Journal of Controlled Release xxx (2012) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Journal of Controlled Release
journal homepage: www.elsevier.com/locate/jconrel
Low-frequency ac electroporation shows strong frequency dependence and yields
comparable transfection results to dc electroporation
Yihong Zhan a, Zhenning Cao b, Ning Bao c, d, Jianbo Li e, Jun Wang a, Tao Geng a, Hao Lin e, Chang Lu b, c,⁎
a
Birck Nanotechnology Center, Purdue University, West Lafayette, IN 47907, USA
School of Biomedical Engineering and Sciences, Virginia Tech-Wake Forest University, Blacksburg, VA 24061, USA
Department of Chemical Engineering, Virginia Tech, Blacksburg, VA 24061, USA
d
School of Public Health, Nantong University, Nantong, 226019, PR China
e
Department of Mechanical and Aerospace Engineering, Rutgers University, Piscataway, NJ 08854, USA
b
c
a r t i c l e
i n f o
Article history:
Received 11 January 2012
Accepted 1 April 2012
Available online xxxx
Keywords:
Gene delivery
Electroporation
Alternating current
Direct current
Frequency
a b s t r a c t
Conventional electroporation has been conducted by employing short direct current (dc) pulses for delivery
of macromolecules such as DNA into cells. The use of alternating current (ac) field for electroporation has
mostly been explored in the frequency range of 10 kHz–1 MHz. Based on Schwan equation, it was thought
that with low ac frequencies (10 Hz-10 kHz), the transmembrane potential does not vary with the frequency.
In this report, we utilized a flow-through electroporation technique that employed continuous 10 Hz-10 kHz
ac field (based on either sine waves or square waves) for electroporation of cells with defined duration and
intensity. Our results reveal that electropermeabilization becomes weaker with increased frequency in this
range. In contrast, transfection efficiency with DNA reaches its maximum at medium frequencies
(100–1000 Hz) in the range. We postulate that the relationship between the transfection efficiency and the
ac frequency is determined by combined effects from electrophoretic movement of DNA in the ac field, dependence of the DNA/membrane interaction on the ac frequency, and variation of transfection under different electropermeabilization intensities. The fact that ac electroporation in this frequency range yields high
efficiency for transfection (up to ~ 71% for Chinese hamster ovary cells) and permeabilization suggests its potential for gene delivery.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
Electroporation has been demonstrated as an efficient physical
tool for delivering genes into mammalian cells [1–4]. Electroporation
dramatically and transiently increases the permeability of the cell
membrane under the influence of an external electric field [5–7].
Macromolecules such as DNA are able to migrate into cells during
the process. The vast majority of electroporation techniques are
based on application of short intense pulses of direct current (dc)
field (rectangular or exponential decay pulses). Similar dc-based approach has also been miniaturized on microfluidic platforms for cell
electroporation or electrofusion [8–22]. In comparison, the reports
on experimental work using alternating current (ac) field for electroporation have been scarce with even fewer papers presenting
systematic investigations on gene delivery [20,23–26]. Electrofusion
of red blood cells and electroporation of eukaryotic cell lines by
⁎ Corresponding author at: Department of Chemical Engineering, Virginia TechBlacksburg, VA 24061, USA. Tel.: + 1 540 231 8681; fax: + 1 540 231 5022.
E-mail address: [email protected] (C. Lu).
lifted oscillating fields were conducted and a strong dependence
on the field frequency was observed [23,24]. Successful fusion and
gene delivery were observed when the field frequency was 40–
100 kHz (with 10 kHz to 1 MHz studied). The authors concluded
that electroporation based on an oscillating field had superior performance than conventional dc pulse electroporation due to less
dramatic polarization and “sonicating effect”. In a separate study,
electroporation with bipolar square wave (60 kHz) pulses were
compared with electroporation done by unipolar and single square
pulses [25]. Transfection efficiency was markedly higher with bipolar pulses than with the other two wave forms due to symmetric
permeabilization of cells under bipolar conditions. The cell viability
after electroporation by either bipolar or unipolar pulses was substantially higher than by single square wave pulses. The authors
mentioned that increasing the frequency to 1 MHz decreased the
transfection efficiency. There was also a conflicting report that
revealed no dependence of electropermeabilization on the ac frequency in the range of 20 kHz–160 kHz when a dye such as calcein
was delivered [26].
Schwan equation was proposed to describe transmembrane potential Δψmembr induced by an oscillating ac field [27,28] and has
been used to analyze ac electroporation data in the frequency range
0168-3659/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconrel.2012.04.006
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
GENE DELIVERY
COREL-06268; No of Pages 7
GENE DELIVERY
2
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx
of 10 kHz–1 MHz (with the assumptions of the cell being a sphere
and the plasma membrane being purely dielectric).
1:5aE cosθ
Δψmembr ¼ 1=2
1 þ ðωτÞ2
ð1Þ
where a is the radius of the cell, θ is the angle between the field line
and a normal from the center of the cell to a point of interest on the
cell membrane, ω is the angular frequency of the external field
(ω = 2πf with f being the frequency of the ac field), E = E 0 sin(ωt + ϕ)
with E 0 being the amplitude of the external field, t being time and
ϕ being the initial phase. The relaxation time τ ¼ aC membr ρint þ ρ2ext
where Cmembr is the membrane capacitance per unit area, ρint and ρext
are the resistivity of the internal fluid and the external medium, respectively. The equation indicates that the transmembrane potential
Δψmembr in general decreases with higher ac frequency, given that the
amplitude of the field is kept constant. Furthermore, based on the
equation the decline of Δψmembr with the frequency is only pronounced
in the range of 104–107 Hz [28,29]. When the frequency is below
10 kHz, Δψmembr effectively does not change with the frequency, due
to the relatively small value of (ωτ) 2.
There has not been systematic experimental study of electropermeabilization and gene delivery by low-frequency (b10 kHz) ac
field in the literature. Based on Eq. (1), low-frequency ac (b10 kHz)
produces higher transmembrane potential than high-frequency ac
(10 kHz–1 MHz). Furthermore, low-frequency ac (50–60 Hz) is widely
used in domestic power systems and may provide drastically simple
solution to power need by electroporation. Ac pulses for electroporation
have conventionally been generated by triggering an ac generator using
discrete dc pulses of several milliseconds or shorter. In this work,
we generated electropermeabilization and DNA delivery using lowfrequency sine-wave or square-wave ac field of 10 Hz-10 kHz by modulating a constant ac field across a microfluidic channel with alternating
wide and narrow sections. Similar to flow-through electroporation
based on dc voltage [30–32], the ac field intensity in the narrow section
was sufficiently high to generate electroporation while the low field in
the wide sections did not disrupt the membrane. Our results indicate
that electropermeabilization becomes more intensive when the ac
frequency decreases in the range of 10 Hz-10 kHz, arguing against the
common belief that Δψmembr does not vary with the ac frequency in
this range. DNA transfection efficiency increases first with the ac frequency and reaches the maximum around 100–1000 Hz before decreasing with further increase in the frequency. The transfection efficiency by
ac electroporation is up to ~71% for Chinese hamster ovary cells after
first-round optimization and this result is very comparable to that of
dc electroporation. We provide mechanistic discussion about the factors
that affect the relationship between the transfection efficiency and the
ac frequency. We believe that low-frequency ac field of 10 Hz-10 kHz offers promising potential for gene delivery.
2. Experimental
The experimental procedures are detailed in the Supplementary
content.
3. Results
We applied a microfluidic flow-through electroporation design to
create cell electroporation by ac field (sine wave or square wave) of
various frequencies (10 Hz-10 kHz) with durations of milliseconds
(1–3 ms). Flow-through electroporation based on constant dc voltage
was applied to electroporate flowing cells in our previous works
[30–32]. As shown in Fig. 1, in this study cells flowed through a
microfluidic channel with wide and narrow sections while a constant
ac field was established across the channel. Due to the difference in
the cross-sectional area between wide and narrow sections, the ac
Fig. 1. The layout of the microfluidic flow-through electroporation device based on
low-frequency ac field. Cells flow through the channel while a constant ac field is
established across the channel. A microfluidic channel with geometric variation
(700 μm wide in the wide sections, 70 μm wide in the narrow section, and 60 μm
deep) was used to produce electroporation in the narrow section. The inset images
show the field variation of ac voltage across the microfluidic channel and over the
time in the cases of sine (left) and square (right) waves.
field intensity in a particular section is inversely proportional to its
cross-sectional area (thus its width when the channel depth is uniform in the device). The ac field variation with time and along the
channel length is indicated in the inset schematics of Fig. 1. In our design, the field intensity in the wide sections was roughly 1/10 of that
in the narrow section (with the width of 700 μm for the wide sections
and 70 μm for the narrow section). Thus the root mean square (rms)
field intensity in the wide sections was substantially lower than the
electroporation threshold and the electroporation occurred exclusively in the narrow section. The microfluidic device functioned as a
resistor due to the very low capacitance and did not produce any alteration to the ac signal. All electroporation was conducted in an electroporation buffer with low ionic strength (8 mM Na2HPO4, 2 mM
KH2PO4, and 250 mM sucrose, pH = 7.2, ρ = 6.25 Ω m). In here, for
the convenience of reference, we refer to the residence time in the
narrow section(s) (determined by the section length and cell velocity) as the electroporation duration (strictly speaking, the electroporation duration should be a fraction of the residence time during which
the field intensity is higher than the electroporation threshold). The
electroporation duration was determined by the dimensions of the
narrow section(s) and the flow rate. It is worth noting that with
very low frequencies (e.g. b100 Hz), the time period of the ac field
(i.e. the time taken by the waveform to complete one cycle) is comparable to or even longer than the electroporation duration. As the result, an individual cell may experience only partial ac wave. The
uniformity of the treatment at the single cell level may be affected
in this circumstance. However, an electroporated cell population collectively still experiences full characteristics of the waveform due to
the fact that cells continuously enter the narrow electroporation section at random initial phases for a period (several minutes) that is
much longer than the ac time period.
The expansion in the cell size is a very distinct and visual phenomenon that can be readily observed during or after electroporation, as
previously described by us and other groups [30,33–35]. As shown
in Fig. 2a, we took a series of snap shots of the same cell flowing
into the narrow section under sine-wave ac field of a certain frequency using phase contrast imaging. The cells had bright and welldefined boundaries before entering the narrow section. Upon entering the narrow section with the strong ac field (the rms field intensity
at ~773 V/cm) with very low frequencies (27–379 Hz), the cells experienced pronounced increase in their diameters that was also associated
with blurring of their outer boundaries. While similar phenomenon
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
3
among different frequencies and also to that produced by dc field of
the same intensity and duration. The data in Fig. 2c indicate that the
electropermeabilization efficiency gradually decreases with the increasing ac frequency in the 27–4870 Hz range. The lower frequency
(b100 Hz) generated similar results to that of dc electroporation.
These results are consistent with the cell expansion and calcein release
data.
DNA delivery is the most important application for electroporation. Successful electroporation-based DNA delivery into cells (or
transfection) requires efficient electropermeabilization of the membrane that allows DNA insertion into the membrane and entry into
cells [36–38], effective transport of DNA into the nucleus, and finally
sufficient expression of DNA into the protein it encodes. Because cell
transfection involves multiple steps besides the electropermeabilization
of the membrane [7], the transfection efficiency (defined by the percent
of cells that expresses the DNA among living cells in here) may not be
determined solely by the electropermeabilization efficacy.
The threshold for electroporation-based transfection was around
500 V/cm with sine-wave ac of 379 Hz (note that this is the rms
field intensity in the narrow section) (Supporting information Fig.
S2). This means that a peak field intensity of ~ 707 V/cm is needed
for CHO cell electroporation by ac field, which is substantially higher
than 300–400 V/cm threshold found for dc electroporation [30]. The
transfection efficiency first increased with higher field intensity before it decreased when the ac field intensity increased further (Supporting information Fig. S2). Similar observation of low gene expression
by live cells after high-intensity electroporation has been reported for
dc electroporation [31,39,40].
Fig. 3a demonstrates that the transfection efficiency (measured
48 h after electroporation) varies significantly with the sine-wave ac
Fig. 2. The electropermeabilization of cells by sine-wave ac electroporation of different
frequencies. In all cases, the rms field intensity in the narrow section is ~ 773 V/cm. The
scale bars are 25 μm. (a) Cell expansion due to ac electroporation observed by phase
contrast imaging. Time-lapse images of the same flowing cell (taken at an interval of
33 ms) were stitched together. (b) Calcein release due to ac electroporation observed
by fluorescence imaging. Time-lapse images of the same flowing cell (taken at an interval of 100 ms) were stitched together. (c) Permeability examined by SYTOX delivery by
ac electroporation. Permeability was calculated as the percent of fluorescent cells
among the total cells. For all frequencies, the electroporation duration was 2 ms. The
dc electroporation had a field intensity of 773 V/cm and a duration of 2 ms. All data
points are based on triplicate experiments, error bars represent +/− standard
deviation.
could still be seen with 3855 Hz ac field to a lower degree, there was no
change with the cell size and appearance when the ac field had a frequency of 10 kHz. The cells observed all had similar sizes. These phase
contrast images indicate that electropermeabilization under an ac
field of the same amplitude exhibits a decreased efficiency with increasing frequency in the 27–10 kHz range. This is further confirmed by
Fig. 2b which shows electroporation-based release of impermeant calcein under different frequencies. Such release of the fluorescent dye is
indicative of the electropermeabilization of the membrane. Calcein release was very rapid with lower frequencies (27–3855 Hz) but could
not be observed at 10 kHz. We also examined the ability of sine-wave
ac electroporation to deliver impermeant molecules into cells. SYTOX
is a nucleic acid stain that is impermeant to live cells and becomes fluorescent once binding to DNA. In our experiment, CHO cells and SYTOX
flowed through the electroporation device in Fig. 1 together under an
rms field intensity of ~773 V/cm in the narrow section and an electroporation duration (i.e. the residence time in the narrow section) of
2 ms. The percentages of fluorescently labeled cells were compared
Fig. 3. The transfection of CHO-K1 cells (by the plasmid vector pEGFP-C1) using sinewave ac electroporation under various frequencies in the range of 10 Hz-10 kHz. For
all ac frequencies, the rms field intensity in the narrow section was 773 V/cm. The
field intensity of dc electroporation was also 773 V/cm. All data points are based on
triplicate experiments, error bars represent +/− standard deviation. (a) The variation
of the transfection efficiency (the percent of EGFP-expressing cells among the viable
cells) over different ac frequencies; (b) The variation of the cell viability (the percent
of viable cells among the total cells) over different ac frequencies.
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
GENE DELIVERY
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx
GENE DELIVERY
4
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx
frequency under different electroporation durations (1, 2, and 3 ms)
and constant ac field intensity (~ 773 V/cm, rms). The transfection efficiency first increased with higher frequency until it peaked in the
range of 300–1000 Hz. The transfection efficiency declined with further increase in the ac frequency. The same trend was observed
with all electroporation durations. At the low end of the frequency
range (b100 Hz), the transfection efficiency under different electroporation durations generally followed the sequence of 1 ms> 2 ms> 3 ms.
With higher frequencies, however, the longer durations (2 or 3 ms)
were favored for more effective transfection. The maximal transfection
efficiencies (30% with 1 ms duration, 42% with 2 ms, and 31% with
3 ms) achieved by ac electroporation were consistently higher than
those generated by dc electroporation of the same duration and intensity (17% with 1 ms, 16% with 2 ms, and 24% with 3 ms). The best performance achieved here (transfection efficiency of 42% and viability of
89%) also compares favorably with that of a single-narrow-section
device under constant dc voltage optimized previously (transfection efficiency of 25% and viability of 30%) [40]. The viability of cells (measured
48 h after electroporation using propidium iodide exclusion) was well
preserved with little variance among different frequencies when the
duration was 1 or 2 ms (96.3% with a S.D. of 1.3% for 1 ms and 92.3%
with a S.D. of 4.4% for 2 ms) (Fig. 3b). The average viability dropped
to 84.8% with the electroporation duration of 3 ms. The lack of strong
correlation between the cell viability and the ac frequency suggests
V 2rms
that the viability
pffiffiffi is mostly affected by Joule heating (P Joule ¼ R with
V rms ¼ V Peak = 2 for sine voltage without dependence on the frequency) instead of the degree of electropermeabilization (which varies
with the frequency). With 3 ms duration and 773 V/cm rms intensity,
the temperature roughly increased by 0.7 °C in the narrow section by
estimation. This agrees with our recent results that there was no substantial decrease in the cell viability even when electroporation occurred to drastically larger cell membrane area as long as the electric
parameters (field intensity and duration) were maintained constant
[41].
We also tested electroporation based on square-wave ac voltage.
We first compared transfection by square-wave ac (Fig. 4) with that
by sine-wave ac (Fig. 3), both in a microfluidic channel with a single
narrow section. We have observed very similar trends in these two
cases. In both cases, the transfection efficiency increased with higher
ac frequency at the low end (b1000 Hz) and decreased with further
increase in the frequency in the range of 1–10 kHz. The cell viability
after electroporation presented no apparent dependence on the ac
frequency. The best transfection efficiency (~39%) produced by
square-wave ac (~1093 V/cm, rms) electroporation in this case is
similar to that (~42%) generated by sine-wave ac electroporation of
lower field intensity (~773 V/cm, rms). This suggests that squarewave ac would be less effective for transfection than sine-wave ac
with the same rms field intensity.
In general, high-efficiency transfection is only practical with multipulse electroporation [39,42]. In the case of flow-through electroporation, this requires a microfluidic channel having multiple narrow sections and fairly high voltage [31,40]. Our square-wave ac power supply
provided peak-to-peak voltage up to 800 V. We conducted squarewave ac electroporation with a microfluidic channel with 4 narrow
sections (shown in Fig. 5a) and the total residence time in these narrow
sections of ~2–6 ms (i.e. ~0.5–1.5 ms in each narrow section). The transfection efficiency increased with both higher field intensity and longer
electroporation duration (Fig. 5b). The cell viability decreased under
the same conditions (Fig. 5c). With the multi-narrow-section configuration, the highest transfection efficiency achieved was 71% (with a cell
viability of 58%) under the rms intensity of 984 V/cm in the narrow sections and a total electroporation duration of 5.4 ms. We believe that
further improvement in the transfection efficiency and the viability
is possible after more optimization of the parameters. The overall performance of ac electroporation is comparable to that of dc electroporation
for DNA delivery with similar devices [31].
Fig. 4. The transfection of CHO-K1 cells (by the plasmid vector pEGFP-C1) under
square-wave ac of various frequencies in the range of 10 Hz-10 kHz with the duration
of 2 ms and the rms field intensity of 1093 V/cm in the narrow section. The microfluidic
channel (with one narrow section) in Fig. 1 was used in this experiment. All data points
are based on triplicate experiments, error bars represent +/−standard deviation. (a) The
variation of the transfection efficiency (the percent of EGFP-expressing cells among viable
cells) over different frequencies; (b) The variation of the cell viability (the percent of
viable cells among the total cells) over different ac frequencies.
4. Discussion
Schwan equation (Eq. (1)) has been conventionally used for evaluation of the transmembrane potential Δψmembr induced by ac field.
When we use typical values (a = 5 μm; Cmembr = 1 μF cm − 2; ρint =
5 Ω m; and ρext = 6.25 Ω m [43]) for the estimation, we obtain a relaxation time τ of ~ 0.41 μs. This means that [1 + (ωτ) 2] 1/2 and Δψmembr
vary only by ~0.03% in the frequency range of 10 Hz-10 kHz. However, in both the electropermeabilization and the transfection experiments, we observed strong dependence on the ac frequency. The
results indicate that the electropermeabilization declines with increasing ac frequency in the range studied and the transfection efficiency peaks in between 100 and 1000 Hz. The frequency
dependence involved in the membrane charging process does not adequately account for the frequency response we observed. In addition, previous literature on ac electroporation has suggested
“sonicating effect” as a factor that affects the performance in the
range of 40–100 kHz [23,24]. However, such effect is unlikely to be
important for our results due to the very low frequency used here.
The data in Fig. 2 strongly point to the conclusion that higher transmembrane potential was generated at lower frequencies in the
range (judged by the increased permeability). Thus we raise the possibility that Schwan equation does not adequately describe the transmembrane potential induced by ac field in the frequency range of
10 Hz-10 kHz. Among possible reasons, Schwan's derivation assumed
zero dielectric permittivity for the cytoplasm and the extracellular
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
5
Fig. 6. The electrophoretic effects on DNA delivery in a low-frequency ac field. (a) Schematic of DNA movement in a low-frequency sine-wave ac field. An ac cycle starts at t0.
The electric field changes direction at half cycle (t1). During the first half cycle (t0–t1),
DNA molecules surrounding a cell are swept against one side of the cell surface.
Such concentration of DNA molecules is largely reversed in the second half of
⇀
the cycle (t1–t2). (b) The variation of 〈Δl 〉 in the sine-wave ac frequency range of
1 Hz-10 kHz with different field durations (τ = 1, 2, and 3 ms).
Fig. 5. The transfection of CHO-K1 cells (by the plasmid vector pEGFP-C1) using
square-wave ac (91 Hz) in a channel with 4 narrow sections under various electroporation durations. All data points are based on triplicate experiments, error bars represent +/− standard deviation. (a) The design of the microfluidic channel (~ 60 μm in
the depth) used in this experiment. Each of the 3 half-circular wide sections has a
length of 5 mm along the centerline. The drawing is not to scale. (b) The transfection
efficiency under different electroporation durations (i.e. the total residence time in
the narrow sections) with the rms field intensity of 820 or 984 V/cm in the narrow sections. (c) The cell viability under different electroporation durations with the rms field
intensity of 820 or 984 V/cm in the narrow sections.
medium [27]. Such assumption may need corrections in the low frequency range we investigated [44].
In order to explain the relationship between the ac frequency and
the transfection efficiency, we first consider the electrophoretic effects on DNA distribution around cells under the oscillating ac field
using the case of sine-wave ac as the example. Such effect has been
suggested in previous works for analysis of DNA delivery by discrete
dc pulses [39,45] but has not been considered for ac electroporation
in a quantitative fashion. Similar argument would also be applicable
to the case of square-wave ac electroporation. Due to the mismatch
between the electrophoretic mobilities of cells and DNA, DNA molecules experience movement relative to cells during an ac cycle. As
shown in Fig. 6a, assuming that DNA has a higher mobility than a
cell (which is typically the case), the movement of DNA relative to
the cell under the ac field may create transient concentration of
DNA against one side of the cell. If the field is applied to the cell/
DNA suspension at the beginning of the cycle, such accumulation of
DNA at (or close to) the cell surface reaches its maximum at t1
when the electric field changes its direction and then the asymmetric
localization of DNA is reversed back to the initial distribution at the
end of the cycle (t2). In order to quantify the electrophoretic effects
of ac field of various frequencies, we examine the net movement of
DNA molecules relative to a cell under the oscillating ac field. We
focus on the process that occurs in the narrow electroporation section
during the electroporation duration (τ) because the electrophoretic
effect is strong due to the high field intensity in there and electrophoresis also coincides with electropermeabilization. With the electrophoretic mobility μ = L/Et, we have
8 τ > ⇀ ⇀⇀
>
< ∫ μ DNA − μ cell E ðt; ϕÞdt if ∃ t ∈½0; τ; ⇀
E ðt; ϕÞ ≥ ET
⇀
Δl ¼ 0
>
>
⇀
: 0
if ∀ t ∈½0; τ; E ðt; ϕÞ b ET
ð2Þ
⇀ ⇀0
⇀
0
with
E
ð
t;
ϕ
Þ¼
E
sin
ð
ωt
þ
ϕ
Þ
(
E ¼1093V=cm in our experiment) and
⇀
Δl being the change in the distance
between DNA and a cell due to
⇀
electrophoresis. Because we use Δl to characterize the enhancement
of gene delivery
into cells by electrophoretic effects, we artificially
⇀
designate Δl to be zero if electroporation does not occur during the
⇀
electroporation duration (i.e. E ðt; ϕÞ never exceeds the electroporation
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
GENE DELIVERY
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx
GENE DELIVERY
6
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx
threshold ET which is ~707 V/cm as suggested by the data in Supporting
μ
and
information Fig. S2). We use typical values in the literature for ⇀
DNA
⇀
μ cell (3× 10 − 4 and 2 × 10− 4 cm2V− 1 s − 1 in the same direction, respec-
tively) in our calculations [46,47]. Finally, because cells enter the electroporation
at random ϕ between 0 and 2π, the ensemble section
⇀
averaged Δl for the cell population
⇀
∫2π
0 Δl dϕ
⇀
Δl ¼
2π
ð3Þ
⇀
Δl is used here as a quantitative parameter to characterize, averagely for the cell population, how much
the electrophoretic effect is.
⇀
In Fig. 6b, we show the variation of Δl with ac frequency
under dif ⇀
ferent electroporation durations τ (1, 2, and 3 ms). Δl has peaks in
decreased the transfection efficiency due to insufficient permeabilization
at higher frequencies (confirmed by the fact that longer durations were
favored for high transfection at this end). Similar peak of the transfection
efficiency at the medium electropermeabilization condition was observed with dc-based electroporation [31,39,40]. Thus the dependence
of transfection efficiency on the ac frequency is likely a result of combined effects from multiple mechanisms.
Acknowledgments
This work was supported by the National Science Foundation
grants CBET 1016547, CBET 0967069, CBET 0747886, United States
Department of Agriculture grant USDA-NRI 2009-35603-05059, and
the National Natural Science Foundation of China Grant 21075070.
Appendix A. Supplementary data
the intermediate range of the frequencies investigated (167, 83, and
56 Hz for 1, 2, and 3 ms, respectively). A simple physical interpretation of this result is that, with very low frequencies and long cycle
times, there is a high probability for the electric field not to reach its
high values during the electroporation duration τ. On the other
hand, with high frequencies and short cycle times, the rapid reversal
of polarity does not allow accumulation of the electrophoretic effect
for a significant period. Based on the calculation, we suggest that
one important reason for the transfection efficiency to maximize in
the intermediate range of the frequencies is that the electrophoretic
effect (which pushes DNA against cell surface) is sensitive to the ac
frequency and becomes the most pronounced with medium frequencies.
Second, we consider how the ac field affects the DNA interaction
with the membrane. The formation of the DNA/membrane complex
is an important step involved in electroporation-based delivery [39,48].
Previous study indicated that once DNA gets into cells or forms strong
interaction with the membrane after dc pulse(s), a reversed field does
not release them [48]. A recent paper by Faurie et al. further demonstrated that when discrete dc pulses of reversed polarity were applied,
the reversal of the DNA/membrane complex by the opposite field was
significant only when the delay between pulses was less than 1 s [45].
Furthermore, they showed that shorter delays between pulses of opposite polarity increasingly made the reversal of the interaction easier. Our
results on the ac frequency dependence (Figs. 3a and 4a) are very consistent with these findings. When the ac frequency is low and the ac
time period is long (relative to the electroporation duration), the vast
majority of cells experience only partial ac wave with one polarity and
there is no reversal of the DNA/membrane interaction. However,
when the electroporation duration was 3–1 ms and the ac frequency
went beyond 300–1000 Hz, cells started to experience full ac wave(s)
with varying polarities. The reversal of DNA/membrane complex
started to occur and became stronger with higher ac frequency due to
the shorter time for the DNA/membrane complex formation. Although
by itself alone this theory still does not explain why the transfection
efficiency increased with higher frequency in the range of 10–300 Hz,
it provides an explanation to the decrease of the transfection efficiency
beyond the medium ac frequency region of 300–1000 Hz.
Finally, there may be other mechanisms that contribute to the fact
that the transfection efficiency has a peak at the intermediate frequencies. The transfection efficiency by ac electroporation is likely
to be highly sensitive to the degree of electropermeabilization. As
revealed by the data in Fig. 2, ac electroporation creates higher permeabilization at lower frequencies. However, Fig. 3 indicates that
the transfection efficiency reaches a peak around 300–1000 Hz. This suggests that the strong electropermeabilization at frequencies b200 Hz
was not optimal for transfection. Such conclusion is further supported
by the fact that short electroporation duration (1 ms) generated better
transfection results than longer durations (2 and 3 ms). After reaching
the optimal frequencies, further increase in the frequency to >4870 Hz
Supplementary data to this article can be found online at doi:10.
1016/j.jconrel.2012.04.006.
References
[1] E. Neumann, M. Schaefer-Ridder, Y. Wang, P.H. Hofschneider, Gene transfer into
mouse lyoma cells by electroporation in high electric fields, EMBO J. 1 (1982)
841–845.
[2] M.R. Prausnitz, V.G. Bose, R. Langer, J.C. Weaver, Electroporation of mammalian
skin: a mechanism to enhance transdermal drug delivery, Proc. Natl. Acad. Sci.
U. S. A. 90 (1993) 10504–10508.
[3] M.P. Rols, C. Delteil, M. Golzio, P. Dumond, S. Cros, J. Teissie, In vivo electrically
mediated protein and gene transfer in murine melanoma, Nat. Biotechnol. 16
(1998) 168–171.
[4] H. Aihara, J.-I. Miyazaki, Gene transfer into muscle by electroporation in vivo, Nat.
Biotechnol. 16 (1998) 867–870.
[5] J.C. Weaver, Y.A. Chizmadzhev, Theory of electroporation: a review, Bioelectrochem. Bioener. 41 (1996) 135–160.
[6] J. Teissie, M. Golzio, M.P. Rols, Mechanisms of cell membrane electropermeabilization: a minireview of our present (lack of ?) knowledge, Biochim. Biophys.
Acta 1724 (2005) 270–280.
[7] J.M. Escoffre, T. Portet, L. Wasungu, J. Teissie, D. Dean, M.P. Rols, What is (still not)
known of the mechanism by which electroporation mediates gene transfer and
expression in cells and tissues, Mol. Biotechnol. 41 (2009) 286–295.
[8] S.-W. Lee, Y.-C. Tai, A micro cell lysis device, Sens. Actuators A: Phys. 73 (1999)
74–79.
[9] Y. Huang, B. Rubinsky, Flow-through micro-electroporation chip for high efficiency single-cell genetic manipulation, Sens. Actuators A: Phys. 104 (2003)
205–212.
[10] Y.C. Lin, M. Li, C.C. Wu, Simulation and experimental demonstration of the electric
field assisted electroporation microchip for in vitro gene delivery enhancement,
Lab Chip 4 (2004) 104–108.
[11] H. Lu, M.A. Schmidt, K.F. Jensen, A microfluidic electroporation device for cell
lysis, Lab Chip 5 (2005) 23–29.
[12] M. Khine, A. Lau, C. Ionescu-Zanetti, J. Seo, L.P. Lee, A single cell electroporation
chip, Lab Chip 5 (2005) 38–43.
[13] Z. Fei, S. Wang, Y. Xie, B.E. Henslee, C.G. Koh, L.J. Lee, Gene transfection of mammalian cells using membrane sandwich electroporation, Anal. Chem. 79 (2007)
5719–5722.
[14] M. Khine, C. Ionescu-Zanetti, A. Blatz, L.P. Wang, L.P. Lee, Single-cell electroporation arrays with real-time monitoring and feedback control, Lab Chip 7 (2007)
457–462.
[15] A. Valero, J.N. Post, J.W. van Nieuwkasteele, P.M. Ter Braak, W. Kruijer, A. van den
Berg, Gene transfer and protein dynamics in stem cells using single cell electroporation in a microfluidic device, Lab Chip 8 (2008) 62–67.
[16] P.J. Marc, C.E. Sims, M. Bachman, G.P. Li, N.L. Allbritton, Fast-lysis cell traps for
chemical cytometry, Lab Chip 8 (2008) 710–716.
[17] A.M. Skelley, O. Kirak, H. Suh, R. Jaenisch, J. Voldman, Microfluidic control of cell
pairing and fusion, Nat. Methods 6 (2009) 147–152.
[18] J.K. Valley, S. Neale, H.Y. Hsu, A.T. Ohta, A. Jamshidi, M.C. Wu, Parallel single-cell
light-induced electroporation and dielectrophoretic manipulation, Lab Chip 9
(2009) 1714–1720.
[19] S. Wang, X. Zhang, W. Wang, L.J. Lee, Semicontinuous flow electroporation chip
for high-throughput transfection on mammalian cells, Anal. Chem. 81 (2009)
4414–4421.
[20] R. Ziv, Y. Steinhardt, G. Pelled, D. Gazit, B. Rubinsky, Micro-electroporation of
mesenchymal stem cells with alternating electrical current pulses, Biomed.
Microdevices 11 (2009) 95–101.
[21] A. Agarwal, M. Wang, J. Olofsson, O. Orwar, S.G. Weber, Control of the release of
freely diffusing molecules in single-cell electroporation, Anal. Chem. 81 (2009)
8001–8008.
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
[22] P.E. Boukany, A. Morss, W.C. Liao, B. Henslee, H.C. Jung, X.L. Zhang, B. Yu, X.M.
Wang, Y. Wu, L. Li, K.L. Gao, X. Hu, X. Zhao, O. Hemminger, W. Lu, G.P. Lafyatis,
L.J. Lee, Nanochannel electroporation delivers precise amounts of biomolecules
into living cells, Nat. Nanotechnol. 6 (2011) 747–754.
[23] D.C. Chang, Cell poration and cell fusion using an oscillating electric field, Biophys.
J. 56 (1989) 641–652.
[24] D.C. Chang, P.Q. Gao, B.L. Maxwell, High efficiency gene transfection by electroporation using a radio-frequency electric field, Biochim. Biophys. Acta 1092 (1991)
153–160.
[25] E. Tekle, R.D. Astumian, P.B. Chock, Electroporation by using bipolar oscillating
electric field: an improved method for DNA transfection of NIH 3T3 cells, Proc.
Natl. Acad. Sci. U. S. A. 88 (1991) 4230–4234.
[26] C. Chen, J.A. Evans, M.P. Robinson, S.W. Smye, P. O'Toole, Measurement of the efficiency of cell membrane electroporation using pulsed ac fields, Phys. Med. Biol.
53 (2008) 4747–4757.
[27] C. Grosse, H.P. Schwan, Cellular membrane potentials induced by alternating
fields, Biophys. J. 63 (1992) 1632–1642.
[28] P. Marszalek, D.S. Liu, T.Y. Tsong, Schwan equation and transmembrane potential
induced by alternating electric field, Biophys. J. 58 (1990) 1053–1058.
[29] V. Vajrala, J.R. Claycomb, H. Sanabria, J.H. Miller Jr., Effects of oscillatory electric
fields on internal membranes: an analytical model, Biophys. J. 94 (2008)
2043–2052.
[30] H.Y. Wang, C. Lu, Electroporation of mammalian cells in a microfluidic channel
with geometric variation, Anal. Chem. 78 (2006) 5158–5164.
[31] T. Geng, Y. Zhan, H.Y. Wang, S.R. Witting, K.G. Cornetta, C. Lu, Flow-through electroporation based on constant voltage for large-volume transfection of cells,
J. Control. Release 144 (2010) 91–100.
[32] T. Geng, Y. Zhan, J. Wang, C. Lu, Transfection of cells using flow-through electroporation based on constant voltage, Nat. Protoc. 6 (2011) 1192–1208.
[33] M. Golzio, M.P. Mora, C. Raynaud, C. Delteil, J. Teissie, M.P. Rols, Control by osmotic pressure of voltage-induced permeabilization and gene transfer in mammalian
cells, Biophys. J. 74 (1998) 3015–3022.
[34] H.Y. Wang, C. Lu, High-throughput and real-time study of single cell electroporation using microfluidics: effects of medium osmolarity, Biotechnol. Bioeng. 95
(2006) 1116–1125.
[35] N. Bao, T.T. Le, J.X. Cheng, C. Lu, Microfluidic electroporation of tumor and blood
cells: observation of nucleus expansion and implications on selective analysis
and purging of circulating tumor cells, Integr. Biol. 2 (2010) 113–120.
7
[36] M. Golzio, M.P. Rols, J. Teissie, In vitro and in vivo electric field-mediated permeabilization, gene transfer, and expression, Methods 33 (2004) 126–135.
[37] S. Satkauskas, M.F. Bureau, M. Puc, A. Mahfoudi, D. Scherman, D. Miklavcic, L.M.
Mir, Mechanisms of in vivo DNA electrotransfer: respective contributions of
cell electropermeabilization and DNA electrophoresis, Mol. Ther. 5 (2002)
133–140.
[38] S. Satkauskas, F. Andre, M.F. Bureau, D. Scherman, D. Miklavcic, L.M. Mir, Electrophoretic component of electric pulses determines the efficacy of in vivo DNA
electrotransfer, Hum. Gene Ther. 16 (2005) 1194–1201.
[39] H. Wolf, M.P. Rols, E. Boldt, E. Neumann, J. Teissie, Control by pulse parameters of
electric field-mediated gene transfer in mammalian cells, Biophys. J. 66 (1994)
524–531.
[40] H.Y. Wang, C. Lu, Microfluidic electroporation for delivery of small molecules and
genes into cells using a common DC power supply, Biotechnol. Bioeng. 100
(2008) 579–586.
[41] J. Wang, Y. Zhan, V.M. Ugaz, C. Lu, Vortex-assisted DNA delivery, Lab Chip 10
(2010) 2057–2061.
[42] M.P. Rols, J. Teissie, Electropermeabilization of mammalian cells. Quantitative
analysis of the phenomenon, Biophys. J. 58 (1990) 1089–1098.
[43] G. Pucihar, T. Kotnik, M. Kanduser, D. Miklavcic, The influence of medium conductivity on electropermeabilization and survival of cells in vitro, Bioelectrochemistry 54 (2001) 107–115.
[44] T. Kotnik, D. Miklavcic, Second-order model of membrane electric field induced
by alternating external electric fields, IEEE Trans. Bio-Med Eng. 47 (2000)
1074–1081.
[45] C. Faurie, M. Rebersek, M. Golzio, M. Kanduser, J.M. Escoffre, M. Pavlin, J. Teissie,
D. Miklavcic, M.P. Rols, Electro-mediated gene transfer and expression are controlled by the life-time of DNA/membrane complex formation, J. Gene Med. 12
(2010) 117–125.
[46] E. Stellwagen, N.C. Stellwagen, Determining the electrophoretic mobility and
translational diffusion coefficients of DNA molecules in free solution, Electrophoresis 23 (2002) 2794–2803.
[47] H.R. Petty, B.R. Ware, S.I. Wasserman, Alterations of the electrophoretic mobility
distribution of rat mast cells after immunologic activation, Biophys. J. 30 (1980)
41–50.
[48] M. Golzio, J. Teissie, M.P. Rols, Direct visualization at the single-cell level of electrically
mediated gene delivery, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 1292–1297.
Please cite this article as: Y. Zhan, et al., Low-frequency ac electroporation shows strong frequency dependence and yields comparable transfection results to dc electroporatio..., J. Control. Release (2012), doi:10.1016/j.jconrel.2012.04.006
GENE DELIVERY
Y. Zhan et al. / Journal of Controlled Release xxx (2012) xxx–xxx