AUTOMATED CELL CHARACTERIZATION PLATFORM:
APPLICATION TO YEAST PROTOPLAST STUDY BY
ELECTROROTATION
J. Laforêt
1
, M. Frénéa-Robin
2
, H. Cérémonie
1
, F. Buret
1
and L. Nicolas
1
1
AMPERE, UMR CNRS 5005, Ecole Centrale de Lyon, Ecully, France
2
Université de Lyon, Lyon, F-69622, France ; AMPERE Villeurbanne, France
Keywords: Dielectrophoresis, electrorotation, yeast cells, yeast protoplasts.
Abstract: This paper is about the development of a new automated platform dedicated to cell manipulation and
characterization by dielectrophoretic methods. We illustrate its possibilities by studying yeast protoplasts
and yeast cells electrorotation spectra, obtained using polynomial microelectrode structures powered by
computer-controlled generators. Measurements were made over the frequency range 100 kHz to 80MHz,
mostly in a suspending medium of conductivity 50 mS/m inside the rotation chamber. The rotation rate of
yeast protoplasts was inferior to that of whole yeast cells. To understand such behavioral differences, yeast
protoplasts were modelled as single-shell spheres in a first approach.
1 INTRODUCTION
The term dielectrophoresis (DEP) is used to describe
the motion and orientation induced by a non-
uniform electric field on polarizable particles, such
as cells. In conventional-DEP (c-DEP), stationary
fields of inhomogeneous strength are used to
translate cells toward field minima or maxima.
Electrorotation (ROT) relies on non-uniformities in
the phase distribution of the applied field to induce
cell rotation at constant velocities.
The effects of chemicals and environmental
factors on the cell electric properties are more and
more adressed. ROT technique enables the study of
various organisms individually without
physiological damage and cell characterization by
angular velocity measurement. This type of
microelectrodes system allows cell handling
(separation, selection, electrofusion…) and transport
and may be basic components to be integrated into
lab-on-a-chips.
Yeasts are eukaryotic cells widely used as model
organism in cell biology, mainly because they are
quick and easy to grow. Preparation and
regeneration of yeast protoplasts are important in
fusion, transformation and cloning studies (Kofod
and al., 1998). Protoplast fusion can be used to
improve anti-bacterial and anti-fungi characteristics
of bakery yeast. In this study, spectra are analysed
according to a two-shell spherical model for whole
yeast cells and single-shell for protoplasts in a first
approach.
2 THEORY
2.1 Yeast Cell Model
In this paper, whole yeast cells are modelled by a
two-shell spherical model. Cytoplasm, membrane
and cell wall are considered as concentric spheres,
according to the individual yeast cell model
developed by Falokun (Falokun and al., 2006). The
complex permittivity of cell interior and membrane
are denoted
*
0
ε
, and
*
1
ε
.To replace the “smeared
out” sphere, we used:
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+
−
−
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+
−
+
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
=
*
1
*
0
*
1
*
0
3
0
1
*
1
*
0
*
1
*
0
3
0
1
*
1
*
1
2
2
2
εε
εε
εε
εε
εε
R
R
R
R
eff
(1)
where R
i
is the radius of the shell index i.
190
Laforêt J., Frénéa-Robin M., Cérémonie H., Buret F. and Nicolas L. (2008).
AUTOMATED CELL CHARACTERIZATION PLATFORM: APPLICATION TO YEAST PROTOPLAST STUDY BY ELECTROROTATION.
In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 190-193
DOI: 10.5220/0001053401900193
Copyright
c
 SciTePress

Then, the complex permittivity of the equivalent
homogeneous cell can be expressed as:
3
**
21eff2
**
11eff2
**
p2
3
**
21eff2
**
11eff2
R
2
R2
R
R2
⎛⎞
⎛⎞
ε−ε
⎜⎟
⎜⎟
+
⎜⎟
⎜⎟
ε+ε
⎝⎠
⎝⎠
ε=ε
⎛⎞
⎛⎞
ε−ε
⎜⎟
⎜⎟
−
⎜⎟
⎜⎟
ε+ε
⎝⎠
⎝⎠
(2)
Relative permittivity and conductivity of the
suspension medium used in the experiments are 78
and 50 mS.m
-1
. Yeast cells average electrical and
geometrical parameters are stored below (Table 1).
Table 1: Properties of different cellular compartments of
yeast cells (Falokun and al, 2006 & Zhou and al, 1996).
Radius or
thickness
Conductivity Permittivity
0:
cytoplasm
5 µm 2 S.m
-1
50
1:
membrane
8 nm 9.10
-6
S.m
-1
6
2: wall 150 nm 6.10
-2
S.m
-1
60
2.2 Electrorotation Theory
The structure and properties of biological cells can
be investigated by observing their ROT spectra
(Gascoyne and al., 2004). Indeed, the rotation
velocity of a spherical particle submitted to a
constant rotating electric field can be expressed as:
[]
2
m
Im K( ) E
R( )
2
εω
ω=−
η
(3)
where η is the solution viscosity, E and ω are the
magnitude and angular frequency of the applied
field.
(
)
K ω is the Clausius-Mossoti factor (CMF),
depending on the particle and its immersion medium
complex permittivities
*
p
ε and
*
m
ε :
**
pm
**
pm
K( )
2
ε−ε
ω=
ε+ε
(4)
The value of
*
p
ε varies according to cell type.
Therefore, each cell type is characterized by a
particular ROT spectrum (as Figures 3 and 4). When
[
]
Im K( ) 0ω> (angle of the induced dipolar moment
with respect to the electric field vector comprised
between 0 and 180°), cells exhibit anti-field rotation
(
R( ) 0
ω
< ). On the contrary case, cells share the
same rotation sense as the field (
R( ) 0ω> ).
3 MATERIALS AND METHODS
3.1 Experimental Setup
The microelectrode structure used in the DEP and
ROT experiments is composed of 4 polynomial
electrodes (Au-Ti deposited on glass) disposed in a
circular arrangement. Those electrodes are powered
by 4 generators delivering sine-wave voltages up to
80 MHz. We simply switch from c-DEP to ROT
according to the phase configuration of the 4 signals
(Figure 1). An advantage is that cells may be
concentrated at the centre of the s by negative DEP
before a ROT experiment. Indeed, all the cells
situated in this area will experiment the same
constant rotating field when undergoing ROT. Only
these cells must be taken into account in rotation
measurements. In this constant field area (Hughes,
1998), cell translation is reduced during the course
of a measurement.
Figure 1: Experimental setup.
Visualization of the applied voltage and
impedance matching are achieved thanks to a wide
band oscilloscope, whose input impedance can be
set to 50 Ohms. All these equipments are controlled
by PC through GPIB interface using a software
developed under LabView
®
which enables
synchronized signal generation. Voltages are kept
constants over the whole frequency range thanks to
automatic gain control.
Cell motion is observed under an inverted
microscope. Image sequences are captured by a high
speed camera. Velocity depends on the electric field
frequency and on the dielectric properties of the cell
and its surrounding medium. Frequency-dependent
rotation rates were first measured with a stopwatch
AUTOMATED CELL CHARACTERIZATION PLATFORM: APPLICATION TO YEAST PROTOPLAST STUDY BY
ELECTROROTATION
191

and then confirmed with a software under
development. After an image processing in order to
detect and label each isolated cell, the angular
velocity is calculated with Matlab
®
by determining
the orientation (angle between its main axis and
horizontal axis) in each sequence of images.
3.2 Cell Preparation
Before experiments, the system was rinsed with
distilled water, washed with ethanol and dried with
an air jet. The samples were centrifuged and the
cells were washed 2 times with a solution whose
conductivity was adjusted to 50 mS.m
-1
by addition
of KCl and directly measured with a conductivity
meter. For protoplasts only, glucose was added to
the solution (at a concentration of 30mM) to adjust
the osmolarity.
Prior to experiments, a drop of cell suspension
(60µL) was deposited onto the electrode system
(gap: 400 µm), in a chamber fabricated with a self-
adhesive silicone bond. Then, a lid was used to close
it and prevent fluid circulation caused by
evaporation.
3.3 Protoplast Forming
Yeast cells (Saccharomyces Cerevisiae) were
suspended and incubated at 35°C during ten minutes
in a pre-treatment solution. After centrifugation,
they were resuspended two times in a buffer solution
which contained 4.7 g.l
-1
sodium citrate, 10.8 g.l
-1
potassium dihydrogenophosphate and 21.8 g.l
-1
sorbitol.
In a second time, they were centrifuged and
resuspended in a 500 U.ml
-1
solution of lyticase
enzymes with the buffer solution. Enzymes digested
the yeast cell wall during an one hour incubation at
room temperature to generate protoplasts.
4 RESULTS AND DISCUSSION
4.1 Simulated ROT Spectra
To obtain the general appearance of the whole yeast
cell spectrum, we use the two-shell model with the
Table 1. The rotation rate is proportional to the
imaginary part of the CMF (3), plotted on Figure 2:
[
]
R( ) Im K( )ω=−χ ω where
2
m
E
2
ε
χ=
η
(5)
In the case of protoplasts, we used the single-
shell model, only taking into account the cytoplasm
and membrane properties (Table 1), in a first
approach.
4.2 Experimental ROT Spectra
First, we have collected typical ROT spectra
exhibited by single viable yeast cells by measuring
the induced rotational velocities at a medium
conductivity of 50 mS.m
-1
under a constant voltage
of 3 V
pp
(Figure 3). The average ROT spectrum is in
agreement with previously reported data (Hölzel,
1997).
10
4
10
5
10
6
10
7
10
8
10
9
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
Frequency (Hz)
Imaginary part of Clausius-Mossotti factor
normal yeast cells
yeast protoplasts
Figure 2:
[
]
Im K( )
−
ω
at 50 mS.m
-1
.
There are two ROT peaks of nearly the same
amplitude. The rotation rate was calculated from
about ten cells per point. Positive and negative
values respectively indicate co-field and anti-field
cell rotation. The results plotted Figures 3 and 4
were averaged across 4 experiments, vertical lines
indicated amplitude between the minimal and
maximal values.
10
4
10
5
10
6
10
7
10
8
-1.5
-1
-0.5
0
0.5
1
1.5
f
(
Hz
)
ROT rate (rad.s
-
1.V
-
2)
norma
l
yeas
t
ce
ll
s
Figure 3: Experimental ROT spectra at 50 mS.m
-1
.
Then, we have collected ROT spectra exhibited
by yeast protoplasts for a conductivity of 50 mS.m
-1
under a 6 V
pp
constant voltage (Figure 4). The zero
crossing frequency was situated around 15 MHz and
11 MHz for normal yeast cells and protoplasts
respectively.
BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices
192

The cells exhibited anti-field rotation at
frequencies below the zero crossing and co-field
rotation above.
10
5
10
6
10
7
10
8
-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
f
(
Hz
)
ROT rate (rad.s
-
1.V
-
2)
yeast cells protoplasts
Figure 4: Experimental ROT spectra at 50 mS.m
-1
.
4.3 Discussion
For whole yeast cells, the overall spectrum (Figure
3) is consistent with that obtained from simulation
(Figure 2), up to a multiplicative scalar factor, which
could be explained by the fact that the simulated
data did not take into account the factor χ (5).
On both experimental spectra, the negative ROT
peak happens near 500 kHz despite rotation rate
attenuation for yeast protoplasts. The viscosity of
the medium suspension, in the case of protoplasts,
was almost 5% higher than water viscosity because
of glucose (Easteal, 1989), which affects the value
of χ. Nevertheless, more precise investigation is
necessary to understand this rotation slowdown.
As can be seen from Table 1, the dielectric
properties of cell wall are close to medium
properties. This may explains the resemblance
between yeast cells and yeast protoplast simulated
spectra (Figure 2) obtained with our approach,
consisting in switching from a model to another by
suppressing the shell representing the cell wall.
The approach consisting in modelling protoplasts
by the two most inside shells presents its limitations.
Indeed, simulated result does not fit the
experimental data well in the co-field rotation part
of the spectra. For 1.1 mS.m
-1
, simulation points out
more differences between the two spectra (Figure 5).
10
3
10
4
10
5
10
6
10
7
10
8
10
9
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
0
.
6
Frequency (Hz)
Imaginary part of Claus ius-Mossotti factor
normal yeast cells
yeas t protoplasts
5 CONCLUSION
As further developments, we need to improve our
experimental setup to cover a wider frequency range
and obtain thereby a complete ROT spectrum,
including the second peak. Future experiments
performed in lower conductivity immersion media
(1.1 mS.m
-1
, for example) may bring more
information about the cell wall influence (Figure 5).
Yeast protoplast and whole cell electric
properties can be extracted from experimental ROT
spectra by parameter identification thanks to a
identification process under Matlab
®
. During this
step, more sophisticated models could be used to
describe cells, as for instance a N-shell ellipsoidal
model. The measurement of cell properties is a step
toward the modelling of electromagnetic field-tissue
interaction using a bottom-up approach.
LabView
®
interface allows to realize several
series of different experiments. Cell motion was
successfully observed over a wide frequency range
for yeast cells. Fabrication of microelectrodes
enabling travelling-wave dielectrophoresis is the last
part of our platform to be developed. Further, these
cell manipulation techniques permit to study the
effects of various treatments on cells such as
response to toxicants for magnetic field exposure
and to detect cell pathologies.
REFERENCES
Kofod, L.V., Andersen, L.N., Kauppinen, M.S., Christgau,
N.S., Dalboge, H., 1998. Enzyme with P-1,6
endoglucanase activity, United States Patent. Patent
Number 5770406.
Falokun, C.D., Markx, G.H., 2006. Electrorotation of
beads of immobilized cells, J. of Electrostatics.
Zhou, X-F., Markx, G.H., Pethig, R., 1996. Effect of
biocide concentration on electrorotation spectra of
yeast cells, Bioch. and Biophys. Acta.
Gascoyne, P. R. C., Vykoukal, J. V., 2004.
Dielectrophoresis-based sample handling in general-
purpose programmable diagnostic instruments,
Proceedings of the IEEE.
Hughes, M.P., 1998. Computer-aided analysis of
conditions for optimizing pratical electrorotation,
Physics in Medecine and Biology.
Hölzel, R., 1997. Electrorotation of Single Yeast Cells at
Frequencies Between 100Hz and 1.6GHz, Biophysical
J.
Easteal, A.J., 1989. Can. Tracer diffusion in aqueous
sucrose and urea solutions, J. Chem..
Fi
g
ure 5:
[
]
Im K
()
−ω
at 1.1 mS.
m
-1
.
AUTOMATED CELL CHARACTERIZATION PLATFORM: APPLICATION TO YEAST PROTOPLAST STUDY BY
ELECTROROTATION
193