptors as therapeutic targets: strate-             acoustic cavitation bubble density and cell concentration: a unified
      gies to inhibit the insulin-like growth factor I receptor. Oncogene 22,         explanation based on cell-to-bubble ratio and blast radius. Ultrasound
      6589-6597 (2003).                                                               Med Biol 29, 1211-1222 (2003).
   6. Chen, Q. R., Zhang, L., Gasper, W., Mixson, A. J. Targeting tumor an-       28. Sundaram, J., Mellein, B. R., Mitragotri S. An experimental and theo-
      giogenesis with gene therapy. Mol Genet Metab 74, 120-127 (2001).               retical analysis of ultrasound-induced permeabilization of cell mem-
   7. Rautioa, J., Chikhale, P. J. Drug delivery systems for brain tumor              branes. Biophys J 84, 3087-3101 (2003).
      therapy. Curr Pharm Des 10, 1341-1353 (2004).                               29. Mir, L. M., Orlowski, S., Belehradek, J., Paoletti, C. Electrochemo-
   8. Lawson, H. C., Sampath, P., Bohan, E., Park, M. C., Hussain, N.,                therapy potentiation of antitumor effect of bleomycin by local electric
      Olivi, A., Weingart, J., Kleinberg, L., and Brem, H. Interstitial che-          pulses. Eur J Cancer 27, 68-72 (1991).
      motherapy for malignant gliomas: the Johns Hopkins experience. J            30. Kinoshita, M., McDannold, N., Jolesz, F. A., et al. Noninvasive local-
      Neurooncol 83, 61-70 (2007).                                                    ized delivery of Herceptin to the mouse brain by MRI-guided focused
   9. Giese, A., Kucinski, T., Knopp, U., Goldbrunner, R., Hamel, W., Me-             ultrasound-induced blood-brain barrier disruption. Proc Natl Acad
      hdorn, H. M., Tonn, J. C., Hilt, D., Westphal, M. Pattern of recur-             Sci USA 103, 11719-11723 (2006).
      rence following local chemotherapy with biodegradable carmustine            31. Zarnitsyn, V. G., Prausnitz, M. R. Physical parameters influencing
      (BCNU) implants in patients with glioblastoma. J Neurooncol 66,                 optimization of ultrasound-mediated DNA transfection. Ultrasound
      351-360 (2004).                                                                 Med Biol 30, 527-538 (2004).
  10. Grossman, S. A. Arguments against the routine use of currently avail-       32. Larina, I. V., Evers, B. M., Esenaliev, R. O. Optimal drug and gene
      able adjuvant chemotherapy in high-grade gliomas. Semin Oncol 30,               delivery in cancer cells by ultrasound-induced cavitation. Anticancer
      19-22 (2003).                                                                   Res 25, 149-156 (2005)
  11. Guzman, H. R., Nguyen, D. X., McNamara, A. J., Prausnitz, M. R.             33. Barth, R. F. Rat brain tumor models in experimental neuro-oncology:
      Equilibrium loading of cells with macromolecules by ultrasound: Ef-             the 9L, C6, T9, F98, RG2 (D74), RT-2 and CNS-1 gliomas. J Neu-
      fects of molecular size and acoustic energy. J Pharm Sci 91, 1693-              rooncol 36, 91-102 (1998).
      1701 (2002).                                                                34. Ueda-Kawamitsu, H., Lawson, T. A., Gwilt, P. R. In vitro pharma-
  12. Wyber, J. A., Andrews, J., D’Emanuele, A. The use of sonication for             cokinetics and pharmacodynamics of 1,3-bis(2-chloroethyl)-1-nitro-
      the efficient delivery of plasmid DNA into cells. Pharm Res 14, 750-            sourea (BCNU). Biochem Pharmacol 63, 1209-1218 (2002).
      756 (1997).                                                                 35. Guzman, H. R., Nguyen, D. X., Khan, S., Prausnitz, M. R. Ultrasound-
  13. Han, Y. W., Ikegami, A., Rajanna, C., Kawsar, H. I., Zhou, Y., Li, M.,          mediated disruption of cell membranes. I. Quantification of molecular
      Sojar, H. T., Genco, R. J., Kuramitsu, H. K., Deng, C. X. Identifica-           uptake and cell viability. J Acoust Soc Am 110, 588-596 (2001).
      tion and characterization of a novel adhesin unique to oral fusobacte-      36. Hallow, D. M., Mahajan, A. D., McCutchen, T. E., Prausnitz, M. R.
      ria. J Bacteriol 187, 5330-5340 (2005).                                         Measurement and correlation of acoustic cavitation with cellular
  14. Mitragotri, S. Healing sound: the use of ultrasound in drug delivery            bioeffects. Ultrasound Med Biol 32, 1111-1122 (2006).
      and other therapeutic applications. Nat Rev Drug Discov 4, 255-             37. Prausnitz, M. R., Lau, B. S., Milano, C. D., Conner, S., Langer, R.,
      260 (2005).                                                                     Weaver, J. C. A quantitative study of electroporation showing a pla-
  15. Koch, S., Pohl, P., Cobet, U., Rainoy, N. G. Ultrasound enhancement             teau in net molecular-transport. Biophys J 65, 414-422 (1993).
      of liposome-mediated cell transfection is caused by cavitation effects.     38. Tata, D. B., Dunn, F., Tindall, D. J. Selective clinical ultrasound sig-
      Ultrasound Med Biol 26, 897-903 (2000).                                         nals mediate differential gene transfer and expression in two human
  16. Pitt, W. G., Husseini, G. A., Staples, B. J. Ultrasonic drug delivery--a        prostate cancer cell lines: LnCap and PC-3. Biochem Biophys Res
      general review. Expert Opin Drug Deliv 1, 37-56 (2004).                         Commun 234, 64-67 (1997).
  17. Tachibana, K. Emerging technologies in therapeutic ultrasound: ther-        39. Greenleaf, W. J., Bolander, M. E., Sarkar, G., et al. Artificial cavita-
      mal ablation to gene delivery. Hum Cell 17, 7-15 (2004).                        tion nuclei significantly enhance acoustically induced cell transfec-
  18. Tsivgoulis, G., Alexandrov, A. V. Ultrasound-enhanced thrombolysis              tion. Ultrasound Med Biol 24, 587-595 (1998)
      in acute ischemic stroke: potential, failures, and safety. Neurothera-      40. Guerin, C., Olivi, A., Weingart, J. D., Lawson, H. C., Brem, H. Recent
      peutics 4, 420-427 (2007).                                                      advances in brain tumor therapy: local intracerebral drug delivery by
  19. Saad, A. H., Hahn, G. M. Ultrasound-enhanced effects of adriamycin              polymers. Invest New Drugs 22, 27-37 (2004).
      against murine tumors. Ultrasound Med Biol 18, 715-723 (1992).              41. Kotnik, T., Macek-Lebar, A., Miklavcic, D., Mir, L. M. Evaluation of
  20. Saad, A. H., Hahn, G. M. Ultrasound enhanced drug toxicity on Chi-              cell membrane electropermeabilization by means of a nonpermeant
      nese hamster ovary cells in vitro. Cancer Res 49, 5931-5934 (1989).             cytotoxic agent. Biotechniques 28, 921-926 (2000).
  21. Rosenthal, I., Sostaric, J. Z., Riesz, P. Sonodynamic therapy--a review     42. Orlowski, S., Belehradek, J., Paoletti, C., Mir, L. M. Transient elec-
      of the synergistic effects of drugs and ultrasound. Ultrason Sonochem           tropermeabilization of cells in culture. Increase of the cytotoxicity of
      11, 349-363 (2004).                                                             anticancer drugs. Biochem Pharmacol. 37, 4727-4733 (1988).

                                                                    Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007
442	                                                                                                                           Zarnitsyn	et	al.

  43. Lodish, H., Berk, A., Zipursky, S. L., Matsudaira, P., Baltimore, D.,    45. Storm, P. B., Renard, V. M., Moriarity, J. L., Tyler, B., Wilentz, R. E.,
      Darnell, J. Molecular Cell Biology. 4th Edition. W.H. Freeman and            Brem, H., Weingart, J. D. Systemic BCNU enhances the efficacy of
      Co, New York (2000).                                                         local delivery of a topoisomerase I inhibitor against malignant glio-
  44. Paliwal, S., Sundaram, J., Mitragotri, S. Induction of cancer-specific       ma. Cancer Chemother Pharmacol 54, 361-367 (2004).
      cytotoxicity towards human prostate and skin cells using quercetin       46. Bourbeau, D., Lavoie, G., Nalbantoglu, J., Massie, B. Suicide gene
      and ultrasound. Br J Cancer 92, 499-502 (2005).                              therapy with an adenovirus expressing the fusion gene CD:UPRT in
                                                                                   human glioblastomas: different sensitivities correlate with p53 status.
                                                                                   J Gene Med 6, 1320-1332 (2004).

                                                                                       Received: December 5, 2006; Revised: July 20, 2007;
                                                                                                   Accepted: August 14, 2007

Technology in Cancer Research & Treatment, Volume 6, Number 5, October 2007

Shared By:
yaohongm yaohongm http://