2311895

Separation of Viable and Nonviable Animal Cell Using Dielectrophoretic Filter
Masaru Hakoda
Dept. of Chemical and Environmental Engineering, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma, Japan
Yoshikazu Wakizaka
Center for Advanced Science and Innovation, Osaka University, Osaka, Japan
Yusuke Hirota
Dept. Environment Management, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma, Japan
DOI 10.1002/btpr.394
Published online February 8, 2010 in Wiley Online Library (wileyonlinelibrary.com).
Selective separation of cells using dielectrophoresis (DEP) has recently been studied and
methods have been proposed. However, these methods are not applicable to large-scale separation because they cannot be performed efficiently. In DEP separation, the DEP force is
effective only when it is applied close to the electrodes. Utilizing a DEP filter is a solution
for large-scale separation. In this article, the separation efficiency for viable and nonviable
cells in a DEP filter was examined. The effects of an applied AC electric field frequency and
the gradient of the squared electric field intensity on a DEP velocity for the viable and nonviable animal cells (3-2H3 cell) were discussed. The frequency response of the DEP velocity
differed between the viable and the nonviable cells. We deducted an empirical equation that
can be used as guiding principle for the DEP separation. The results indicate that the viable
and the nonviable cells were separated using the DEP filter, and the best operating condiC 2010 American Institions such as the applied voltage and the flow rate were discussed. V
tute of Chemical Engineers Biotechnol. Prog., 26: 1061–1067, 2010
Keywords: bioseparation, dielectrophoresis, animal cell, AC electric field
Introduction
The dielectrophoresis (DEP) phenomenon was studied in
detail by Pohl1 in 1960. DEP is the motion of dielectric particles caused by polarization effects in a nonuniform electric
field. This motion depends on the dielectric properties of the
particle and medium, particle size, as well as the gradient of
the squared electric field intensity (!E2), which depends on
the applied voltage and electrode geometry.1,2
Recently, the dielectric properties of bioparticles have
been analyzed using the DEP phenomenon to identify methods of selective manipulation and separation.3 DEP has
many useful biotechnological applications, for example, in
the separation of viable and nonviable yeasts from their mixture, 4,5 in the separation of blood cells,6–8 and in the manipulation and the separation of submicron particles such as
latex particles and viruses.9–11 For more details, see Refs.
12–15. However, the DEP force is effective only when it is
close to the electrode because !E2 decreases rapidly away
from the electrode. Therefore, these have not been applied to
large-scale separations because they are the separation in the
narrow domain on a micro device. However, the DEP filter
has the large volume of separation products, and the possibility of continuous operation compared with the micro device. Few studies on separation devices using the DEP filter
have been reported.16–21 Docoslis et al.16,17 used an etched
Correspondence concerning this article should be addressed to M.
Hakoda at [email protected]
C 2010 American Institute of Chemical Engineers
V
silicon wafer between the electrode gaps to develop a filter,
thereby resulting in continuous separation. Authors discussed
the electrode geometries to improve the separation efficiency
of the DEP filter.20,21 On the basis of the aforementioned
paper, the electrodes of the DEP filter of present research
were made, and the separation experiments were performed.
This filter has a selective retention capability for viable cells,
which is caused by a negative-DEP force that moves toward
the low-electric field region. It retains only the viable cells
in mixture cells and discharges the nonviable cells.
The purpose of this research was to examine the practicality of a cell separator that used the DEP filter. The effects of
the experimental conditions on the separation efficiency were
examined using a wire–wire type DEP filter. In addition, the
electrical difference between viable and nonviable cells was
clarified by measuring the dielectric characteristic of animal
cells. On the basis of those results, separation of the viable
and nonviable cells was performed using the DEP filter.
Moreover, the DEP separation by the difference in the activity of the cells was examined.
Theory
A time average dielectrophoresis force FDEP is applied in
particles in the nonuniform AC electric field and is theoretically shown by the following equation4:
FDEP ¼ 2pr3 eM Re½KðxÞ rE2
(1)
1061
1062
Biotechnol. Prog., 2010, Vol. 26, No. 4
Figure 2. Conceptual diagram of viable and nonviable cell separation by dielectrophoretic filter.
Materials and Methods
Materials
Figure 1. Schematic diagrams of experimental apparatus for
dielectrophoretic velocity measurement and the electrode geometry.
where r is the particle radius, eM is the real part of the medium permittivity, and E is the electric field intensity.
Re[K(x)] indicates the real part of the Clausius-Mossotti
function and is given by the following equations:
KðxÞ ¼
ep eM
ep þ 2eM
(2)
where e*
p and e*
M are the complex permittivity of the particle and the medium, respectively. Re[K(x)] depends on the
frequency of the applied electric field and also on the
dielectric property of the particles and medium. When
Re[K(x)] [ 0, the particles move toward the high electric
field region, called the positive DEP. On the other hand,
when Re[K(x)] \ 0, the particles move toward the low
electric field region, called the negative DEP. !E2 depends
on the electric field intensity, the electrode shape, and the
shape of the device.
Stokes law is shown by the following equation:
Fdrag ¼ 6pg rv
(3)
where Fdrag is the drag force, g is the viscosity of the medium, and v is the particle velocity. For small particles and
ignoring the Brownian motion and buoyancy force, it can be
assumed that the DEP force and the drag force are equal.
Therefore, the DEP velocity can be given by the following
equation:
vDEP ¼
r eM Re½KðxÞ
rE2
3g
2
(4)
As seen in Eqs. 1 and 4, the most important factors for
determining the DEP force and the DEP velocity are
Re[K(x)] and !E2. The !E2 is another factor that greatly
depends on the applied voltage and the electrode shape.
The DEP force is influenced by the electrode geometry and
the geometry of the separation device. However, !E2
decreases abruptly with increase of the distance from the
electrode.
Mouse hybridoma 3-2H3 cells (RCB0867, Riken Gene
Bank, Japan) of free-floating cells, HeLa cells (RCB0007,
Riken Gene Bank, Japan), and MDCK cells (03-360, Dainippon Sumitomo Pharmaceutical Co., Ltd., Japan) of anchorage-dependent cells were used. The growth medium used
was DMEM (D6429, Sigma Co., Ltd.) supplemented with
10% FCS, 100 mg/L streptomycin sulfate, and 100 U/mL
crystalline potassium penicillin G. The 3-2H3 cells were cultured in a petri dish at 310 K in a humidified 5% CO2 incubator. The cells were suspended in isotonic solution
consisting of 8.5 %(w/v) sucrose plus 0.3 %(w/v) dextrose
buffer, and the cell concentrations were adjusted to about
106 cells/mL. The nonviable cells were obtained by an autoclave (SS-325, TOMY) at 80 C for 10 min. The cell concentration was measured using a hemocytometer.
Measuring method for DEP velocity
Schematic diagrams of the experimental apparatus for the
DEP velocity measurement and the geometry of the electrodes are shown in Figure 1. The electrodes were made of
nickel, and both the electrode wires were 50 lm in diameter.
The distance between the wire-wire electrodes installed horizontally is 350 lm. To measure the DEP velocity of the
cells, the position of individual cells was recorded as a function of time in an arbitrary area divided into 50 lm units.
An ac voltage was applied to the electrodes using a function
generator (Model 33250A, Agilent Technologies, USA). The
movement of a single cell subjected to the DEP force was
observed by a CCD camera system (CCD Micro Scope Inf500, Moritex Co., Japan and Trinitron, SONY Co., Japan).
All experiments were carried out in a temperature controlled
room at 298 1 K.
DEP filter apparatus
Conceptual diagram of the separation of viable and nonviable cell separation by the DEP filter (cross sectional view)
is shown in Figure 2. This filter has a selective retention
capability for viable cells that is caused by a negative-DEP
force which moves toward the low electric field region. On
the other hand, since nonviable cell has small DEP force, it
passes through between the DEP filters by the flow of liquid.
The cells are separated by this method close to the electrode
which the DEP force is effectively acted.
A schematic diagram of the cell separation apparatus with
the DEP filter and a photograph of the filter electrode are
Biotechnol. Prog., 2010, Vol. 26, No. 4
1063
Figure 3. Schematic diagrams of dielectrophoretic separation apparatus and photograph of filter electrode.
(a) Schematic diagram of the experimental apparatus for cell separation. (b) Side viewof DEP filter unit. (c) Photograph of wire electrode for
DEP filter.
shown in Figure 3. The upper part of the DEP filter unit consisted of a circulation liquid gateway, a permeation liquid
exit, and the filter electrode part (Figure 3b). The electrodes
were made of nickel, and both the electrode diameter and
electrode gap were 200 lm (Figure 3c). Moreover, the diameter of the permeation part of the DEP filter was 10 mm. The
DEP filter unit middle part and the lower part consisted of a
silicon rubber spacer of 2 mm in thickness and an acrylic resin
plate, respectively. A cell suspension in a reservoir was sent
to the filter unit with a peristaltic pump, released at from Exit
B and returned to the reservoir again (Figure 3a). The permeation flowing rate to the DEP filter was kept constant with
another pump connected to Exit A. The retention ratio of cells
was defined by the following equation:
Retention ratio ½% ¼
XB
100
XA þ XB
(5)
where, XA and XB are the cell concentrations at Exit A and
Exit B, respectively.
Results and Discussions
Effect of frequency on DEP velocity
To examine the optimum frequency for the separation of
viable and nonviable 3-2H3 cells, the DEP velocity was
measured by using the apparatus shown in Figure 1. For the
viable cells, cells that were in the logarithm growth period
1064
Biotechnol. Prog., 2010, Vol. 26, No. 4
Figure 4. Effect of applied electric field frequency on dielectrophoretic velocity.
Figure 5. Effect of !E2 on dielectrophoretic velocity for viable
cells.
of 3 days after seeding were used, and the nonviable ones
were obtained by heating the viable cells at 356 K for 10
min. The effect of frequency on the DEP velocity of the 32H3 cells is shown in Figure 4. The nonviable cells displayed a weak positive-DEP at the frequency range of 1
kHz–1 MHz. On the other hand, the viable cells have shown
a negative-DEP at 10 kHz or less and a positive-DEP over
100 kHz. This result indicates that viable and nonviable cells
are separable under the experimental conditions for a frequency of 10 kHz or less using the DEP filter.
Effect of !E2 on DEP velocity
To examine the influence of !E2 on the DEP velocity for
the viable cells, the DEP velocities of the cells that move
between the wire electrodes were measured. The relation
between the DEP velocity and !E2 in a certain position
between the electrodes was measured. The factor !E2,
which affects the DEP force, was analyzed using 2D finite
element analysis software (Maxwell 2D, Ansoft Corp.,
UAS). Figure 5 shows the influence of the analytic value of
!E2 on the DEP velocity measured under the conditions of
the frequency of 1 kHz and applied voltage of 14 Vpp. Based
on these experimental results, the empirical equation was
obtained:
Figure 6. Effect of frequency on retention ratio for viable
cells.
(Cultivated 3 days, viability: more than 90%).
They found that the DEP velocity is equal to Eq.4. Therefore, the constant (3 1017) must be equal to (r2eMRe
[K(x)])/3g in Eq.4.
2 mL/min, and permeation flow rate of 0.4 mL/min. The experimental results are shown in Figure 6. The retention ratio
decreased with the increase in frequency and in the case of 1
kHz was the highest. This result agrees with the measurement results of the DEP velocity shown in Figure 4. The
aforementioned result shows that the optimum frequency for
the retention of viable cells was 1 kHz.
Effect of frequency on the retention of viable cells
Effect of applied voltage on viable cell retention
The retention ratios of viable 3-2H3 cells were measured
using the DEP separation apparatus shown in Figure 3. Cells
at 3 days after seeding are defined as viable ones. The cell
separation experiments were carried out under the conditions: frequency of 1, 5, and 10 kHz, circulation flow rate of
The effect of the applied voltage on the retention ratio of
the 3-2H3 viable cells was examined under a frequency of 1
kHz. The experiment was carried out at 2 mL/min circulation flow rate, 0.4 or 0.8 mL/min permeation flow rate. The
experimental results are shown in Figure 7. The retention
vDEP ¼ 3 1017 rE2
(6)
Biotechnol. Prog., 2010, Vol. 26, No. 4
1065
Figure 8. Effect of frequency on retention ratio.
Figure 7. Effect of applied voltage on retention ratio for viable
cells.
(Mixture cell suspension of cultivated 3 days and 5 days cells,
viability: 55%)
(Cultivated 3 days, viability: more than 90%).
ratios were over 50% regardless of the applied voltage. As
defined in Eq. 5, the retention ratio becomes 50% when the
cells concentration at Exit A and that at Exit B are equal. At
the permeation flow rate of 0.4 mL/min, the retention ratio
increased over the applied voltage of 20 Vpp and the retention ratio of 95% was obtained at 40 Vpp. At the permeation
flow rate of 0.8 mL/min, the retention ratio increased over
30 Vpp, and the retention ratio was 95% at 50 Vpp. In the
cell separation using the DEP filter, it is thought that the
minimum voltage holding the cells can be estimated by Eq.
6. As a result, the critical voltage necessary for retaining the
viable cells was 20 Vpp under the condition of a frequency
of 1 kHz and a permeate flow rate of 0.4 mL/min, and it
was 27 Vpp under the condition of a permeate flow rate of
0.8 mL/min. The experimental results of the applied voltage
on which the retention rate of the viable cells increased
agreed well with the critical voltage obtained by Eq. 6. In
the case of the cell separation using the DEP filter, Eq. 6
was effective for deciding the permeation flow rates and the
applied voltage, etc.
Figure 9. Effect of frequency on retention ratio.
Separation of viable and nonviable cells
using the DEP filter
The viable cells and the nonviable cells were separated
using the DEP filter. Authors reported that there are relationship between the dielectric characteristic and the growth activity in the cultivation process of the 3-2H3 cells.22 Cells
cultured for 3 days after seeding were chosen as the viable
ones and cells cultured for 5 days after seeding were chosen
as the nonviable cells. In the separation experiment of the
viable and nonviable cells, both of the cells were adjusted to
the same concentration.
The circulation flow rate was 2 mL/min, and the permeation flow rate was 0.4 mL/min. The experimental results are
shown in Figure 8. Since the retention ratios of the nonviable cells were 50–60%, it is concluded that the nonviable
cells permeated the DEP filter without the action of the DEP
(Cultivated 5 days, viability: 20%).
force. On the other hand, the retention ratio of the viable
cells was around 80%, so the viable and the nonviable cells
were separated by using the DEP filter. The mixture suspension of cells cultured for 3 days and for 5 days was separated using the DEP filter. However, the retention ratios for
the viable cells in the experiment using the mixture suspension shown in Figure 8 were lower than those in the experiments using only the cells cultured for 3 days in Figure 6.
To examine this phenomenon, the cells cultured for 5 days
were separated under the same conditions, and the experimental results are shown in Figure 9. The retention ratios of
the nonviable cells were about 50–60%, which agrees well
with the results of the mixture suspension shown in Figure 8,
1066
Biotechnol. Prog., 2010, Vol. 26, No. 4
Figure 10. Dielectrophoretic behaviors of various species of viable and nonviable cells.
and the nonviable cells were not retained with the DEP filter.
On the other hand, the retention ratios of the viable cells cultured for 5 days were about 60%, and the viable cells were
not almost retained either. The retention ratios of the viable
cells of the mixture cells suspension cultured for 3 days and
5 days were lower than that of the cells cultured for 3 days.
This is because the viable cells cultured for 5 days permeated the DEP filter. As a result of the microscopy, no difference in size could be observed in viable cells cultured for 3
days and the viable cells cultured for 5 days. From the
results, it found that there are ‘‘dielectric differences’’
between the viable cells cultured for 3 days and that for 5
days. Moreover, the results shown in Figure 8 show that
there was no influence of cell size on the retention ratios of
the viable and nonviable cells since the retention ratios of
both the viable cells and the nonviable cells were not influenced by the applied voltage. The aforementioned results
suggest that the differences of the DEP force between the
viable and the nonviable cells are not the difference in the
cell size but the differences in the electrical characteristics
of the cells. Moreover, it was proved that the electrical characteristics of the cells are influenced by the cultivation days
of the cells.
The DEP characteristic of various viable
and nonviable cells
To examine the possibility of separating cells of a different species, the direction of the DEP was observed using the
apparatus in Figure 1. The effects of the frequency on the
DEP characteristics of the 3-2H3 cells, the MDCK cells, and
the HeLa cells are shown in Figure 10. Consequently, in the
three species of the cells, the DEP characteristic of the viable cells and the nonviable cells differed. Furthermore, a
specific frequency that gives negative-DEP force only to the
viable cells existed. It has to be proven that the viable cells
and the nonviable cells can be separated using the DEP filter
even in cells other than the 3-2H3 cells. Moreover, it is con-
sidered that separation by cell species is possible, since the
DEP characteristics differed according to the cells species.
Conclusions
This article shows that separation by retaining only the
viable cells by negative DEP and removing the nonviable
cells is possible using the DEP filter. The results of the measurement of the DEP velocity using a wires-type electrode
unit indicate that the optimum frequency for separating the
viable and the nonviable 3-2H3 cells is 1 kHz. Furthermore,
an empirical equation for separation using the proposed DEP
filter was obtained. In the separation experiment using the
DEP filter, the retention ratio of the viable cells was maximum at the frequency of 1 kHz. The empirical equation was
an indicator for determining the permeation flux and the
applied voltage. However, the retention ratio of the cells in
the logarithm growth phase was higher than that in the dead
phase. It was concluded that the reason for this is the change
of the electrical characteristics of cells as a result of a
change in the metabolism, cell cycle, etc. The result of this
study suggests the possibility of separation of high-activity
cells and low-activity cells. It was suggested that the separation of viable and nonviable cells is possible not only for
3-2H3 cells but also for MDCK cells and HeLa cells and
separation by cell species could also be possible.
Literature Cited
1. Pohl HA. Dielectrophoresis, New York: Cambridge University
Press; 1978.
2. Jones TB. Electromechanics of Particles, New York: Cambridge
University Press; 1995.
3. Pethig R, Markx GH. Applications of dielectrophoresis in biotechnology. Trends Biotechnol. 1997;15:426–432.
4. Markx GH, Talary MS, Pethig R. Separation of viable and nonviable yeast using dielectrophoresis. J Biotechnol. 1994;32:29–
37.
Biotechnol. Prog., 2010, Vol. 26, No. 4
5. Markx GH, Pethig R. Dielectrophoretic separation of cells: continuous separation. Biotechnol Bioeng. 1995;45:337–343.
6. Becker FF, Wang X-B, Huang Y, Pethig R, Vykoukl J, Gascoyne PRC. Separation of human breast cancer cells from blood
by differential dielectric affinity. Proc Natl Acad Sci USA.
1995;92:860–864.
7. Gascoyne PRC, Wang X-B, Yang J, Becker FF. Dielectrophoretic separation of cancer cells form blood. IEEE Trans Ind
Appl. 1997;33:670–678.
8. Wang X-B, Yan J, Huang Y, Vykoukal J, Becker FF, Gascoyne
PRC. Cell separation by dielectrophoretic field-flow-fractionation. Anal Chem. 2000;72:832–839.
9. Morgan GH, Green NG. Dielectrophoretic manipulation of rodshaped viral particles. J Electrostatics. 1997;42:279–293.
10. Morgan H, Hughes MP, Green NG. Separation of submicron
bioparticles by dielectrophoresis. Biophys J. 1999;42:279–293.
11. Schnelle T, Muller T, Gradl G, Shirley SG, Fuhr G. Dielectrophoretic manipulation of suspended submicron particles. Electrophoresis. 2000;21:66–73.
12. Pethig R, Markx GH. Applications of dielectrophoresis in biotechnology. Trends Biotechnol. 1997;15:426–432.
13. Gascoyne PRC, Vykoukal J. Particle separation by dielectrophoresis. Electrophoresis. 2002;23:1973–1983.
14. Hughes MP. Strategies for dielectrophoretic separation in laboratory-on-a-chip systems. Electrophoresis. 2002;23:2569–2582.
15. Gascoyne PRC, Huang Y, Pethig R, Vykoukal J, Becker FF.
Dielectrophoretic separation of mammalian cells studied by
1067
16.
17.
18.
19.
20.
21.
22.
computerized image analysis. Meas Sci Technol. 1992;3:439–
445.
Docoslis A, Kalogerakis N, Behie LA, Kaler KIVS. A novel
dielectrophoresis-based device for the selective retention of viable cells in cell culture media. Biotechnol Bioeng. 1997;54:239–
250.
Docoslis A, Kalogerakis N, Behie LA. Dielectrophoretic forces
can be safely used to retain viable cells in perfusion cultures of
animal cells. Cytotechnology. 1999;30:133–142.
Abidin ZZ, Downes L, Markx GH. Large scale dielectrophopretic construction of biofilms using textile technology. Biotechnol Bioeng. 2007;96:1222–1225.
Abidin ZZ, Downes L, Markx GH. Novel electrode structures
for large scale dielectrophoretic separations based on textile
technology. J Biotechnol. 2007;130:183–187.
Wakizaka Y, Hakoda M, Shiragami N. Effect of electrode geometry on dielectrophoretic separation of cells. Biochem Eng J.
2004;20:13–19.
Wakizaka Y, Hakoda M, Shiragami N. Numerical simulation of
electrode geometry and its arrangement of dielectrophoretic filter for separation of cells. J Chem Eng Jpn 2004;37:908–911.
Hakoda M, Hachisu T, Wakizaka Y, Mii S, Kitajima N. Development of a method to analyze single cell activity by using
dielectrophoretic levitation. Biotechnol Prog. 2005;21:1748–
1753.
Manuscript received Apr. 24, 2009, and revision received Dec. 14, 2009.