See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/47510909

Electrical cell-substrate impedance sensing as a non-invasive tool for cancer

cell study

Article in The Analyst · October 2010

DOI: 10.1039/c0an00560f · Source: PubMed

CITATIONS READS

171 775

5 authors, including:

Jongin Hong Mohana Marimuthu

Chung-Ang University Dhanalakshmi Srinivasan University
215 PUBLICATIONS 2,983 CITATIONS 40 PUBLICATIONS 674 CITATIONS

SEE PROFILE SEE PROFILE

Cheol Soo Choi Sanghyo Kim

Gachon university college of medicine, Korea Gachon University
168 PUBLICATIONS 10,223 CITATIONS 149 PUBLICATIONS 3,462 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Mohana Marimuthu on 15 February 2018.

The user has requested enhancement of the downloaded file.

 View Online

MINIREVIEW www.rsc.org/analyst | Analyst

Electrical cell-substrate impedance sensing as a non-invasive tool for cancer

cell study
Jongin Hong,†*a Karthikeyan Kandasamy,b Mohana Marimuthu,b Cheol Soo Choic and Sanghyo Kim*b
Received 22nd July 2010, Accepted 20th September 2010
DOI: 10.1039/c0an00560f

Cell-substrate interactions are investigated in a number of studies for drug targets including
angiogenesis, arteriosclerosis, chronic inflammatory diseases and carcinogenesis. One characteristic of
malignant cancerous cells is their ability to invade tissue. Cell adhesion and cytoskeletal activity have
served as valuable indicators for understanding the cancer cell behaviours, such as proliferation,
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F

migration and invasion. This review focuses on bio-impedance based measurement for monitoring the
behaviours in real time and without using labels. Electric cell-substrate impedance sensing (ECIS)
provides rich information about cell-substrate interactions, cell-cell communication and cell adhesion.
Downloaded by Imperial College London on 26 October 2010

High sensitivity of the ECIS method allows for observing events down to single-cell level and achieving
nanoscale resolution of cell-substrate distances. Recently, its miniaturization and integration with
fluorescent detection techniques have been highlighted as a new tool to deliver a high-content platform
for anticancer drug development.

Introduction year of 2020. Accordingly, it has been of significant importance

Cancer is a disease in which abnormal cells vigorously divide
without control, invade adjacent tissues and spread through the
body via bloodstreams or lymphatic vessels.1,2 Last decades
cancer has been a major public concern and become the leading
cause of natural death. For example, breast cancer in women and
prostate cancer in men are common malignancies in the world.
New cases over ten millions are diagnosed annually worldwide,
and the number is expected to increase to twenty millions by the
a
Department of Chemistry and Institute of Biomedical Engineering,
Imperial College London, South Kensington Campus, London, SW7
2AZ, United Kingdom. E-mail: hong.jongin@gmail.com
b
College of Bionanotechnology, Kyungwon University, San-65, Bokjeong-
dong, Sujeong-gu, Seongnam-si, Gyeonggi-do, 461-701, Korea. E-mail:
samkim@kyungwon.ac.kr
c
Korea Mouse Metabolic Phenotyping Centre, Lee Gil Ya Cancer and
Diabetes Institute, Gachon University of Medicine and Science, 7-45,
Songdo-dong, Yeonsu-gu, Incheon, 406-840, Korea
† Present address: Department of Material Science & Engineering,
KAIST, Daejeon 305-701, Korea
to uncover the roles of various genes, protein products and
growth factors in the abnormal behavior of cancer cells. In
particular, vascular endothelial growth factors (VEGFs) play
a major role on inducing the proliferation, migration and
survival of endothelial cells and leading to the formation of new
blood vessels that are vital for tumour growth and metastases.3
Recently, special attention has been paid to test safety and effi-
ciency of new anticancer drug candidates4,5 and thus in vitro
cytotoxicity testing with live cells is invaluable in the assessment
of their adverse effects and environmental threats. Importantly,
label-free and non-invasive techniques have been highly on
demand since conventional cell-based assays for the analysis of
cell proliferation and viability are labor intensive, involve
labeling steps and provide a snapshot of the experiment. A
number of label-free technologies including electrical cell-
substrate impedance sensing (ECIS), quartz microbalance and
optical lightmode spectroscopy, have emerged as a means of
monitoring cellular processes in real-time.6–9 Among them, ECIS
has been recognized as a powerful tool to investigate cellular

Jongin Hong is currently a research professor in the Department of Material

Science & Engineering at KAIST and also an academic visitor in the
Department of Chemistry at Imperial College London (ICL). Before his
current position, he has worked as a postdoctoral research associate at ICL
for 3 years (2007–2010). Sanghyo Kim is an assistant professor in the
College of Bionanotechnology at Kyungwon University. Karthikeyan Kan-
dasamy and Mohana Marimuthu are studying for a Master degree at
Kyungwon University under the supervision of Jongin Hong and Sanghyo
Kim. Cheol-Soo Choi is currently an associate professor at Gachon Univer-
sity of Medicine and Science. This group is mainly engaged with the devel-
opment of high-throughput cellular assays and drug discovery for cancer.

Sanghyo Kim (Left), Mohana Marimuthu, Jongin Hong and

Karthikeyan Kandasamy (Right)

This journal is ª The Royal Society of Chemistry 2010 Analyst


properties and physiological functions of cells after Giaever and
Keese firstly invented it in 1984.10,11 To date, ECIS measurements
have been successfully established to monitor motion, attach-
ment and spreading of cultured cells, quantify effects of
biochemical compounds including cytotoxicity and assess cell
migration in wound healing assays.11–21
Typically, an ECIS system consists of gold electrodes of
a small sensing electrode and a large counter electrode in the
bottom of cell culture dishes, a lock-in amplifier with an internal
oscillator and a personal computer that controls the measure-
ment and stores the data. The system is based on measuring the
impedance change with weak and non-invasive AC signals that
can theoretically be modeled as a RC circuit. The oscillator that
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
relays to switch between the different wells applies an AC signal
of 1 mA (usually in the frequency range from 1 to 40 kHz)
through a series of a 1 MU resistor between the relatively small
microelectrode (103 cm2) and the larger counter electrode. The
Downloaded by Imperial College London on 26 October 2010
working electrode deposited on a cell culture dish should be 103
cm2 or even smaller since a resistance component from a bulk
electrolyte (normally cell culture medium) dominates the
measured impedance when the larger electrode is used. The
amplifier collects the in-phase and out-of-phase voltages, which
are converted to time series of resistance and capacitance, across
the system. When cells adhere and spread on the electrode
surface, the impedance is altered because the cells restrict the
current flow. Accordingly, this change can be utilized to analyze
dynamic information about cell shape and locomotion. The
measured impedance is also susceptible of external conditions,
such as pH, temperature and an addition of biological and
chemical compounds. Different types of configurations including
eight wells with one electrode (8W1E), eight wells with ten elec-
trodes (8W10E) and 96 wells with integrated electrodes (96WE)
are to monitor the cell behavior in real-time and without the use
of labels. A schematic diagram of an ECIS system and its
working principle are shown in Fig. 1. Basically, the cells are
Fig. 1 (a) A schematic diagram of the ECIS measurement system and (b) current flows without cells (top) and with cells (bottom) from the surface of Au
sensing electrodes.
Analyst
View Online
considered as dielectric particles due to the insulating properties
of their membranes and thus, when they attach and spread on the
small electrode, interfere the current flow between the two elec-
trodes. This leads to large and measurable changes in impedance
and the impedance increases until they are fully spread
(a confluent layer of the cells is established). After reducing the
area available for the current flow, the measured impedance
fluctuates with time. If the change in cell shape or motion or both
occurs, this results in the corresponding change in the current
flow between the ventral surface (i.e. the side closest to the
ground) of the cells and the substratum (i.e. a solid surface over
which a cell moves or upon which a cell grows) and the resistance
between the cells. The degree of the changes is determined by the
number of cells seeded on the sensing electrode, cell-cell inter-
action, cell morphological changes and cell motility. In addition,
the impedance fluctuation provides information about cellular
dynamics depending on different cell types. Accordingly, time-
resolved impedance measurements provide real-time fingerprints
of the cells under a variety of culture conditions. To explain the
resulting impedance, mathematical models have been developed
to calculate cell morphological parameters, such as the narrow
spaces between cells (barrier resistance, Rb), the narrow spaces
between the ventral surface of the cells and the electrode (h), and
the cell membrane capacitance (Cm).22–25 Such parameters can be
determined by fitting the theoretical model to the experimental
data. Recently, Lovelady et al. have demonstrated that the
signatures of electrical noise in ECIS measurements can be useful
to distinguish different cell types and their dynamics.26 There-
fore, the ECIS has been considered as a non-invasive tool to
quantitatively study cell behaviours, such as cell spreading,
attachment, proliferation, differentiation and migration, and
cytoxicity of compounds or drugs. ECIS measurements can also
give information about electrochemical processes in the tissues
and be used to monitor their physiological changes against
environmental conditions. In addition, the utilization of
This journal is ª The Royal Society of Chemistry 2010

impedance as a cellular readout allows for the preclusion of
conventional labels. This can circumvent their negative impact
on cellular processes and place cells in their native and physio-
logically relevant status. Importantly, the ECIS system provides
kinetic data of initial cell attachment and prolonged steady-state
adhesion in real time and the process is fully automated with
conventional attachment assays. It also detects vertical motions
of the cells as they move on or off the working electrode with the
spatial resolution less than 1 nm and thus this technique is
a unique measurement of cell motility. In this mini-review, we
will evaluate ‘Electrical Cell-substrate Impedance Sensing
(ECIS)’ as a label-free and non-invasive tool for cancer biology.
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
Cell adhesion, migration and proliferation
When normal cells are grown up in a culture vessel (test tube,
bottle, flask or dish), they tend to divide until its surface is
Downloaded by Imperial College London on 26 October 2010
covered by a single layer of cells and then further cell division is
inhibited (called density-dependent inhibition). In contrast,
cancer cells tend to proliferate and gradually pile up on top of
another, forming multilayered aggregates. Another difference
between normal and cancer cells involves the requirement of
anchorage. Most normal cells only grow well when they are
provided with an appropriate solid surface to which they can
adhere (called anchorage-dependent). If the anchorage is pre-
vented, normal cells cannot be divided and may commit suicide
by apoptosis (i.e. an orchestrated programme for cell death).
However, cancer cells are not subject to this normal safeguard.
Accordingly, the continual monitoring of cell attachment,
spreading and proliferation are crucial for understanding cellular
behaviors of both normal and cancer cells. Traditional methods
using an optical microscope are tiring and tedious because of
many manipulations to maintain the cells in a healthy state while
they are on the microscope state. The quantification of the cell
adhesion and spreading requires a sophisticated live imaging
system and further data processing. When compared to the live
cell imaging, the ECIS system offers distinguishing features. For
example, this can quantitatively and continuously monitor cell
growth and dynamic behaviour over days with a temporal
resolution of minutes. The impedance data represent a statisti-
cally relevant ensemble and thus no further data processing is
needed. Moreover, this label-free and non-invasive approach can
be combined with conventional fluorescent assays to obtain an
additional information about the cellular behaviours.
Giaver and Keese have firstly monitored the behavior of
mammalian fibroblasts lines, such as WI-38 and WI-38/VA-13,
by measuring the impedance for long periods of time.11 Fig. 2
shows a typical result of the in- and out-of-phase voltages as
a function of time. The cells act as insulating particles and thus
play a passive role by blocking the current. After they are fully
spread, the measured impedance fluctuates with time. These
fluctuations result from the constant locomotion of the cells.
They also observed that normal and transformed cells showed
different signatures of impedance fluctuations. Subsequently,
they reported how normal and transformed cells interact with
four different protein layers (bovine serum albumin, gelatin,
bovine fetuin and human plasma fibronectin) coated on the
microelectrodes (i.e. fibronectin and gelatin were good
substrates).12 In addition, Kowolenko et al. monitored bone-
This journal is ª The Royal Society of Chemistry 2010
View Online
Fig. 2 A typical result of the ECIS measurement during cell culture. At
time zero, human fibroblastic cells were inoculated in an ECIS well giving
a final cell density of 105 per cm2.16
marrow macrophage (BMDM) attachment on protein-coated
surfaces and spreading in response to activation agents, such as
interferon-g (IFN-g), lipopolysaccharide (LPS) and heat-killed
Listeria monocytogenes (HKLM).13 The ECIS measurements
detected that the degree of spreading induced by either LPS or
HKLM (i.e. second signal agents related to cytolytic activity of
macrophages and protein synthesis) was different from that
induced by IFN-g (i.e. directly bind to its receptor and increase
protein kinase C activity). Furthermore, this simple and accurate
AC method can be useful for studying epithelial transport and
morphology. For example, Lo et al. investigated transepithelial
impedance of a confluent layer of Madin-Darby canine kidney
(MDCK) cells and developed a new theoretical model applicable
to the epithelial cells because their apical cell membranes were
connected and sealed with tight junctions.17 Wegener et al.
examined the kinetics of MDCK cell spreading on the electrodes
coated with different proteins and the influence of divalent
cations on its kinetics.20 They also quantified the inhibitory effect
of soluble Arg-Gly-Asp-Ser (RGDS) peptides, which could
compete with extracellular matrix proteins for the integrin
binding sites, on cell attachment. Similarly, the impedance
measurements can monitor cellular transformation (i.e. the first
step in cancer development) and assess the difference between
normal and transformed cells. For example, Park et al. quanti-
tatively examined the proliferation of NIH3T3 mouse fibroblasts
and the morphological changes associated with cellular trans-
formation.27 D143V_CXCR2 overexpression caused the contin-
uous activation of signaling pathways leading to the
transformation and this was reflected in the great increase of the
impedance.
Cell proliferation is an increase in cell populations as a result
of cell development and cell division (reproduction). In cells that
are dividing, the nuclear DNA should be replicated and then
distributed in a way that the two new cells receive a complete set
of genetic information each other. To prepare and accomplish
Analyst

these tasks, an orderly sequence of events (G1 phase, S phase, G2
phase and M phase) takes place in a cell leading its division and
duplication and the cell cycle is an essential mechanism in which
all living things reproduce. Obviously, normal cells have mech-
anisms for regulating progression through the cell cycle, such as
restriction point, DNA replication checkpoint and spindle
checkpoint. Contrary to normal cells, cancer cells continue to
grow and divide without halting at the restriction point of the cell
cycle even if under extremely adverse conditions. The most
common method for distinguish the different cell cycle stages is
to use cell cycle markers with fluorescence.28,29 However, these
markers can undergo dynamic changes and interactions
throughout the cell cycle. In addition, the cell cycle may be
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
perturbed by the markers. Recently, Wang et al. successfully
demonstrated the real-time dynamic analysis of the cell cycle in
Downloaded by Imperial College London on 26 October 2010
Fig. 3 (a) A schematic diagram of cell cycle. Under normal conditions,
the ability to pass through the restriction point is mainly governed by
external conditions, such as the presence of growth factors. On the other
hand, checkpoint pathways monitor conditions inside the cell and tran-
siently halt the cell cycle at various points if conditions are not appro-
priate for continuing. (b) normalized impedance as a function of time of
HeLa cells measured at 60 kHz. The cells on CIS1 and CIS2 were
synchronized by double thymidine block, whilst those on CIS3 were not
synchronized. No cell on CIS4 was control. The six numbered black
arrows indicate the exchange of culture medium.30
Analyst
View Online
live Human cervical carcinoma cells (HeLa) by the ECIS
measurements.30 To monitor the progression of the cell cycle,
HeLa cells were synchronized with double thymidine block and
then the time-course of the cellular impedance was obtained over
5 days (Fig. 3). The impedance increased in G1 phase and S phase
(i.e. Cells reattach and adhere to the electrode in G1 phase and
they grow and spread out in S phase), whilst it decreased in G2
phase and M phase (i.e. Cells round up and detach partially from
the electrode in G2 phase and divide into two daughter cells in M
phase). Notably, time-course impedance curves mirror the real-
time progression of the cell cycle in mammalian cells and the
shape of the curves is related to the synchronicity of the cell
population. We expect that this time-resolved cell cycle moni-
toring has great potentials in biomedical applications including
pharmacology, toxicology and high-throughput screening.
ECIS-based cancer research
To understand the framework in regard to the ECIS-based
cancer research, we intend to outline ‘signal transduction’ and
‘cell migration’. The cell (i.e. the functional basic unit of our life)
receives and acts on signals from beyond the plasma membrane.
Each information is converted into a specific cellular response
and this is a universal property of living cells (called ‘signal
transduction’). Cell behaviour is regulated by a complex network
of intracellular and extracellular signal transduction pathways.
Extracellular signaling molecules, such as cytokines or growth
factors or extracellular matrix proteins, bind to cell-surface
receptors and then the receptor-ligand interactions trigger events
inside the cell. The intracellular signaling cascades recruit the
signaling molecules including GTPases, Ca2+-regulated proteins
and protein kinases. Fig. 4 shows the major signal pathways
relevant to cancer in human cells.31 In particular, cancerous cells
possess mutated versions of oncogenes and tumor suppressors.
Such mutations alter signal pathway components in a way that
causes them to inappropriately deliver proliferative signals,
switching on cell growth, DNA replication and cell division.
Cancer cells ignore the signals from their environment that
normally keep cell proliferation under tight control. To date, two
types of genes, oncogenes and tumor suppressor genes, have been
recognized as playing a big role in cancer. Their discovery and
understanding have facilitated the development of new kinds of
cancer therapies. The ECIS-based assessment has been one of
important tools for these studies. Recently, Zudaire et al.
quantitatively assessed that knockdown of the aryl hydrocarbon
receptor repressor (AHRR), a bHLH/Per-ARNT-Sim tran-
scription factor located in a region of chromosome 5 (5p15.3), in
human lung cancers significantly enhanced in vitro anchorage-
dependent and –independent cell growth.32 Importantly, the
AHRR was suggested as a putative new tumor suppressor gene
in multiply type of human cancers. Transforming growth factors
(TGFs) are known as proteins to induce oncogenic trans-
formation in a specific cell culture system. Sawhney et al. applied
quantitative cell micromotion experiments to studying multiple
cell signaling mechanisms, such as extracellular signal-regulated
kinase/mitogen-activated protein kinase (ERK/MAPK), regu-
lating cell adhesion functions in human colon cancer cells.33 They
demonstrated that autocrine transforming growth factor-
a (TGF-a) controlled biological functions via activation of
This journal is ª The Royal Society of Chemistry 2010

Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
Downloaded by Imperial College London on 26 October 2010
Fig. 4 A schematic representation of the seven signaling pathways: JAK/STAT (orange), Notch, MAPK/ERK (blue), PI3K/AKT (green), NFkB
(brown), Wnt (yellow) and TGF-b (purple). All ligands are colored in red and transcription factors in white. Factors that exert an inhibitory effect onto
a particular pathway are framed with a black line. Cellular membrane and nuclear membrane are drawn in green and grey, respectively.31
p70S6K enzyme at specific phosphorylation sites. The prostate
transglutaminase (TGase-4) regulates the interaction between
prostate cancer and vascular endothelial cells. Jiang et al. found
that overexpression of prostate transglutaminase (TGase-4)
facilitated the adhesion of prostate cancer cells onto endothelial
cells and its action was independent of the Rho-associated kinase
(ROCK) pathway.34 Placenta growth factor (PlGF) is a key
molecule in angiogenesis and vasculogenesis and a member of the
VEGF family. Chen et al. investigated the direct biological
function of endogenous PlGF in human lung cancer cell line
A549.35 They used the ECIS system to determine the migration of
A549 cells and the knockdown of PlGF resulted in remarkably
reduced migrating capacity.
On the other hand, metastasis is the most feared aspect of
cancer. An ability to invade neighboring tissues is an unique
feature of malignant tumors. To date, although such invasiveness
is not thoroughly understood, it is of significant importance to
understand mechanisms of the adhesion and migration of the
malignant cancers. In general, cell migration has been well
studied in non-neoplastic cells, such as fibroblasts and epithelial
cells. It is a highly complex integrated process depending on the
actin-rich cortex beneath the plasma membrane. Three distinct
activities involved in cell crawling are as follows. Actin-rich
structures are pushed out at the front of cell (protrusion). The
actin skeleton connects across the plasma membrane to the
substratum (attachment) and the bulk of the trailing cytoplasm is
drawn forward (traction). The cycle is repeated and this results in
moving the cell forward in a stepwise function. It should be noted
that cell migration involves multiple processes regulated by
various signaling molecules. For example, the actin cytoskeleton
and its regulatory proteins are crucial for the migration in most of
cells. Integrins also mediate the signaling between the cell interior
and the extracellular matrix (ECM). These interactions serve not
only to anchor cells to the ECM but also to provide means of cell
migration. Importantly, cancer cells move within tissues during
invasion and metastasis by their own motility. Recently, new
cancer treatments, such as anticancer drugs, have been developed
This journal is ª The Royal Society of Chemistry 2010
View Online
for targeting specific signal transduction molecules. Importantly,
the ECIS measurements provide several advantages and/or
unique features for drug discovery when compared to other
chemotaxis assays including Boyden chamber assays: better
mimic of in vivo events, high sensitivity and real-time monitoring.
Malignant tumors undergo a series of changes when pro-
gressing the formation of metastasis. Many genes and factors
have been involved in the tumor invasion including adhesion
molecules, cytoskeletal proteins and proteases.36 To investigate
the function of such factors in physiological and pathological cell
invasion, a quantitative invasion assay in real-time is of signifi-
cant importance. Ren et al. conducted invasion assays of ovarian
cancer cells (SKOV3 and HEY) using the ECIS system.37 They
assessed the ability of SKOV3 cells to invade into a confluent
monolayer of peritoneal mesothelial cells producing lysophos-
phatidic acid (LPA) since LPA is a potential diagnostic marker
and a therapeutic target for ovarian cancer. Cell invasion
through the mesothelial cell layer resulted in a dramatic drop in
resistance, whilst other cell types, such as NIH3T3 and MCF7
cells, didn’t cause a significant drop in resistance (Fig. 5a).
Through the transfection with siRNAs against LPA1–3 as well as
control siRNAs against glyceraldehyde-3-phosphate dehydro-
genase (GAPDH) or green fluorescent protein (GFP), they
determined whether the receptors for LPA were involved in cell
invasion (Fig. 5b). siRNAs against LPA2 and LPA3 significantly
inhibited the invasion process, whilst those against LPA1 and the
controls had no influence on cell invasion. They suggested that
LPA2 and LPA3 were strongly related to the cellular functions of
SKOV3 cells. In addition, they pretreated the mesothelial cells
with HELSS (an inhibitor of the calcium-independent phos-
pholipase A2, iPLA2) and AACOCF3 (an inhibitor of both
cytosolic PLA2 (cPLA2) and iPLA2). They showed both inhibi-
tors delayed invasion time and reduced invasion extent (Fig. 5c).
CD44 is a cell-surface glycoprotein associated with various
adhesion-dependent cellular processes including cell migration,
tumor cell metastasis and invasion. Kuo et al. studied the inva-
sive behavior of human breast carcinoma cell lines by the time-
Analyst
 View Online
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
Downloaded by Imperial College London on 26 October 2010

Fig. 5 (a) Invasion of ovarian cancer cells (SKOV3) into peritoneal mesothelial cells (LP9). SKOV3, NIH3T3 or MCF7 cells (1  105 cells in 300 ml
medium) were added on top of the LP9 cell layer. 300 ml-medium without cells was added as control. (b) the inhibitory effect of siRNAs against LPA1–3,
GAPDH and GFP on cell invasion. (c) the effect of inhibitors (HELSS and AACOCF3)on the invasion of SKOV3 cells. LP9 cell monolayers were
pretreated with HELSS (1 mmol/l, 30 min) or AACOCF3 (100 mmol/l, 30 min) before adding SKOV3 cells.37

course impedance measurements.38 They monitored the real-time Cytotoxicity

invasion of human breast carcinoma cell lines (MDA-MB-435
cells) to human umbilical vein endothelial cells (HUVECs). In
Fig. 6a, MDA-MB-435 cells were pretreated with TGF-b for 12
or 24 h and then they added to the HUVEC monolayer after 1 h
recording. TGF-b treated cells substantially invaded the typical
endothelial cells when compared to untreated cells. The break-
down in the cell-cell contact between HUVEC cells resulted the
rapid decrease in resistance after 4 h. They elucidated that TGF-
b would induce CD44 cleavage on the cell surface and this
enhanced the migration and invasiveness of the MDA-MB-435
cells. Similarly, Saxena et al. quantitatively investigated the effect
of leptin, a key in the regulation of energy balance and body
weight control, on hepatocellular carcinogenesis.39 After
HUVECs were fully spread, the confluent HUVEC layers were
challenged with hepatocellular carcinoma cell lines, such as
HepG2 and Huh 7 cells. The reduction of impedance indicated
that the invasion resulted from retraction of the endothelial cell
junctions and extravasation of the two cells on the substratum
(Fig. 6b). Interestingly, leptin-treated cells showed more steeper
decrease in impedance when compared to controls. In addition,
they evaluated the effect of pharmacologic inhibitors on the
signaling pathways regulated by the leptin. The leptin was related
to activation of signal transducers and activators of transcription
3 (STAT3), AKT, and extracellular signal-regulated kinase
(ERK) signaling pathways and leptin-induced phosphorylation
of AKT and ERK was dependent on Janus-activated kinase
(JAK)/STAT activation. The inhibitors of JAK/STAT effectively
blocked leptin-induced invasion.
Cytotoxicity is the cell-killing property of chemical compounds
including food, cosmetics and drugs or a mediated cell such as
a cytotoxic T cell. This is independent from the mechanisms of
death; necrosis (accidental cell death) and apoptosis (pro-
grammed cell death). Cell-based assays can provide essential
information about the potential effects of chemical compounds
on specific cell properties. It should be noted that cells are finely
balanced homeostatic machines that respond to external stimuli
through complex pathways. Accordingly, toxicity could be the
result of a multitude of cellular events including the changes in
cell morphology, differentiation, proliferation, function and/or
communication. Cytotoxicity assays are widely used to screen
cytotoxic compounds in the pharmaceutical industry or ‘hits’
without unwanted cytotoxic effects. For example, vital dyes,
such as trypan blue or propidium iodide, are utilized to assess cell
membrane integrity. Biochemical methods, such as the MTT or
MTS assay,40,41 the neutral red uptake assay42 and ATP-based
assays, have been commonly used to assess viable (living) cells.
Other cytotoxicity tests are based on the pH changes induced by
acidic metabolites in the neighborhood of cultured cells.43 In
addition, intracellular motion of small vesicles or organelles
responded to toxicants are characterized by image analysis and
video-enhanced microscopy. Although each of these techniques
has its own advantages and disadvantages, these traditional
methods are time consuming and require complex steps with
multiple reagents. Accordingly, it is highly on demand to
continuously monitor cell behavior against toxic effects and to

Analyst This journal is ª The Royal Society of Chemistry 2010

 View Online
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
Downloaded by Imperial College London on 26 October 2010

Fig. 6 (a) the effect of TGF-b treatment on transendothelial migration.38 Resistance changes in impedance are challenged with MDA-MB-435 cells.
The cancer cells with or without TGF-b (10 ng/ml) were added to the HUVECs monolayer after 1 h of recording. (b) that of leptin treatment on cell
invasion potential of hepatocellular carcinoma cells.39 HepG2 and Huh7 cells were cultured in Matrigel invasion chambers in serum-free media con-
taining 100 ng/ml leptin (L) for 24 h. For combined treatment, cells were pretreated with 100 mmol/l AG490 (L + AG), 10 mmol/l PD98059 (L + PD) or
10 mmol/l LY294002 (L + LY) for 45 min followed by leptin treatment. Resistance changes in the impedance at 4 kHz as confluent layers of HUVECs
were challenged with HepG2 and Huh7 cell suspensions. The control was the addition of media without HepG2 or Huh7 cells.

accurately analyze the change in cell viability for a long term.
Recently, ECIS has been considered as a tool for in vitro toxicity
testing.44–50 Xiao et al. continuously monitored the impedance of
attached cells exposed to toxicants, such as cadmium chloride,
sodium arsenate and benzalkonium chloride.44–46 Fig. 7 shows
the cytotoxicity effect as a function of concentration of all three
compounds tested. Importantly, the length course of toxicant
exposure made the attached and infected cells die or detach from
the working electrode and this corresponded to a gradual
decrease in capacitance. In addition, the inhibition assays
provided half-inhibition concentration in real time that agreed
with that of the standard neutral red assay. They also assessed
the cytotoxicity of mercury chloride as a model of acute toxicants
and 1,3,5-trinitrobenzene (TNB) as that for studying long-term
cytotoxicity effects using the ECIS measurements. Recently,
a multiarrayed electrode system has been developed to achieve
high-throughput cytotoxicity testing.47–50 Yeon et al. demon-
strated a microfabricated cell chip to monitor cell growth and
cytotoxic effects of human hepatocellular carcinoma cells
(HepG2) after treatments with several toxicants, such as
tamoxifen and menadione.47 Opp et al. evaluated dose-depen-
dent responses of HUVEC in response to cytochalasin B known
to disrupt actin filaments and rapidly affect cellular
morphology.48 Interestingly, Curtis et al. showed that four of 10
cell lines were appropriate for use in the ECIS system and two of
these, bovine pulmonary artery endothelial cells (BLMVECs)
and iguana heart (IgH-2) cells, responded well with improved
sensitivity to waterborne industrial chemicals.49 Very recently,
Yun et al. determined the effect of magnesium ions on osteoblast
(U2-OS) cell activity using an eight-well ECIS electrode.50

This journal is ª The Royal Society of Chemistry 2010 Analyst


Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
Downloaded by Imperial College London on 26 October 2010
Fig. 7 (Left) Time-course responses of fibroblast V79 cells exposed to
cytotoxic chemicals (mM): (top) cadmium chloride, (a) 2.9, (b) 4.6, (c) 6.2
and (d) 8.1; (middle) benzalkonium chloride, (a) 15.2, (b) 18.3, (c) 21.3
and (d) 30.4; sodium arsenate, (a) 45, (b) 60, (c) 140 and (d) 200. and
(Right) inhibition curves derived from the time-course responses at t0 ¼
16 h. Half-inhibition concentrations of CdCl2, BAK and NaHAsO4 were
4.0, 14.0 and 50.1 mM, respectively.44
Wound healing
Wound healing assays allow for studying cell migration, cell-cell
interactions and proliferation rates under a variety of culture
conditions. In general, a small area of a cell monolayer is disrupted
or displaced (creating a ‘wound’), the images at the beginning and
at regular intervals are captured and then the images are
compared to quantify both migration and proliferation rates of
Fig. 8 Electrical wound-healing assays: (a) different time exposures of an elevated field, (b) different size electrodes and (c) multiple wounding and
recovery.21 BS-C-1 cells were grown to confluence in cell culture wells before measurements started. The impedance measured at 4 kHz is plotted as
a function of time.
Analyst
View Online
the cells. This healing can take from several hours to over a day
and this depends on cell types, culture conditions and the extent of
the wound region. Instead of mechanical damage to the cells, an
electrical means to wound confluent cell layers can be imple-
mented to the ECIS measurements. Noiri et al. firstly used a DC
pulse to kill the cells and then monitored the repopulation of the
wounded area.51 However, DC signals were not suitable to con-
trollably damage the cells because of the electrochemical genera-
tion of toxic species or the damage of the electrode area depending
on the polarity of the wounding current. Recently, they circum-
vented the problems by using milliampere-level AC wounding
currents at relatively high frequency in the ECIS system.21
Accordingly, coupling invasive and non-invasive features would
enable the ECIS system to comprise an entirely automated means
to carry out wound-healing assays. When the wounding current
was applied to the cell-covered electrode, the plasma membranes
were electroporated and this resulted in cell killing. Fig. 8a shows
the response of cells to different time exposures of the wounding
current and their recovery. In the next few hours depending on
a degree of killed cells, the resistance was increased back to the
levels of the cell-covered electrode because of the migration of cells
from its perimeter inward to replaced the dead cells. In addition,
they also investigated the effects of electrode sizes and cell types on
the wound-healing assays (Fig. 8b) and their reproducibility after
multiple wounding (Fig. 8c). The ECIS-based wound healing
assays have been successfully demonstrated with different cell
lines including BS-C-1 (African green monkey kidney cells; CCL-
26), NRK (normal rat kidney cells; CRL-6509), MDCK (Madin-
Darby canine kidney cells; CLL-34) and endothelial cells.
Conclusions
In this Minireview, we have described that impedance-based
cellular assays have distinct advantages for cancer biology and
This journal is ª The Royal Society of Chemistry 2010

drug discovery. Fortunately, entire ECIS systems, such as
ECIS from Applied Biophysics, xCELLigence from ACEA
Biosciences and Roche and Cell Key from MDS Sciex, are now
commercially available and thus they have allowed for better
approaches for cancer diagnosis, treatment and prevention.
Accordingly, we believe that these label-free and non-invasive
measurements in real-time are undoubtedly crucial to understand
the behaviours of cancer cells and eventually to design new
anticancer drugs. In addition, we expect that the ECIS
measurements will be readily integrated with standard end-point
assays to improve the quality and reproducibility of measured
data. In particular, it is desirable to simultaneously perform both
optical and electrochemical investigations. For this purpose,
Published on 20 October 2010 on http://pubs.rsc.org | doi:10.1039/C0AN00560F
transparent electrode materials, such as indium tin oxide (ITO)
and Poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate)
(PEDOT:PSS), can receive much attention if ECIS is combined
with inverted phase-contrast and fluorescence microscopes.
Downloaded by Imperial College London on 26 October 2010
Furthermore, the addition of nano- and microfluidic systems will
enable us to perform complex processing and analysis operations
in the ECIS format.52,53
Acknowledgements
J. Hong acknowledges the Brain Korea 21 Project in 2010. This
research was supported by the Kyungwon University Research
Fund in 2010. This work was also supported by the WCU
Program (R33-2008-000-10067-0) of Korean Ministry of
Education & Science Technology and by the grant (Grant No:
2010-B04) from Gyeonggi Regional Research Centre (GRRC)
Program, Republic of Korea.
Notes and references
1 B. Alberts, A. Johnson, J. Lewis, M. Raff, K. Roberts, P. Walter,
‘‘Molecular Biology of the Cell’’, 2002, 4th ed., Garland Science.
2 L. J. Kleinsmith, ‘‘Principles of Cancer Biology’’, 2006, Pearson
International Edition.
3 D. Nahari, R. Satchi-Fainaro, M. Chen, I. Mitchell, L. B. Task,
Z. Liu, J. Kihneman, A. B. Carroll, L. S. Terada and
F. E. Nwariaku, Mol. Cancer Ther., 2007, 6, 1329.
4 American Cancer Society, Cancer Facts and Figures, American
Cancer Society, Inc., 2003.
5 R. E. White, Annu. Rev. Pharmacol. Toxicol., 2000, 40, 133.
6 J. Voros, R. Graf, G. L. Kenausis, A. Bruinink, J. Mayer, M. Textor,
E. Wintermantel and N. D. Spencer, Biosens. Bioelectron., 2000, 15,
423.
7 T. S. Hug, Assay Drug Dev. Technol., 2003, 1, 479.
8 J. M. Atienza, N. Yu, S. L. Kirstein, B. Xi, X. Wang, X. Xu and
Y. A. Abassi, Assay Drug Dev. Technol., 2006, 4, 597.
9 B. Xi, N. Yu, X. Wang, X. Xu and Y. A. Abassi, Biotechnol. J., 2008,
3, 484.
10 M. Tarantola, A.-K. Marel, E. Sunnick, H. Adam, J. Wegner and
A. Janshoff, Integr. Biol., 2010, 2, 139.
11 I. Giaever and C. R. Keese, Proc. Natl. Acad. Sci. U. S. A., 1984, 81,
3761.
12 I. Giaever and C. R. Keese, IEEE Trans. Biomed. Eng., 1986, 33, 242.
13 M. Kowolenko, C. R. Keese, D. A. Lawrence and I. Giaever, J.
Immunol. Methods, 1990, 127, 71.
14 I. Giaever and C. R. Keese, Proc. Natl. Acad. Sci. U. S. A., 1991, 88,
7896.
15 C. Tiruppathi, A. B. Malik, P. J. D. Vecchio, C. R. Keese and
I. Giaever, Proc. Natl. Acad. Sci. USA, 1992, 89, 7919.
16 I. Giaever and C. R. Keese, Nature, 1993, 366, 591.
This journal is ª The Royal Society of Chemistry 2010
View publication stats
View Online
17 C.-M. Lo, C. R. Keese and I. Giaever, Biophys. J., 1995, 69, 2800.
18 L. Reddy, H.-S. Wang, C. R. Keese, I. Giaever and T. J. Smith, Exp.
Cell Res., 1998, 245, 360.
19 E. Noiri, E. Lee, J. Testa, J. Quigley, D. Colflesh, C. R. Keese,
I. Giaever and M. S. Goligorsky, Am. J. Physiol., 1998, 274, C236.
20 J. Wegener, C. R. Keese and I. Giaever, Exp. Cell Res., 2000, 259, 158.
21 C. R. Keese, J. Wegener, S. R. Walker and I. Giaever, Proc. Natl.
Acad. Sci. U. S. A., 2004, 101, 1554.
22 I. Giaever and C. R. Keese, Proc. Natl. Acad. Sci. U. S. A., 1991, 88,
7896.
23 C.-M. Lo and J. Ferrier, Eur. Biophys. J., 1999, 28, 112.
24 B. Burns, D. Walker, E. Brown, L. Thurmon, R. Bowden,
C. R. Keese, S. Simon, M. Entman and C. Smith, J. Immunol, 1997,
159, 2893.
25 A. R. Burns, R. A. Bowden, S. D. MacDonell, D. C. Walker,
T. O. Odebunmi, E. M. Donnachie, S. I. Simon, M. L. Entman and
C. W. Smith, J. Cell. Sci, 2000, 113, 45.
26 D. C. Lovelady, T. C. Richmond, A. N. Maggi, C.-M. Lo and
D. A. Rabson, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys.,
2007, 76, 041908.
27 G. Park, C. K. Choi, A. E. English and T. E. Sparer, Cell Biol. Int.,
2009, 33, 429.
28 N. Thomas and I. D. Goodyer, Targets, 2003, 2, 26.
29 H. P. Easwaran, H. Leinhardt and M. C. Cardoso, Cell Cycle, 2005, 4,
453.
30 L. Wang, L. Wang, H. Yin, W. Xing, Z. Yu, M. Guo and J. Cheng,
Biosens. Bioelectron., 2010, 25, 990.
31 O. Dreesen and A. H. Brivanlou, Stem Cell Reviews and Reports,
2007, 3, 7.
32 E. Zudaire, N. Cuesta, V. Murty, K. Woodson, L. Adams,
N. Gonzalez, A. Martınez, G. Narayan, I. Kirsch, W. Franklin,
F. Hirsch, M. Birrer and F. Cuttitta, J. Clin. Invest., 2008, 118, 640.
33 R. S. Sawhney, M. M. Cookson, B. Sharma, J. Hauser and
M. G. Brattain, J. Biol. Chem., 2004, 279, 47379.
34 W. G. Jiang, R. J. Ablin, H. G. Kynaston and M. D. Mason,
Microvasc. Res., 2009, 77, 150.
35 J. Chen, L. Ye, L. Zhang and W. G. Jiang, J. Cell. Biochem., 2008,
105, 313.
36 G. K. Wang and W. Zhang, Histol Histopathol, 2005, 20, 593.
37 J. Ren, Y. J. Xiao, L. S. Singh, X. Zhao, Z. Zhao, L. Feng,
T. M. Rose, G. D. Prestwich and Y. Xu, Cancer Res., 2006, 66, 3006.
38 Y.-C. Kuo, C.-H. Su, C.-Y. Liu, T.-H. Chen, Chie-Pein Chen and H.-
S. Wang, Int. J. Cancer, 2009, 124, 2568.
39 N. K. Saxena, D. Sharma, X. Ding, S. Lin, F. Marra, D. Merlin and
F. A. Anania, Cancer Res., 2007, 67, 2497.
40 R. Ni, M. A. Leo, J. Zhao and C. S. Lieber, Alcohol and Alcoholism,
2001, 36, 281.
41 A. P. Li, C. Lu, J. A. Brent, C. Pham, A. Facket, C. E. Ruegg and
P. M. Siber, Chem.-Biol. Interact., 1999, 121, 17.
42 G. Repetto, A. del Peso and J. L. Zurita, Nat. Protoc., 2008, 3, 1125.
43 J. W. Parce, J. C. Owicke, K. M. Kercso, G. B. Sigal, H. G. Wada,
V. C. Muir, L. J. Bousse, K. L. Ross, B. I. Sikic and
H. M. McConnell, Science, 1989, 246, 243.
44 C. Xiao, B. Lachance, G. Sunahara and J. H. T. Luong, Anal. Chem.,
2002, 74, 5748.
45 C. Xiao and J. H. T. Luong, Biotechnol. Prog., 2003, 19, 1000.
46 C. Xiao and J. H. T. Luong, Toxicol. Appl. Pharmacol., 2005, 206,
102.
47 J. H. Yeon and J.-K. Park, Anal. Biochem., 2005, 341, 308.
48 D. Opp, B. Wafula, J. Lim, E. Huang, J.-C. Lo and C.-M. Lo,
Biosens. Bioelectron., 2009, 24, 2625.
49 T. M. Curtis, J. Tabb, L. Romeo, S. J. Schwager, M. W. Widder and
W. H. van der Schalie, J. Appl. Toxicol., 2009, 29, 374.
50 Y. H. Yun, Z. Dong, Z. Tan and M. J. Schulz, Anal. Bioanal. Chem.,
2010, 396, 3009.
51 E. Noiri, Y. Hu, W. F. Bahou, C. R. Keese, I. Giaever and
M. S. Goligorsky, J. Biol. Chem., 1997, 272, 1747.
52 L. Kang, B. G. Chung, R. Langer and A. Khademhosseini, Drug
Discovery Today, 2008, 13, 1.
53 J. Hong, J. B. Edel and A. J. deMello, Drug Discovery Today, 2009,
14, 134.
Analyst