Gene Therapy (2006) 13, 163–172
& 2006 Nature Publishing Group All rights reserved 0969-7128/06 $30.00
www.nature.com/gt
ORIGINAL ARTICLE
Therapeutic ultrasound-mediated DNA to cell and
nucleus: bioeffects revealed by confocal and atomic
force microscopy
M Duvshani-Eshet, L Baruch, E Kesselman, E Shimoni and M Machluf
The Faculty of Biotechnology and Food Engineering, The Technion – Israel institute of Technology, Haifa, Israel
Therapeutic ultrasound (TUS) has the potential of becoming a
powerful nonviral method for the delivery of genes into cells
and tissues. Understanding the mechanism by which TUS
delivers genes, its bioeffects on cells and the kinetic of gene
entrances to the nucleus can improve transfection efficiency
and allow better control of this modality when bringing it to
clinical settings. In the present study, direct evidence for the
role and possible mechanism of TUS (with or without Optison)
in the in vitro gene-delivery process are presented. Appling
a 1 MHz TUS, at 2 W/cm2, 30%DC for 30 min was found to
achieve the highest transfection level and efficiency while
maintaining high cell viability (480%). Adding Optison further
increase transfection level and efficiency by 1.5 to three-fold.
Confocal microscopy studies indicate that long-term TUS
application localizes the DNA in cell and nucleus regardless
of Optison addition. Thus, TUS significantly affects transfection efficiency and protein kinetic expression. Using innovative direct microscopy approaches: atomic force microscopy,
we demonstrate that TUS exerts bioeffects, which differ from
the ones obtained when Optison is used together with TUS.
Our data suggest that TUS alone affect the cell membrane in
a different mechanism than when Optison is used.
Gene Therapy (2006) 13, 163–172. doi:10.1038/sj.gt.3302642;
published online 22 September 2005
Keywords: therapeutic ultrasound; gene delivery; contrast agents; atomic force microscopy; bioeffects
Introduction
Ultrasound (US) has emerged as a unique method for the
delivery of genes into cells and tissues,1–5 thus placing
it as a promising nonviral approach for gene delivery
in the clinical settings. Most of the studies using US
for gene delivery have applied low-frequency levels
(o1 MHz). These frequencies are known to create
cavitation (creation and oscillation of gas bubbles in
aqueous media), which is suggested to play a major role
in cell membrane permeabilization.6 However, these
frequencies are also known to induce tissue damage
if not properly controlled.7 As a result, studies using
therapeutic US (TUS), which utilizes higher frequencies
(1–3 MHz), intensity levels of 0.5–2 W/cm2, pulse mode,
have emerged.2,3,5 Using TUS rather than low-frequency
US has several advantages: It is approved for clinical
application, has good penetration through soft tissues,
does not damage cell and tissue and does not affect DNA
integrity.5,8
The mechanisms by which TUS mediates gene
delivery into cells remains poorly understood. Cavitation
is believed to play a role in almost all US modalities used
for gene delivery.6,9 However, TUS is known to be
Correspondence: Dr M Machluf, The Laboratory of Cancer Drug
Delivery and Mammalian Cell Technology, The Faculty of Biotechnology and Food Engineering, Technion – Israel Institute of
Technology, Haifa 32000, Israel.
E-mail: [email protected]
Received 3 March 2005; revised 20 July 2005; accepted 1 August
2005; published online 22 September 2005
subcavitational due to the high frequencies and low
intensities used,4,10 thus making it hard to give an
explanation for its bioeffects. Other studies also suggested that TUS might enhance gene transfection by
accelerating DNA escape from endosomes, alteration of
intracellular trafficking, upregulation of genes, protein
translation,5,8 or altering cell membrane permeability,
thus creating pores.11,12 Understanding the mechanism
by which TUS delivers genes to cells, its bioeffects on cell
membrane, the kinetics of gene entrance to the nucleus
and its time of expression are crucial in order to improve
transfection efficiency and allow better control of this
modality.13,14
Recently, Optison, an US contrast agents (USCA), has
been used to enhance gene delivery when using TUS
in vitro and in vivo.11,15 The use of USCA is suggested to
lower the energy threshold for cavitation,10,16 making it
feasible in the therapeutic frequencies. However, the
bioeffects of USCA in combination with TUS need to be
further studied.
This study presents direct evidence for the ability
of TUS to deliver the DNA to the nucleus and its
bioeffects on cells, with or without Optison. Using
direct innovative scanning probe microscopy approach;
atomic force microscopy (AFM) we demonstrate that
TUS exerts bioeffects which may be different from
the ones obtained when Optison is added. AFM is a
unique technology which offers cell topography analyses
in the nano-scale.17 Using confocal laser scanning
microscopy (CLSM), DNA trafficking in the different
cell compartments, the kinetic of DNA entrance to the
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
cells and nucleus, and its expression to protein are
elucidated.
endothelia (BCE) which are primary cells (Supplementary, Figure 1). Therefore, these parameters were used
throughout the rest of the study.
Results
Effect of Optison on TUS transfection
To evaluate the effect of Optison during TUS application,
cells were exposed to TUS with the addition of Optison.
As seen in Figure 2, when cells were exposed to TUSOptison at 20% DC (Figure 2a), 30% (Figure 2b) or
50%DC (Figure 2c) for 10, 20, 30, or 40 min, luciferase
activity was increases about three-, 2.5-, 1.5- and 1.1-fold,
respectively, compared to transfection with TUS alone
(Figure 1). The use of Optison did not have a significant
effect on cell viability when compared to TUS alone.
When using pGFP and TUS at 30% DC, 2 W/cm2 for
30 min, addition of Optison led to almost two-fold
of increase in transfection efficiency (Figure 2d). No
significant expression of GFP was detected in the control
experiments.
a
Luciferase Activity
(RLU/mg protein)
80000
60000
40000
20000
120
b 360000
120
100
300000
100
80
240000
80
60
180000
60
40
120000
40
20
% Transfection
30
20
10
0
0
10
20
30
40
Exposure Time (min)
100
80
60
40000
40
20000
20
0
0
0 10 20 30 40
Exposure Time (min)
e
40
d
120
60000
0
0
0 10 20 30 40
Exposure Time (min)
c 80000
20
60000
0
0
2 W/cm2
% Viability
Relative to Control
1 W/cm2
Effect of TUS on the kinetic and localization of DNA
As seen in Figures 3(a and b), 52%78 of cells fixed
immediately post-TUS (without Optison), contained
pGeneGrip plasmid (pGG) in cell cytoplasm. Moreover,
18%76 of the cells contained pGG in the nucleus (Figure
3c). At 5 h post TUS, a slight increase in the number of
cells containing pGG in the cytoplasm and nucleus was
observed (Figures 3a–c) and cells were already expressing GFP (Green cells, Figure 3a). At 1-day post TUS, the
number of cells containing pGG in cytoplasm or nucleus
increased by 14 and 7% respectively, compared to cells
fixed immediately post TUS (Figures 3b and c). At all
time points, the percentage of cells containing pGG in the
different cell compartments was significantly higher than
the control (Po0.001).
0 10 20 30 40
Exposure Time (min)
f
200000
40
150000
30
100000
20
50000
10
% Transfection
Optimization of TUS transfection
Cells were successfully transfected with pGL3-Luc, using
TUS at 1 W/cm2 (0.144 MI) and 2 W/cm2 (0.159 MI)
for 20–40 min. Applying TUS at 20% duty cycle (DC),
2 W/cm2 for 30 min resulted with luciferase activity of
40 30074600 RLU/mg protein, and viability of 90%710
(Figure 1a). Increasing DC to 30% resulted with the
highest luciferase activity of 190 000721 000 RLU/mg
protein and viability of 82%78 (Figure 1b). However,
increase of the DC to 50% for 30 min did not increase cell
transfection and decreased cell viability to 62%78
(Figure 1c). Increasing exposure time to 40 min, using
20 or 30% DC and 2 W/cm2, did not increase transfection, and also reduced cell viability to 75%710 and
62%79, respectively (Figures 1a and b). Similar results
were obtained when using pIRES2-EGFP (pGFP), with
the highest transfection achieved when TUS was applied
at 2 W/cm2, 30% DC for 30 min (Figure 1d). These
parameters also led to the highest transfection efficiency
(28%77) with even distribution of cells expressing GFP
(Figure 1e). In all experiments and parameters used for
transfection, temperature was measured using a thermocouple. The temperature increase was found to be
insignificant (o1.51C) in all in vitro studies. Using
Lipofectaminet, a transfection of 120 000730 000 RLU/
mg protein, and transfection efficiency of 30%78 (Figure
1f) was achieved. Using TUS at 30% DC, 2 W/cm2 for
30 min, also led to the highest transfection in three other
cell type: PC3 – human prostate cancer cell line, PC2 –
murine prostate cancer cell line, and bovine capillary
Luciferase Activity
(RLU/mg protein)
164
0
0
Luciferase GFP
Figure 1 Effects of TUS parameters on cell transfection and viability. TUS of 1 MHz with different intensities and exposure times was used
to transfect cells with pGL-Luc at: (a) 20% DC, (b) 30% DC, (c) 50% DC and (d) transfection efficiency of cells treated with pGFP at 30% DC,
2 W/cm2 for 0–40 min. (e) Representative fluorescent micrograph of cells expressing GFP 24 h post-TUS at 30% DC, 2 W/cm2 for 30 min. Bar
– 100 mm. (f) Transfection of cells with pGL-Luc or pGFP using Lipofectaminet. Values are mean7s.e.m. of 18 different repeats.
Gene Therapy
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
165
Luciferase Activity
(RLU/mg protein)
a
120000
100000
120
b 480000
100
420000
120
c 200000
120
100
160000
100
360000
80
80000
60
60000
240000
60
80000
40
120000
20
20000
0
0
0
0 10 20 30 40
Exposure Time (min)
∗∗
d 60
∗
50
% Transfection
20
0
0 10 20 30 40
Exposure Time (min)
0 10 20 30 40
Exposure Time (min)
40
40000
20
60000
0
0
80
120000
60
180000
40
40000
80
300000
% Viability
Relative to Control
2 W/cm2
1 W/cm2
40
∗
30
20
10
0
+DNA +DNA+Opti
No TUS No TUS
+DNA
+TUS
+DNA+Opti
+TUS
Figure 2 Effect of Optison on TUS transfection and viability. Cells transfected with pGL-Luc using TUS of different intensities and exposure
times with the addition of 10% Optison. (a) 20% DC, (b) 30% DC, (c) 50% DC and (d) transfection efficiency of cells treated with pGFP at 30%
DC, 2 W/cm2 for 30 min with Optison. Values are mean7s.e.m. of 18 different repeats. **Po0.01. *Po0.0001 vs no TUS.
Immediately post-TUS–Optison, 68%77 of the cells
contained pGG in the cells and 32%76 in the nucleus,
(Figures 3a–c). At 1-day post-TUS application, 75%77 of
the cells contained pGG in the cytoplasm, and 39%75 in
the nucleus (Figures 3a–c). As with TUS alone, GFP
expression was also observed 5 h post-TUS–Optison–
DNA (Figure 3a). The percentage of cells containing pGG
in cytoplasm and nucleus, at all time-points post-TUS–
Optison, was significantly higher (Po0.05) than cell
treated without Optison.
Effect of TUS exposure time on DNA localization
TUS was operated for 0, 10, 20 or 30 min in the presence
of pGG. Cells were fixed immediately post-TUS and
imaged using CLSM (Figure 3d) or analyzed by FACS
(Figure 3e). As seen, when applying TUS for 10 min,
20%75 of the cells contained pGG in cytoplasm and
o2% in the nucleus. Applying TUS for 20 or 30 min
increased significantly (Po0.05) the percentage of cells
containing pGG to 34%78 and 49%76, respectively
(Figure 3d). The percentage of cells containing pGG in
the nucleus was also increased to 5%74 and 18%77,
respectively. These results were supported by FACS,
demonstrating an increase in the number of cells
containing pGG with the increase of TUS time-application (Figure 3e).
Kinetic of protein expression
Cells were transfected with pGFP using TUS or
Lipofectaminet. As seen in Figure 4(a), cells exposed
to TUS expressed GFP as early as 3 h post-TUS.
Maximum GFP expression was achieved 15–48 h post-
TUS. However, GFP expression in cells transfected with
Lipofectaminet was detected no sooner than 10–20 h
post-transfection. The addition of Optison during TUS
application did not change the kinetic of GFP expression
by cells when compared to TUS alone (Figure 4b).
Localization of DNA when added post-TUS
In order to study whether TUS is the main driving force
delivering the plasmid to the cells, pGG was added to the
cells at different time points post-TUS or TUS-Optison
application (Figure 5). The cells were fixed 24 h later and
imaged by CLSM. When pGG was added immediately
post TUS application, 11%73 of the cells contained pGG
(Figures 5a and b). The number of cells containing pGG
decreased significantly (Po0.05) to 6%73 and 4%72
when pGG was added 2 or 5 h post-TUS, respectively
(Figures 5a and b). When Optison was added during
TUS application, while pGG was added immediately
post-TUS, 15%74 of the cells contained pGG (Figure 5a
and b). Adding pGG 2 or 5 h post-TUS–Optison resulted
with 8%73 and 6%72 of the cells having pGG
respectively (Figures 5a and b). No pGG was detected
in the nucleus and in the control cells, which were not
exposed to TUS or TUS–Optison and GFP expression
was detected in o1% of the cells.
These results were further supported by FACS
analysis (Figures 5c and d). Applying TUS–pGG and
Optison at the same time, led to a significant (Po0.01)
increase in the number of cells containing pGG compared to cells treated with TUS alone (Figure 5c).
Moreover, when Optison was added, the curve (blue)
was shifted to the right indicating that more copies of
Gene Therapy
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
166
a
TUS-Fix 24 hours
TUS-Fix 5 hours
TUS-Fix Immediately
With Optison
No Optison
Control - No TUS
Control - No TUS
TUS-Fix Immediately
TUS-Fix 5 hours
TUS-Fix 10 hours
TUS-Fix 24 hours
% Cells with pGG inside cells
(per field)
100
80
c
60
40
20
40
30
20
10
0
0
No Optison
60
With Optison
e
Cytoplasm
50
No Optison
250
With Optison
No TUS (0.1%)
Nucleus
40
30' TUS (47%)
Events
% Cells with pGG inside
d
Control - No TUS
TUS-Fix Immediately
TUS-Fix 5 hours
TUS-Fix 10 hours
TUS-Fix 24 hours
50
% Cells with pGG in nucleus
(per field)
b
30
20
20' TUS (29%)
10' TUS (17%)
10
0
0
0
10
20
TUS Exposure Time (min)
30
100
101
102
103
Relative Fluorescence
104
Figure 3 Localization of DNA in cells post TUS. (a) Confocal micrographs of cells transfected with pGG encoding for GFP. Cells with only
pGG or pGG-Optison were exposed to 1 MHz TUS, 2 W/cm2, for 30 min and fixed immediately, 5 and 24 h post-TUS. Cells not exposed to
TUS served as control. White arrows indicate plasmids (seen as dots) inside the cell nucleus. Green cells (seen in clear gray) are cells
expressing GFP (Bar, 10 mm). Micrographs are representatives of 12 different fields scanned in three separate experiments. (b, c)
Quantification of the number of cells containing pGG at different times post-TUS inside the cell (b) or nucleus (c) per total number of cells in
field as images by CLSM; *Po0.05. (d) Quantification of the number of cells containing pGG immediately post-TUS application for 10, 20 or
30 min. All analyses were performed on 12 random confocal images. (e) FACS analysis of cells transfected with pGG immediately post-TUS
applied for 10, 20 and 30 min. Control – no ultrasound (see online for color figures).
plasmid were detected per cell. Adding pGG immediately after TUS or TUS-Optison did not increase
significantly the number of cells containing PGG (Figure
Gene Therapy
5d). Similar data were obtained when pGG was added
2 or 5 h post-TUS with or without Optison (data not
shown).
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
% Normalized cells
expressing GFP
a
120
100
80
60
40
Ultrasound
20
Lipofectamine
0
0
% Normalized cells
expressing GFP
b
5
10 15 20 25 30 35 40
Time post transfection (hours)
45
50
120
100
80
60
40
– Optison
20
+ Optison
5
10 15 20 25 30 35 40
Time post transfection (hours)
167
Discussion
0
0
covered with pits in the size of 100–500 nm, which were
much deeper (100–800 nm) than the ones observed
immediately post-TUS alone (10–15 nm). A similar
phenomenon was observed 5 h post-Optison–TUS
(Figure 6p). At 1-day post-Optison–TUS application, a
decrease in height and width of these features was
detected, resembling those detected 24 h post-TUS alone
(Figures 6i and q).
AFM micrographs were also analyzed by Quartz
software for their height differences (pkpk) and
average roughness (Ra) (Table 1). Height differences in
membrane surface decreased immediately post TUS
when compared to cells not exposed to TUS (Po0.05).
The Ra values also decreased from 7677 nm to 4576 nm
(Po0.05), indicating a smoother surface. The membrane
bioeffect observed when Optison was added during TUS
exposure were significantly (Po0.05) increased, in terms
of Ra and pkpk, compared to the ones obtained with
TUS alone. Similar roughness and topography features
were observed when cells were imaged by scanning
electron microscopy (SEM, Supplementary Figure 2).
45
50
Figure 4 Kinetic of GFP expression post-transfection with TUS or
Lipofectaminet. (a) Kinetic of GFP expression in cells transfected
using TUS (1 MHz, 2 W/cm2, 30 min) or Lipofectaminet. (b)
Kinetic of GFP expression in cells transfected using TUS with or
without Optison. Each sample was read four times, and each
experiment was repeated three times. Values are mean7s.e.m. of 12
different samples.
Effect of TUS on cell membrane – AFM analyses
To detect the effects that TUS induces on cell membrane,
AFM microscopy was used. All nontreated and treated
cells possessed normal shape and size (Figure 6a–c). Cell
surface characterization was performed at magnifications of 5 5 mm (Figures 6d–i) and 10 10 mm (Figure
6j–q). As seen, the surface of nontreated cells consisted
of randomly distributed pits of 100–800 nm wide, and
20–140 nm high (Figures 6d and j). Immediately postTUS application, without Optison, membrane surface
was characterized by uniform prolonged flatter structures ordered predominantly in one direction (Figures 6e
and k). A significant change (Po0.05) in cell membrane
topography was observed 5 h post-TUS (Figure 6l). The
pits were larger from the ones observed in nontreated
cells with higher topography differences (200–700 nm
wide, 20–80 nm high). At 1-day post-TUS, pits similar in
size to the ones observed in nontreated cells were seen,
indicating the reversibility of this phenomenon (Figures
6f and m).
When adding Optison with no TUS, structural
changes in pits height and surface roughness were
immediately detected (Figures 6g and n). However,
when Optison was used with TUS, a significant
difference in membrane structural topography was seen
immediately post-TUS, when compared to cells treated
with TUS alone (Figures 6h and o). The membrane was
TUS-assisted gene delivery is a promising emerging
technology, that may safely deliver and target gene
in vitro and in vivo. To bring such technology closer to the
clinical setting major barriers such as low transfection
efficiency delivering DNA to the nucleus need to be
further addressed. Moreover, there is a clear need to
elucidate the bioeffects and mechanism involved in TUSmediated gene delivery.
In the present study, a 1 MHz TUS modality, operated
at 2 W/cm2, 30% DC for 30 min (without Optison) was
found to achieve the highest transfection efficiency while
maintaining high cell viability. The transfection efficiency
achieved using TUS operated at 2 W/cm2, 30% DC for
30 min is similar to the one achieved with Lipofectaminet supporting the efficiency of this technology.
The kinetics of DNA entrance to cells and its
trafficking to the nucleus are major barriers facing
nonviral gene-delivery systems,18 which is particularly
true when using TUS. To date, no direct evidence has
been published demonstrating DNA entrance to cell
nucleus when using TUS. Therefore, in this study
fluorescent plasmid – pGG,19 which is extensively used
for intracellular trafficking of DNA in the cells,13 was
used. Our confocal studies reveal that pGG is present not
only in cell cytoplasm but also in the nucleus. All
plasmids entering the cell cytoplasm and nucleus
occurred during TUS operation, while no significant
increase was observed in the number of cells containing
DNA at different time points post-TUS. Furthermore,
when looking at the kinetics of protein expression, it
is evident that when using TUS protein expression
occurred within short time period post-TUS application.
A possible explanation for these findings is the long-term
application (30 min) of TUS. When applying TUS for
20 min or less, DNA was detected mostly in the
cytoplasm but almost none in the nucleus. In most
reported studies, TUS was applied for only 1–2 min, and
DNA was not detected in the nucleus, which may also
explain the lower transfection achieved.4,13 All together
these data suggest that TUS plays an important role in
Gene Therapy
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
168
Control-No TUS+pGG
TUS-Fix Immediately
TUS-Fix 2 hours
TUS-Fix 5 hours
With Optison
No Optison
a
c
Control
30
pGG added 5 hr post TUS
∗
25
20
With Optison (63%)
pGG added 2 hr post TUS
Events
35
% Cells with pGG inside
(per field)
pGG added Immediately post TUS
b
pGG added before TUS
256
No Optison (47%)
0
100
∗
15
d 256
101
102
103
104
pGG added immediately post TUS
Events
10
5
0
No Optison
With Optison (13%)
No Optison (10%)
With Optison
0
100
101
102
Relative Fluorescence
103
104
Figure 5 Kinetic of DNA entrance to cells post-TUS with or without Optison. (a) Confocal micrographs of cells in which plasmid DNA
(pGG, seen in dots) was added: immediately, 2 and 5 h post-TUS (1 MHz, 2 W/cm2, 30 min) and were fixed 24 h later. Cells not exposed
to TUS served as control. Micrographs are representatives of 12 different fields scanned in three separate experiments (Bar, 20 mm).
(b) Quantification of the percentage of cells with pGG inside the cell per total number of cells in the field. Analyses were performed on 12
random confocal images; *Po0.05. (c, d) FACS analysis of cells exposed to TUS with or without the addition of Optison. (c) Addition of pGG
before TUS application. (d) Addition of pGG immediately post-TUS application. Red – negative control, cell with pGG without TUS (curve
up to 101). Black – negative control, cell with pGG-Optison, without TUS (curve up to 101). Green – cells exposed to TUS alone (no Optison).
Blue – cells exposed to TUS-Optison (with Optison) (see online for color figures).
delivering DNA to cell cytoplasm and nucleus, and the
entrance of DNA into cells occurs mainly during TUS
application. To support this hypothesis, we have performed experiments in which the DNA was added at
different time point post-TUS application and found that
the number of cells containing DNA was significantly
lower than the number of cells containing DNA when it
was added during TUS application. Moreover, DNA was
located inside the cytoplasm but not in the nucleus and
no GFP expression was observed.
Different groups suggested that TUS mediates gene
delivery by altering cell membrane porosity and/or
creating holes.11,12,20 This is believed to be due to the
Gene Therapy
physical or thermal effects induced by US application.8,9
In our studies, no significant increase in temperature was
detected (o1.51C) even when using 30 min of TUS. Thus,
indicating that thermal effects are not the main mechanism affecting membrane permeability and other forces,
such as the mechanical ones, may play a more significant
role in alteration of cell membrane.8,9 However, this
study is performed in vitro and when going to in vivo
studies the thermal effect should be monitored and
considered due to different absorption of the US energy
by the different tissues.14 The other mechanism proposed
to account for TUS bioeffect on cell membrane is based
on the mechanical and physical properties of the US
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
169
a 25
No TUS
No Optison
5 10 15 20 25
869nm
+TUS - Optison
20
15
10
5
0
µm 0
With Optison
+TUS + Optison
c 25
No TUS
No Optison
h
i
713nm
662nm
5 µm 5 µm
5 µm
5 µm 5 µm
5 µm
5 10 15 20 25
TUS-Fix immediately
TUS-Fix 5 hours
k
j
1271nm
10 µm
10 µm
1327nm
10 µm
m
l
863nm
n
With Optison
5 µm 5 µm
689nm
5 µm
688nm
5 µm 5 µm
g
5 10 15 20 25
TUS Fix 24 hours
f
489nm
5 µm
b 25
20
15
10
5
0
µm 0
e
d
No TUS
20
15
10
5
0
µm 0
TUS-Fix immediately
10 µm
1082nm
10 µm
10 µm
1173nm
10 µm
10 µm
o
p
q
1481nm
1236nm
1248nm
10 µm
10 µm
10 µm
10 µm
TUS-Fix 24 hours
10 µm
10 µm
10 µm
10 µm
Figure 6 AFM scans of the outer cell membrane post-TUS with or without the addition of Optison. Cells were exposed to TUS at 30% DC,
2 W/cm2 for 30 min and fixed without the application of TUS, immediately, 5 and 24 h post-TUS. (a–c) AFM-height images of the whole cell
at 25 25 mm: (a) without the application of TUS. (b) Immediately post-TUS alone. (c) Immediately post-TUS–Optison application. (d–q)
AFM three-dimensional images of the surface membrane of cells not exposed to TUS (d, g, j and n), fixed immediately (e, h, k and o), 5 h
(l and p) or 24 h post-TUS (f, i, m and q), without (top row) or with (bottom row) the addition of Optison. AFM scans were performed at
5 5 mm (d–i) and 10 10 mm (j–q). All images are representative of six different cells fixed at the same time and exposed to the same
parameters, scanned at three different occasions.
waves. We have used a new approach, AFM, to try and
understand the possible mechanism responsible for such
bioeffects. AFM is a powerful tool in biology, which
provides three-dimensional imaging of cell surface
topography and detects changes in surface stiffness,
elasticity and roughness, with a resolution of 10 nm.17
Some of the major advantages of AFM over other
scanning microscopy such as SEM are the prevention
of dehydration process, gold sputtering and high
vacuum needed for electron microscopy samples preparations. In addition, AFM scanning enables to measure
size of features in the x-, y-, z-axes, thus determine the
depth of pits seen in the membrane. The AFM studies
demonstrated that immediately post-TUS, the cell
membrane becomes smoother and flatter as indicated
by the significant decrease in cell membrane height and
roughness. These changes in cell membrane structure
may be a result of the mechanical pressure applied by the
TUS. This mechanical pressure may in turn reduce the
unstirred layer in the local environment of the cell, thus
leading to increase in DNA concentration near the cell
membrane.6,8,14 Tachibana et al.21 observed, using SEM,
minor disruption of membrane cell surface due to the
application of low frequency US (255 kHz). While
Taniyama et al.,11 which also used SEM to study the
surface of cells post-TUS application, observed almost no
changes when using TUS alone.
Optison, or other USCA have been used by different
research groups to enhance US transfection.11,15 In the
case of TUS, most of the studies argue that the addition
of USCA is crucial to obtain efficient transfections. These
USCA are suggested to lower the cavitation threshold,10
Gene Therapy
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
170
Table 1
Surface characterization analysis of the outer cytoplasmatic membrane of cells exposed to TUS evaluated by AFM scans
Features width (W) and height (H)a (nm)
Cells
(no TUS)
Fix immediately
post TUS
Fix 5 h
post TUS
Fix 24 h
post TUS
pKpKb (nm)
No Optison
With Optison
No Optison
W: 100–800,
H: 20–140
W: 100–200,
H: 10–15
W: 200–500,
H: 20–80
W: 100–700,
H: 30–120
W: 200–1000
H: 20–250
W: 100–500
H: 100–800
W: 200–800
H: 50–700
W: 200–800
H: 50–150
640720
Rac (nm)
With Optison
668718
d
d,e
No Optison
7677
d
With Optison
8677
103713d,e
701722
4576
530736d
590729d
42713
9379e
581748
623757
6379
78711
524753
a
Features width (W) and height (H) were determined using cross-section analysis.
pkpk (peak-to-peak) is the height difference between the higher point and the lower point in the field.
c
Ra is the average roughness. Values are mean7s.e.m. of six different cells scanned at 5 5 mm.
d
Po0.05 vs no TUS.
e
Po0.05 vs no Optison.
b
thus increasing cell permeability and distributing the
DNA to more cells.11,13,16 Nevertheless, the possible
involvement of other mechanisms and the relative role
of USCA in the process of gene delivery still remains
limited.9 Moreover, most methods applied for evaluating
the bioeffects of USCA and their possible role in the
cavitation process are based on secondary effects, such as
cell lyses, creation of sonoluminance, creation of peroxides and uptake of molecules by cells in vitro and
in vivo.13,22–24
In the present study, adding Optison during TUS
application resulted with a significant increase in
transfection level (1.5-fold) and efficiency (two-fold)
compared to transfection with TUS alone. Moreover,
using TUS–Optison led to a transfection efficiency, which
was significantly higher (two-fold) than the one achieved
with Lipofectaminet. Thus, more cells were transfected
when using Optison. The use of Optison with TUS also
resulted with an increase of 16% in the percentage of cells
containing DNA inside the cytoplasm. Similarly, a 14%
increase was obtained in the percentage of cells containing DNA in the nucleus compared to TUS alone. FACS
analyses demonstrated that Optison increased the
number of plasmid copies per cell. These data support
the hypothesis that using Optison with DNA during TUS
application results with higher distribution of the DNA
to more cells as well as increasing cell membrane
permeability, which was also suggested by other studies.20,25 Yet, the use of Optison did not affect the kinetics
of protein expression, ruling out the possibility that
Optison affects DNA trafficking to the nucleus. Another
interesting finding was that when increasing TUS time
application from 10 to 30 min, a significant increased in
transfection level was achieved, regardless of Optison
addition. This observation, together with the observation
that adding DNA post-TUS–Optison did not lead to a
significant increase in the number of cells having DNA,
strengthen our hypothesis that TUS by itself plays a
major role in delivering DNA to the nucleus.
The bioeffects attributes by the use of Optison when
compared to TUS alone were further studied using AFM.
The addition of Optison during TUS application led to
membrane changes, which were different in nature and
size from the ones achieved without the addition of
Optison. When using TUS–Optison, height and roughGene Therapy
ness differences were significantly higher than the ones
obtained when cells were exposed to TUS alone. It is
possible that the collapse of Optison during TUS
application8,16 gave an uneven patchy surface. Still these
data did not indicate for possible hole formation in the
membrane due to the TUS application. Several reports
showed the effect of TUS with USCA on cell membrane
using SEM.11,20,25 Mehier-Humbert et al.20 observed a
smoother surface post-TUS application with USCA in
MATBIII cells, but rougher surface in erythrocytes.
Taniyama et al.11,25 observed holes in cells immediately
post TUS with Optison. Using SEM, we also detected
hole-like formation in the cell membrane (Supplementary Figure 2), however, such holes were seen in the
control cells as well as in cells exposed to TUS. The AFM
analyses and the SEM studies raises the possibility that
we may be looking at additional mechanisms, rather
than hole formation, which is involved in TUS-mediated
gene delivery with Optison. Moreover, taking in account
the low mechanical index (MI) (0.159) obtained in our
studies and the therapeutic mode used, which is less
expected to induce cavitation,10 it is more likely that TUS
alone induces local shear forces and/or acoustic microstreaming rather than cavitaion. On the other hand, it
is possible that when using Optison concurrent with TUS
a cavitation-based mechanism may be involved.8,11,15
Nevertheless, further studies need to be performed to
strengthen this hypothesis.
Finally we have demonstrated that TUS bioeffcts, with
or without Optison are reversible. DNA added 5 h postTUS almost did not enter the cell cytoplasm, and AFM
revealed that cell membrane changes were reversed in a
period of 24 h post-TUS, which is consistent with the
findings of Taniyama et al.11,25 These findings further
support the safeties in using TUS for gene delivery
although it is also needed to be confirmed in vivo.
In summary, these in vitro studies support the major
role of TUS in the gene-delivery process. TUS affect DNA
trafficking to cell nucleus, demonstrating that long-term
application of TUS is necessary to achieve significant
increase in transfection and protein expression. Adding
Optison further contribute to this process particularly
when TUS, Optison and DNA are used simultaneously.
Overall the findings of the present study should be also
taken into account when considering TUS, with or
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
without Optison, for in vivo application. A long-term
application of TUS may increase transfection level and
efficiency as opposed to the short-time application
currently used. Finally, the use of a novel approach,
AFM, for the direct imaging and quantification of the
bioeffects add incremental data to the understanding of
the role of TUS in the gene-delivery process in vitro.
Materials and methods
Plasmids
Two reporter plasmids: pGL3-Luc (Promega) and
pIRES2-EGFP (Clontech) were used for the transfection
studies. Plasmids amplification and purification was
performed using a JET-Star kit (Genomed) according to
the manufacturer’s protocol. Plasmid concentrations
were measured using absorbance at 260 nm. A pGeneGrip plasmid (pGG, Gene Therapy Systems19), labeled
with rhodamin and encoding for GFP was used for DNA
intracellular tracking experiments.
Cell culture
Baby hamster kidney cells were grown in DMEM-F12
(Biological Industries, Israel) supplemented with 10%
FCS, 1% penicillin/streptomycin solutions (Biological
Industries) and fungizone (Gibco, Life Technologies).
The cell cultures were maintained at 371C and 5% CO2.
TUS apparatus
TUS with a 1 MHz applicator and a 2 cm2 surface area
probe (UltraMax XLTEK, Canada) was used for all studies.
The coupling quality and total energy delivered were
monitored at all times. Temperature was measured by a
needle tip thermometer. The acoustic pressure and the
amount of voltage were measured using a hydrophone
(Specialty Engineering Associates) connected to an oscilloscope (Tektronix). The mechanical index (MI) at all TUS
parameters was calculated according to the equation:10
qffiffiffi
MI ¼ P= f
where f is the ultrasonic frequency (MHz) and P is the
refractory acoustic pressure (MPa) measured by the
hydrophone (detailed description of TUS apparatus in
Supplementary data, Figure 3).
In vitro gene transfection
Cells were seeded in six-well plates at a density of
1 105 cells/well, and incubated for 24 h. DNA was
added to the cells at concentration of 7.5 mg/ml for pGLLuc or pEGFP and 0.5 mg/ml for pGG, and the well was
applied with 4 ml of medium. The ultrasonic probe was
immersed directly into the well, and placed 0.5 cm above
the cells. The plate was placed on rubber platform to
absorb reflected waves. For USCA studies, 10% (v/v)
Optison (human albumin microspheres, Amersham
Health), with or without DNA, was added to the cells
before TUS application. In the control experiments, cells
received DNA alone or DNA with only Optison.
Optimization of TUS was performed using 20–50% DC,
0–2 W/cm2 (correspond to 0.045–0.159 MI) and exposure
times of 0–40 min. In all other studies, unless otherwise
stated, TUS was applied at 2 W/cm2, 30% DC for 30 min.
Luciferase activity was determined 3 days post-TUS
application using TD20/20 Luminometer (Turner
Designs). For each sample, total protein was detected
using BCA Protein Assay Reagents (Pierce Biotechnology). Luciferase activity is reported as relative light units
divided by total protein measured for each sample. Cells
expressing GFP were observed under fluorescent microscope. Transfection efficiency was evaluated 24 h postTUS, as the number of cells expressing GFP per total
number of cells, and by flow cytometry (FACS Calibur,
BD). For FACS analysis, 2 104 cells were analyzed. For
each parameter, cell viability was measured using MTT
assay (Sigma) 2 h post-TUS application. Cell viability is
presented as the percentage of viable cells post-TUS
application relative to control – with no TUS, which is
indicated as 100% viability.
171
Effect of TUS on DNA localization using CLSM
Cells seeded on cover-slide glass (Lab-Tek) were supplemented with pGG and TUS was applied with or without
Optison. At different time points post-TUS application
(immediately, 5, 10 and 24 h), cells were washed in PBS
and fixed in 4% paraformaldehyde. Cells with pGG alone
or pGG with Optison, without the application of TUS
served as the controls. Cell nuclei were stained with
Draq-5t (Biostatus, 0.5 ml/ml) and cells were mounted
using Fluoromount-G (Electron Microscopy Science).
Confocal analysis was performed using a Bio-Rad 1024
confocal microscope (MRC 1000, Hercules, CA, USA)
and micrographs were taken across 8 mm of the z range at
steps of 0.5 mm to determine the location of the DNA
inside the nucleus or the cytoplasm. Data were analyzed
using CAS software (BioRad).
Effect of TUS-time application on DNA localization
in cells
Cells were exposed to TUS for 0, 10, 20 or 30 min in
the presence of pGG. Immediately post-TUS application,
cells were washed in PBS and either fixed in 4%
paraformaldehyde for confocal studies, or analyzed by
FACS. For FACS analysis, 2 104 cells were analyzed.
Kinetics of protein expression post-TUS transfection
Transfection efficiency was evaluated by FACS at
different time points post-TUS application, and by
counting the number of cells expressing GFP under
fluorescent microscope. For FACS studies, 2 104 cells
were analyzed. Percentage of normalized cells expressing GFP was calculated as the number of cells
expressing GFP at each time point divided by the
maximum number of cells expressing GFP and multiplied by 100. Since no significant differences in the
percentage of normalized cells expressing GFP was
observed between the two methods (o10%), the data
presented is from the FACS analyses.
DNA entrance to cell post-TUS
pGG was added to the cells immediately, 2 and 5 h postTUS application. Cells were fixed 24 h later in 4%
paraformaldehyde and were analyzed using CLSM as
described above.
Quantification of DNA in cells
The number of cells containing pGG in the different
studies was quantified using Image-Pro (Media Cybernetics). Cells were divided into three groups: (I) Cells
containing pGG in cytoplasm and nucleus, (II) Cells
Gene Therapy
TUS-DNA delivery: bioeffect using microscopy methods
M Duvshani-Eshet et al
172
containing pGG in the nucleus and (III) Cells with no
plasmid. For each time point, the number of cells
containing pGG (cytoplasm and nucleus) or nucleus
was evaluated and divided by the total number of cells in
a field. The relative fluorescence of cells containing pGG
in all the different studies was evaluated by FACS.
Sample preparation and scanning protocol for AFM
Cells exposed to TUS with or without the addition of
Optison, fixed immediately, 5, 10 and 24 h post-TUS with
3% gluteraldehyde and dehydrated in ethanol series.
AFM scans were performed in a tapping mode, in air,
using cantilevered optic fiber AFM probe with a tip
diameter of 20 nm (Nanonics, Israel) and imaged in
Nanonics AFM/NSOM/Confocal-100 system. Cells were
scanned at magnitudes of 50 50 and 25 25 mm to
visualize the entire cells, and at 10 10, 5 5 and
2.5 2.5 mm, to detect surface changes in cell membrane.
Quartz software (Cavendish Instrument, UK) was
used to analyze cell membrane surface topography and
presented as peak-to-peak (pk–pk), and surface roughness (Ra). Cross-section analyses were used to measure
the dimensions of typical topographic features in the
x-, y-, z-axes.
Statistical analysis
All of the data are presented as mean value7standard
error of the mean (s.e.m.). All transfection conditions
were performed in six repeats and each experiment was
repeated at three separate occasions. Confocal micrographs are representatives from three different experiments and six random fields. For each experiment, FACS
analyses were performed four times. AFM scans are
representatives of six cells scanned at each parameter.
Statistical significance was determined using t-test for
independent sample.
Acknowledgements
This work was in part supported by The Israel Science
Foundation (ISF) to Marcelle Machluf.
References
1 Fechheimer M, Boylan J, Parker S, Sisken J, Patel G, Zimmer S.
Transfection of mammalian cells with plasmid DNA by scrape
loading and sonication loading. Proc Natl Acad Sci USA 1987; 84:
8463–8467.
2 Kim HJ, Greenleaf JF, Kinnick RR, Bronk JT, Bolander ME.
Ultrasound mediated transfection of mammalian cells. Hum
Gene Ther 1996; 7: 1339–1346.
3 Tata DB, Dunn F, Tindall DJ. Selective clinical ultrasound signals
mediate differential gene transfer and expression in two human
prostate cancer cell lines: LnCap and PC-3. Biochem Biophys Res
Com 1997; 234: 64–67.
4 Bao S, Thrall BD, Miller DL. Transfection of a reporter plasmid
into cultured cells by sonoporation in vitro. Ultrasound Med Biol
1997; 23: 953–959.
5 Lawrie A, Brisken AF, Francis SE, Tayler DI, Chamberlain J,
Crossman DC et al. Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro. Circulation 1999;
99: 2617–2620.
6 Miller MW, Miller DL, Brayman AA. A review of in vitro
bioeffects of inertial ultrasonic cavitation from a mechanistic
perspective. Ultrasound Med Biol 1996; 22: 1131–1154.
7 Guzman HR, McNamara AJ, Nguyen DX, Prausnitz MR.
Bioeffects caused by changes in acoustic cavitation bubble
density and cell concentration: a unified explanation based on
cell-to-bubble ratio and blast radius. Ultrasound Med Biol 2003;
29: 1211–1222.
8 Ng KY, Liu Y. Therapeutic ultrasound: its application in drug
delivery. Med Res Rev 2002; 22: 204–223.
9 Miller DL, Pislaru SV, Greenleaf JE. Sonoporation: mechanical
DNA delivery by ultrasonic cavitation. Somat Cell Mol Genet
2002; 27: 115–134.
10 Apfel RE, Holland CK. Gauging the likelihood of cavitation from
short-pulls, low-duty cycle diagnostic ultrasound. Ultrasound
Med Biol 1991; 17: 179–185.
11 Taniyama Y, Tachibana K, Hiraoka K, Aoki M, Yamamoto S,
Matsumoto K et al. Development of safe and efficient novel
nonviral gene transfer using ultrasound: enhancement of
transfection efficiency of naked plasmid DNA in skeletal muscle.
Gene Therapy 2002; 9: 372–380.
12 Brayman AA, Coppage ML, Vaidya S, Miller MW. Transient
poration and cell surface receptor removal from human
lymphocytes in vitro by 1 MHz ultrasound. Ultrasound Med Biol
1999; 25: 999–1008.
13 Lawrie A, Brisken AF, Francis SE, Tayler DI, Wyllie D, Kiss-Toth
E et al. Ultrasound-enhanced transgene expression in vascular
cells is not dependent upon cavitation-induced free radicals.
Ultrasound Med Biol 2003; 29: 1453–1461.
14 Dalecki D. Mechanical bioeffects of ultrasound. Annu Rev Biomed
Eng 2004; 6: 229–248.
15 Lawrie A, Brisken AF, Francis SE, Cumberland DC, Crossman
DC, Newman CM. Microbubble-enhanced ultrasound for
vascular gene delivery. Gene Therapy 2000; 7: 2023–2027.
16 Unger EC, Hersh E, Vannan M, McCreery T. Gene delivery
using ultrasound contrast agents. Echocardiography 2001; 18:
355–361.
17 Alonso JL, Goldmann WH. Feeling the forces: atomic force
microscopy in cell biology. Life Sci 2003; 72: 2553–2560.
18 Munkonge FM, Dean DA, Hillery E, Griesenbach U, Alton EW.
Emerging significance of plasmid DNA nuclear import in gene
therapy. Adv Drug Del Rev 2003; 55: 749–760.
19 Zelphati O, Liang X, Hobart P, Felgner PL. Gene chemistry:
functionally and conformationally intact fluorescent plasmid
DNA. Hum Gene Ther 1999; 10: 15–24.
20 Mehier-Humbert S, Bettinger T, Yan F, Guy RH. Plasma
membrane poration induced by ultrasound exposure: Implication for drug delivery. J Control Rel 2005; 104: 213–222.
21 Tachibana K, Uchida T, Ogawa K, Yamashita N, Tamura K.
Induction of cell-membrane porosity by ultrasound. Lancet 1999;
353: 1409.
22 Cochran SA, Prausnitz MR. Sonoluminescence as an indicator of
cell membrane disruption by acoustic cavitation. Ultrasound Med
Biol 2001; 27: 841–850.
23 Shohet RV, Chen S, Zhou YT, Wang Z, Meidell RS, Unger RH
et al. Echocardiographic destruction of albumin microbubbles
directs gene delivery to the myocardium. Circulation 2000; 101:
2554–2556.
24 Skyba DM, Price RJ, Linka AZ, Skalak TC, Kaul S. Direct in vivo
visualization of intravascular destruction of microbubbles by
ultrasound and its local effects on tissue. Circulation 1998; 98:
290–293.
25 Taniyama Y, Tachibana K, Hiraoka K, Namba T, Yamasaki K,
Hashiya N et al. Local delivery of plasmid DNA into rat carotid
artery using ultrasound. Circulation 2002; 105: 1233–1239.
Supplementary Information accompanies the paper on the Gene Therapy website (http://www.nature.com/gt).
Gene Therapy