Embed Size (px)
Citation preview
Ai
OCD
a
ARRAA
KTIBMICH
1
atMncpcpeMwsoc(
0d
Biosensors and Bioelectronics 26 (2011) 2376–2382
Contents lists available at ScienceDirect
Biosensors and Bioelectronics
journa l homepage: www.e lsev ier .com/ locate /b ios
cell-based impedance assay for monitoring transient receptor potential (TRP)on channel activity
liver Pänke ∗, Winnie Weigel, Sabine Schmidt, Anja Steude, Andrea A. Robitzkientre for Biotechnology and Biomedicine (BBZ), University of Leipzig, Division of Molecular Biological-Biochemical Processing Technology, Deutscher Platz 5,-04103 Leipzig, Germany
r t i c l e i n f o
rticle history:eceived 4 August 2010eceived in revised form 8 October 2010ccepted 11 October 2010vailable online 16 October 2010
eywords:RP cation channelmpedance spectroscopyioimpedanceicroelectrode array
a b s t r a c t
Transient receptor potential (TRP) channels are non-selective ion channels permeable to cations includingNa+, Ca2+ and Mg2+. They play a unique role as cellular sensors and are involved in many Ca2+-mediatedcell functions. Failure in channel gating can contribute to complex pathophysiological mechanisms. Dys-functions of TRP channels cause diseases but are also involved in the progress of diseases. We present anovel method to analyse chemical compounds as potential activators or inhibitors of TRP channels to pro-vide pharmaceutical tools to regulate channel activity for disease control. Compared to common methodssuch as patch clamp or Ca2+ imaging, the presented impedance assay is automatable, experimental lessdemanding and not restricted to Ca2+ flow. We have chosen TRPA1 from the TRPA (‘ankyrin’) family as amodel channel which was stimulated by allyl isothiocyanate (AITC). HEK293 cells stably transfected withhuman TRPA1 cDNA were grown on microelectrode arrays. Confluent cell layers of high density were
ndium tin oxidehip-based screening assayigh content screening
analysed. Impedance spectra of cell-covered and non-covered electrodes yielded a cell-specific signal atfrequencies between 70 and 120 kHz. Therefore, 100 kHz was chosen to monitor TRPA1 activity there-upon. An average impedance decrease to about 70% of its original value was observed after application of10 �M AITC indicating an increased conductance of the cell layer mediated by TRPA1. Transfected cellspretreated with 10 �M of inhibitor ruthenium red to prevent channel conductance, as well as controlcells lacking TRPA1, showed no impedance changes upon AITC stimuli demonstrating the specificity of
y.
the novel impedance assa. Introduction
The transient receptor potential (TRP) superfamily consists oflarge number of cation channels. Most of them are permeable
o both monovalent and divalent cations including Na+, Ca2+ andg2+ (Pedersen et al., 2005; Ramsey et al., 2006). TRP cation chan-
els are expressed in almost every tissue and cell type. Due to theirentral role in Ca2+-mediated cell function and many homeostaticrocesses such as Ca2+ and Mg2+ reabsorption, failure in correcthannel gating or permeation will likely contribute to complexathophysiological mechanisms that cause human diseases (Inouet al., 2006; Nilius, 2007; Kiselyov et al., 2007; Venkatachalam andontell, 2007; Yamamoto et al., 2007; Watanabe et al., 2008). Here,e present a novel method based on impedance spectroscopy to
creen chemical compounds as potential activators or inhibitorsf TRP channels to provide pharmaceutical tools to regulate TRPhannel activity for symptomatic and therapeutical disease controlPatapoutian et al., 2009). Compared to common methods used for
∗ Corresponding author. Tel.: +49 0341 97 31242; fax: +49 0341 97 31249.E-mail addresses: [email protected], [email protected] (O. Pänke).
956-5663/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.bios.2010.10.015
© 2010 Elsevier B.V. All rights reserved.
the analysis of TRP channel gating such as patch clamp or opti-cal techniques like Ca2+ imaging (Hill and Schaefer, 2007, 2009),the presented cell-based impedance assay employs microelectrodearrays (MEA) (Rahman et al., 2006; Wolf et al., 2008; Jahnke et al.,2009; Krinke et al., 2009) which is a promising approach of increas-ing interest for high throughput applications where a high degreeof parallelisation is needed (Asphahani and Zhang, 2007; Yang andBashir, 2008; Sadik et al., 2009). The presented assay is automat-able, less demanding in respect to the experimental abilities of theconducting technician (patch clamp) and not restricted to the flowof Ca2+ ions (Ca2+ imaging).
Cell-based impedance spectroscopy is stimulated by thepioneered work of Giaever and Keese (1984) and allows a com-prehensive electrical characterisation of adherent cell layers onmicroelectrodes. It is qualified to detect morphological changes(Giaever and Keese, 1993; Keese and Giaever, 1994), cell attach-ment and spreading (Mitra et al., 1991; Asphahani et al., 2008) as
well as cell motility and migration (Giaever and Keese, 1991; Loet al., 1993; Wegener et al., 2000) in real-time without any inter-fering labelling techniques. Cell-based impedance spectroscopy iscommonly used to detect morphological changes within an unitedcell structure as a response to an external stimulus such as viral
O. Pänke et al. / Biosensors and Bioelectronics 26 (2011) 2376–2382 2377
F uctane microi m2.p ion su
ie1e(mobotiiid2Na(cc
(aTamaetv
ig. 1. (A) Microelectrode array (MEA) used for detection of TRPA1 ion channel condlectrodes with diameters of 50, 100 and 200 �m, which are transparent and allows approximately 570 �m and the ring-shaped counter electrode has an area of 1.8 maths defining the measured impedance. For better cell adherence a collagen adhes
nfection (Niikura et al., 2004; Mccoy and Wang, 2005; Campbellt al., 2007) or chemical compounds of cytotoxic (Tiruppathi et al.,992; Keese et al., 1998; Ko et al., 1998; Arndt et al., 2004; Xingt al., 2005; Krinke et al., 2009; Opp et al., 2009) or pharmacologicalWolf et al., 2008; Jahnke et al., 2009) relevance. Such stimuli pri-
arily affect the electrical conductance of the intercellular spacer the cell–electrode interface rather than changes of cell mem-rane conductance or capacitance. The applied microelectrodes areften non-transparent and do not allow microscopic imaging ofhe adherent cells. Ion channel characterisation using cell-basedmpedance spectroscopy is rarely found in the literature, althoughon channels located in cell membranes have crucial roles in phys-ology and pathophysiology and are important targets for drugiscovery (Curran, 1998; Wissenbach et al., 2004; Okuhara et al.,007; Jegla et al., 2009; Patapoutian et al., 2009; Mathie, 2010;assini et al., 2010). Impedimetric single cell analysis of ion channelctivity in bovine chromaffin cells might serve as a rare exampleHan and Frazier, 2006) where the observed impedance changesould be directly attributed to a molecular compound within theell membrane.
We have chosen human transient receptor potential ankyrin1TRPA1, also known as ANKTM1) from the TRPA (‘ankyrin’) familys model for the impedimetric analysis of ion channel conductance.RPA1 is expressed in sensory neurons (Story et al., 2003) and hascentral role in the pain response to endogenous inflammatory
ediators (Kwan et al., 2006; Bautista et al., 2006) and to a diverserray of chemical irritants such as allyl isothiocyanate (AITC) (Jordtt al., 2004; Hinman et al., 2006; Macpherson et al., 2007). AITC ishe main component of mustard oil and often used for TRPA1 acti-ation in in vitro analyses. Channel-specific inhibitors are rare, less
ce. The custom-made MEA 9W6E has nine wells, each containing three pairs of ITOscopic imaging of the adherent cell layer. The distance between adjacent electrodes(B) Scheme of adherent HEK293 cells on the electrode surface and possible currentpport was applied (not to scale).
specific, targeting various members of the TRP channel family andrequire high working concentrations. However, inhibitor ruthe-nium red is known to suppress AITC-stimulated ion conductanceof TRPA1 in in vitro analyses (Nagata et al., 2005).
Here, we demonstrate the monitoring of stimulated ion con-ductance of heterologously expressed TRPA1 using cell-basedimpedance spectroscopy which gauges changes of the TRPA1-mediated membrane conductance. HEK293 cells were used asexpression system and cultured on microelectrode arrays (MEA)consisting of indium tin oxide (ITO) or platinum microelectrodes,respectively. The application of transparent ITO electrodes alloweddirect optical control of cell growth and adherence on the electrode.Fig. 1A shows the geometry of the self-developed ITO-based MEA,the arrangement of the electrodes and adherent cells on a trans-parent ITO microelectrode. Under these conditions, the measuredimpedance of the cell-covered microelectrode is defined by threepossible current paths (Fig. 1B, see also Giaever and Keese, 1991;Arndt et al., 2004): The channel-specific current depends on theion gating status of the membrane-embedded channel protein, theinter cell leak current is influenced by morphological changes ofthe united cell layer, and finally to a minor extent, current can bepassed unspecifically across the cell membrane. For reliable detec-tion of TRPA1 activity, the fraction of the latter two needs to besufficiently small and the TRPA1 channel-dependent current has todominate the measured impedance behaviour.
The pursued goal of this work was to demonstrate that cell-based impedance spectroscopy is qualified to monitor the activityof specific ion channels within cell membranes which opens theopportunity for automated analysis of ion channel related pharma-ceutical compounds. Additionally, we successfully tested the ability
2 Bioele
on
2
2
(DGtiID11Hef1clesmimmgwAtw
2
fds(phmdsattrbswadiOa23ccMr
378 O. Pänke et al. / Biosensors and
f ITO as transparent electrode material that allows combining theovel impedance assay with optical techniques.
. Materials and methods
.1. Cell culture
HEK293 cells, stably transfected with human TRPA1 cDNAHEK TRPA1), were kindly provided by Dr. Kerstin Hill and Prof.r. Michael Schaefer (Rudolf-Boehm-Institut, Universität Leipzig,ermany) (Hill and Schaefer, 2007). TRPA1 expression within
he transfected HEK293 cells was checked by immunocytochem-cal imaging and mRNA expression analysis (see Supplementalnformation). Cells were cultured in non-covered culture flasks inulbecco’s modified Eagle medium (DMEM), supplemented with0% fetal bovine serum, 2 mM glutamax, 100 units/ml penicillin,00 �g/ml streptomycin and 500 �g/ml geneticin. Non-transfectedEK293 cells lacking TRPA1 (HEK wt) were maintained in minimumssential medium (MEM) with Earle’s salt, supplemented with 10%etal bovine serum, 2 mM glutamax, 100 units/ml penicillin and00 �g/ml streptomycin. Impedimetric control experiments werearried out using also HEK293 FT cells from Invitrogen. This celline is usually used for simplified establishment of recombinantxpression systems with lentiviral constructs. However, these cellshowed also an optimised cell growth and better cell adherenceaking them perfectly suitable for impedimetric control exper-
ments. HEK293 FT were cultured in Dulbecco’s modified Eagleedium, supplemented with 10% fetal bovine serum, 2 mM gluta-ax, 100 units/ml penicillin, 100 �g/ml streptomycin and 20 �g/ml
entamicin. All compounds were supplied by Invitrogen. All cellsere maintained at 37 ◦C in a humified atmosphere with 5% CO2.ccutase (PAA Laboratories) was used to detach adherent cells gen-
ly to protect the membrane embedded TRPA1 channel proteinhen splitting the confluent cell cultures.
.2. Microelectrode arrays
Two different microelectrode arrays (MEA) have been evaluatedor their ability to monitor ion channel activity. MEA 9W6E wasesigned and fabricated by ourselves. It consists of nine separatedample chambers (wells) with six electrodes of three different sizesFig. 1A). The 9-well geometry matches the standardised 96-well-late format. The electrode material is indium tin oxide (ITO) whichas the important advantage of being transparent and allowedicroscopic imaging of cell growth on the electrode (Fig. 1A). The
iameters of the electrodes are 50, 100 and 200 �m. Preliminarytudies revealed that the smallest electrode already yielded anppropriate cell-specific impedance signal. Therefore and for bet-er comparison, all data presented in this report originate fromhe 50 �m electrodes. The circular working electrodes are sur-ounded by a ring-shaped counter electrode. The size differenceetween working and counter electrode guaranteed that the mea-ured impedance was controlled by the adherent cell layer on theorking electrode. The area ratio between the counter and the
nalysed 50 �m working electrode is 900. MEA 9W6E has beenesigned using the AutoCAD software (Autodesk) and fabricated
n our own clean room facility. The borosilicate substrate (Docterptics, 49 mm × 49 mm × 1 mm) was cleaned consecutively withcetone, propanol and piranha solution (95% H2SO4 and 30% H2O2,:1 (v/v)) prior to coating. Afterwards, 3.5 �m positive resist AR-
510 (Allresist) was spin-coated onto the glass substrate (spinoater Primus SB15, SSE). MEA structures were transferred usingustom-made photomasks (MLC Jena), the mask aligner MJB4 (SüssicroTec) and an UV light exposure of 350 nm. Subsequently, theesist was developed using AR 300-35 (Allresist). Afterwards, the
ctronics 26 (2011) 2376–2382
ITO layer was deposited by alternating current sputtering (BAE 250,BalTec). The remaining positive resist and the ITO sputtered there-upon were removed by acetone. Finally, 5 �m of the negative resistSU-8 (MicroChem) were deposited for isolation of the conductorpaths.
The commercially available MEA 4W15E was purchased fromAyanda Biosystems (MEA60 4Well) and served as control. MEA4W15E consists of only four wells, each containing a rectangulararray of 15 circular platinum working electrodes with a diameter of40 �m. Although the platinum electrodes are non-transparent, theyare deposited on a transparent glass substrate. This allows a lim-ited evaluation of cell growth and adherence next to the platinumelectrodes.
2.3. Impedance spectroscopy
Even though all HEK293 cells grew on electrode surfaceswithout adhesion support, cells were cultured on collagen-coated microelectrode arrays (MEA) to provide robust fixation ofthe cell layer and prevent mechanical disruption upon reagentaddition. Collagen coating was obtained after incubation with40 �g/ml collagen R (Serva) in 0.02 M acetic acid for 1 h at 37 ◦C.50,000–100,000 cells were seeded in each well (150 �l) and cul-tured until a confluent cell layer was achieved (typically after 2–3days culturing the cells at 37 ◦C in a humified atmosphere with5% CO2). Cell growth was followed by phase contrast microscopy(Nikon Eclipse TE2000-U) and only densely cell-covered elec-trode arrays were selected for impedimetric analysis. MEA wereplaced into an adaptor board (MicroChannelSystems) on a hotstage at 37 ◦C (Minitüb HT200). Impedance was recorded with afrequency response analyser (1260A, Solartron Analytical) con-nected to a dielectrical interface (1296A) and an alternatingvoltage of 10 mV in 2-electrode configuration (0 mV bias betweenelectrodes) that is proven and tested in cell-based impedanceapplications (Wolf et al., 2008; Kloss et al., 2008; Rahman et al.,2008; Jahnke et al., 2009; Krinke et al., 2009). Each cell layerwas initially analysed by recording a complete frequency spec-trum ranging from 600 kHz to 100 Hz (equidistant logarithmicspacing, 10 points per decade) using the software package SMaRTv2.8 (Solartron Analytical). For data analysis, a relative impedancespectrum was calculated according to �|Z|rel = (|Z|cov − |Z|o)/|Z|owhere |Z|cov and |Z|o were the impedance magnitudes of thecell-covered and non-covered electrode, respectively. The asso-ciated reference spectrum |Z|o of the non-covered electrode wasrecorded after removal of the cell layer by trypsin/EDTA treatment(2.5 g/l trypsin, 0.38 g/l EDTA-Na4, Invitrogen) at the end of eachexperiment. Only experiments showing a relative impedance max-imum of at least �|Z|rel,max = 0.7 were used for statistical analysis.For measuring TRPA1 activity, impedance was recorded subse-quently at a single frequency of 100 kHz every 3 s for 10 minin the presence of the respective culture medium. After 2.6 min,10 �M of the TRPA1-specific activator allyl isothiocyanate (AITC,Sigma-Aldrich) was added (containing additionally 0.01% (v/v)DMSO). Inhibition of TRPA1 activation was achieved by adding10 �M ruthenium red (RR) 1 min prior to the kinetic record-ing. Control HEK293 cells lacking TRPA1 (HEK wt and HEK FT)were treated identically. The recorded time courses were nor-malised to the relative impedance at the beginning of the recordingaccording to �|Z|norm = (|Z|cov,t − |Z|o)/(|Z|cov,o − |Z|o) where |Z|cov,t,|Z|cov,o and |Z|o were the time course of the impedance magnitude
from the cell-covered electrode, the impedance magnitude of thecell-covered electrode at the beginning and of the cell-free elec-trode after trypsin treatment, respectively. All calculations wereperformed with self-written software routines in Matlab 7 (Math-Works) or Origin 7.5 (OriginLab).
O. Pänke et al. / Biosensors and Bioelectronics 26 (2011) 2376–2382 2379
F Pt) eleo Relat
2
Osttes
3
3
fnmdfaeaBantwaqticbfvba
mAarA
ig. 2. (A) Impedance magnitude spectra from indium tin oxide (ITO) and platinum (f 10 �M AITC (dashed) as well as non-covered electrodes as reference (dotted). (B)
.4. Statistical analysis
Statistical analysis was carried out using the software packagerigin 7.5 (OriginLab). All values are expressed as mean (± SD,
tandard deviation) of at least ten independent experiments. Addi-ionally, data were analysed by a one-way ANOVA and a Bonferroniest was used for pair wise comparisons between means. Differ-nces between two means with p < 0.001 were considered highlyignificant.
. Results and discussion
.1. Identification of the cell-specific impedance signal
Impedance spectra of cell-covered and non-covered electrodesrom 600 kHz to 100 Hz were recorded. Although impedance mag-itude and phase information were obtained, only impedanceagnitude data that provide the essential information are
iscussed in the following. This limitation has already been success-ully applied in impedimetric assays that analyse cellular behaviourfter stimulating changes in cell morphology and spreading (Wolft al., 2008; Jahnke et al., 2009; Krinke et al., 2009). It has also beendopted in commercial systems (e.g. ECIS Z system from Applied-ioPhysics). From the relative impedance �|Z|rel, which is defineds the impedance magnitude difference of the cell-covered andon-covered electrode normalised to the impedance magnitude ofhe non-covered electrode (see Section 2), the cell-specific signalas derived. Representative impedance spectra are shown in Fig. 2
nd revealed that cell-specific impedance was observed at a fre-uency of approximately 90 kHz as indicated by the maximum ofhe relative impedance �|Z|rel,max. The detection frequency wasndependent of the applied MEA, electrode material or HEK293ell line. The average value from different cell seeding and MEA ofoth types was 91 ± 26 kHz for MEA 9W6E (n = 15) and 87 ± 18 kHzor MEA 4W15E (n = 10). This detection frequency matches pre-ious findings where human neuroblastoma SH-SY5Y cells haveeen cultured on ITO electrodes showing cell-specific detection atfrequency of 100 kHz (Jahnke et al., 2009).
The maximum of the relative impedance �|Z|rel,max is a general
easure of cell coverage and cell layer sealing on the electrodes.fter microscopic selection of microelectrode arrays covered withdensely packed cell layer, the observed �|Z|rel,max showed aather high spreading, independent of the applied MEA system.ssuming irregularities within the cell layer morphology that were
ctrodes covered with non-treated HEK TRPA1 cells (solid) and 10 min after additionive impedance calculated from the data in A.
microscopically not always observable but probably causing a con-siderable reduction of �|Z|rel,max, only experiments showing valuesof at least 0.7 have been considered for statistical analysis and aver-aging. However, the range of �|Z|rel,max was identical for cell layersfrom TRPA1-transfected cells (HEK TRPA1) and non-transfectedHEK FT cells indicating equal cell adherence for both cell types. Incontrast, the range of �|Z|rel,max was different for both MEA sys-tems analysed. Maximum values of �|Z|rel,max = 2.8 were routinelyfound for the ITO electrodes. The application of the Ayanda MEA4W15E with platinum electrodes resulted in an higher upper limitfor �|Z|rel,max of 4.5. However, the measured values were more thanqualified for further analyses. The observed difference might bedue to differences in the applied electrode material, isolation resistor surface morphology. It is worthy to note that cell layers fromstandard wild type cells (HEK wt) grown on MEA 4W15E showedlower values (≤2.5) throughout, which is indicative for a reducedcell adherence of the standard HEK wild type cells.
Comparison of impedance spectra of non-covered electrodesand electrodes covered only with the adhesion supporting filmrevealed that collagen contributed only to a minor extent to theimpedance of the cell-covered electrode. The relative impedanceof the collagen-covered electrode was ≤0.2 which was negligi-ble compared to the relative impedance of cell-covered electrodes(≤8%).
3.2. Impedimetric monitoring of specific activation and inhibitionof TRPA1
For detecting the influence of TRPA1-specific activators,impedance spectra of cell-covered electrodes before and 10 minafter addition of 10 �M allyl isothiocyanate (AITC) were recorded(Fig. 2). The concentration had been shown to be sufficient to acti-vate the TRPA1 channel (Nagata et al., 2005). The AITC-inducedopening of TRPA1 causes a transmembrane current of cationswhich lowers the membrane resistance of the cells within theadherent cell layer. This should result in a decrease of the mea-sured impedance which was in fact observed in the experiments(Fig. 2B). Although capacitance changes after TRPA1 activationwere not likely, the stimulation of TRPA1 could be accompa-
nied by structural changes of the cell membrane that might havealso induced small changes of the membrane capacitance. More-over, although morphological changes were not observed by phasecontrast microscopy, it cannot be completely excluded that the spe-cific TRPA1-mediated decrease of the membrane resistance also
2380 O. Pänke et al. / Biosensors and Bioelectronics 26 (2011) 2376–2382
Fig. 3. (A) Normalised impedance magnitude recorded at a constant frequency of 100 kHz. Added compounds, used cell cultures and MEA chips are indicated. Activator AITCand control compound DMSO were added 2.6 min after start of recording as indicated by the dashed line. If applied, inhibitor RR was added immediately before start of ther chan1 ordin(
ihd
cfa2ipoptodbtTf
riqwaiFmtrodtsRsar
ecording. Curve set I and II indicate two different behaviours of the activated TRPA1.5 min after compound addition (see arrow in A) from at least 10 independent recv/v)). The significance level was ***p < 0.001.
nduced small changes of the cell layer structure, which mightave caused an additional contribution to the observed impedanceecrease due to a slightly increased inter cell leak current.
Not every AITC addition resulted in an observed reduction of theell-specific impedance. This unexpected behaviour was observedor both MEA systems and can be explained by a process knowns desensitization of TRPA1 (Akopian et al., 2007; Karashima et al.,008). Pharmacological desensitization, also known as ‘run down’,
s a decline in activity with time although the activating agent is stillresent. It is a fundamental mechanism for regulating the activityf neuronal systems. As the TRPA1 channel plays a key role in therocessing of noxious information it undergoes functional desensi-ization by unknown mechanisms. If the process of desensitizationccurred faster than the recording of two consecutive spectra, theetection of the AITC-stimulated impedance decrease could haveeen missed since the membrane resistance had already recoveredo its original value. This necessitated a continuous recording ofRPA1 channel activity by a kinetic protocol as presented in theollowing.
For real-time monitoring of TRPA1 activity which is a pre-equisite for screening applications or analysis of TRP-specificnhibitors, the cell-specific impedance was recorded at a single fre-uency of 100 kHz over time, where the relative impedance �|Z|relas maximal. The marginal difference between the determined
verage of 90 kHz (see above) and the chosen 100 kHz resultedn a deviation of �|Z|rel that was below 1.5% throughout (seeig. 2B). This was negligible and cancelled by the additional nor-alisation procedure of the kinetic data (see Section 2). For high
emporal resolutions, only one electrode per well was impedimet-ically analysed. The appropriate electrode was chosen in respectf high cell layer sealing and high homogenous cell coverage aserived from the impedance spectra and supported by phase con-rast microscopy. The transparent ITO electrodes of MEA 9W6Eubstantially helped to evaluate the cell layers microscopically.
epresentative time courses that reflect average behaviour arehown in Fig. 3A. The data were additionally normalized to the rel-tive impedance �|Z|rel at the beginning of the kinetic impedanceecording (see Section 2). Again, the addition of AITC to cell lay-nels (see text for details). (B) Mean values of the normalised impedance magnitudegs. Concentrations applied were 10 �M AITC, 10 �M RR and 1.4 mM DMSO (0.01%
ers expressing TRPA1 ion channels caused an impedance decreaseindicating conductive ion flow mediated by TRPA1. The stimula-tion of TRPA1-mediated conductance showed a time lag of about30 s. This means that the activator AITC had to diffuse into thedensely packed cell layer before activation of TRPA1 proceeded.This is in contrast to patch clamp experiments where single cellsor even single ion channels are analysed and infusion of AITCimmediately stimulates channel activity. However, for impedimet-ric analysis a maximum impedance decrease was observed 1.5 minafter compound addition and remained constant for two-thirds ofthe analysed electrodes at least for the time window under investi-gation (Fig. 3A, set I). In about one-third of the cases, the impedancesignal recovered to its original value within the next 5 min (Fig. 3A,set II). This behaviour was observed for both MEA systems and canbe explained by desensitization as mentioned above. In those cases,the impedance decrease caused by the stimulated ion conductanceof TRPA1 was cancelled by desensitization that followed. The reasonfor the irregular behaviour that impedance recovery was observedonly in some cases is unknown, but might depend on the cells’live cycle and cell layer growth. However, the kinetic assay thatis based on the detection of the initial AITC-stimulated impedancedecrease, which has been observed in all cell cultures expressingTRPA1, is perfectly suited to test chemical compounds for the abilityto activate or inhibit TRP ion channel conductance within screeningapplications.
For demonstrating that the presented activity assay is also qual-ified to screen for potential TRP inhibitors, 10 �M of ruthenium red(RR) was added to the culture medium 1 min prior to the impedancerecording. In this case, nearly no impedance decrease was observedafter AITC addition as shown in Fig. 3A, demonstrating that thepresence of the TRP blocking agent RR prevented the stimulationof TRPA1 ion conductance by AITC.
In control experiments where only 0.01% DMSO was addedto cell layers consisting of TRPA1-transfected HEK293 cells, no
impedance change was observed. The applied concentrationmatched the DMSO impurity from the AITC stock solution and didnot affect cell layer sealing. However, DMSO was critical since 1%addition yielded an impedance decrease close to the level of a non-
Bioele
caclsDtcilftpatp
4
liioiamcacweahosirnpicaf2tdcenaatrftwtp
A
w(h
O. Pänke et al. / Biosensors and
overed electrode, even though morphological changes had notlways been observed microscopically. Additionally, no impedancehange was observed when 10 �M of AITC was added to cell layersacking TRPA1. These controls prove that AITC induced a TRPA1-pecific signal in TRPA1-transfected HEK293 cells and that neitherMSO nor mechanical stress during compound addition influenced
he impedance signal. Fig. 3B shows mean values of the impedancehange 1.5 min after compound addition proving that the initialmpedance decrease after AITC stimuli is highly significant. Theowered value of the normalised impedance magnitude was equalor both MEA under investigation and were 0.65 ± 0.21 (n = 26) forhe ITO electrodes of MEA 9W6E and 0.72 ± 0.14 (n = 10) for thelatinum electrodes of MEA 4W15E indicating that both materialsre qualified for impedimetric ion channel analysis. However, theransparent ITO electrodes simplified the validation of the appro-riate cell layers by optical microscopy.
. Conclusions
We have demonstrated that TRP channel-specific gating iniving cells can be monitored in vitro by means of real-timempedance spectroscopy. For that purpose, HEK293 cells express-ng membrane-embedded TRPA1 ion channels have been culturedn novel microelectrode arrays. Upon addition of allyl isoth-ocyanate (AITC), a TRPA1-specific activator, impedance of thedherent cell layer decreased significantly indicating channel-ediated ion conductance. The initial impedance change was
ompleted within minutes whereas a cell-mediated response ton external stimulus inducing morphological changes of the unitedell structure occurs typically within hours. The impedance signalas independent of the analysed MEA system indicating that both
lectrode materials, indium tin oxide (ITO) and platinum, allow reli-ble TRP ion channel analysis. However, electrodes made of ITOave the advantage of being transparent, which is necessary forptical control. The induced impedance decrease can be utilised forearching potential inhibitors of members of the TRP channels fam-ly by applying automated cell-based impedance spectroscopy thateports the functional status of a single group of molecular chan-el proteins within the cell membrane. We demonstrated that theresence of ruthenium red (RR), a known blocking agent for TRPA1,
nhibited the AITC-induced stimulation of conductance. Especiallyompounds that block members of TRP ion channels specificallynd with low working concentrations are required and wantedor symptomatic and therapeutical disease control (Okuhara et al.,007; Mathie, 2010). Only a few inhibitors are known so far. Fur-hermore, the impedimetric real-time recording overcomes theetection problems of two-point recordings typically used in bio-hemical assays where desensitization of TRP channels (Akopiant al., 2007; Karashima et al., 2008) possibly masks the activity sig-al of interest. The presented technique has also the potential tonalyse natively expressed ion channels in living cells or withinrtificial membrane systems where Ca2+ imaging and patch clampechniques fail to work, e.g. when the ion channels are incorpo-ated in small liposomes and vesicles. Our findings provide the basisor developing novel impedimetric real-time screening systems foresting pharmaceutical compounds that target diseases connectedith ion channels from the TRP channel family. Upscaling from
he presented 9-well layout to a complete 96-well plate format isrojected.
cknowledgements
HEK293 cells, stably transfected with human TRPA1 cDNA,ere kindly provided by Dr. K. Hill and Prof. M. Schaefer
Universität Leipzig). We would like to thank Dr. K. Hill forelpful discussion. Financial support by the Deutsche Forschungs-
ctronics 26 (2011) 2376–2382 2381
gemeinschaft DFG (SFB610, project Z05 Rob) is also kindlyacknowledged.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.bios.2010.10.015.
References
Akopian, A.N., Ruparel, N.B., Jeske, N.A., Hargreaves, K.M., 2007. J. Physiol. Lond. 583,175–193.
Arndt, S., Seebach, J., Psathaki, K., Galla, H.J., Wegener, J., 2004. Biosens. Bioelectron.19, 583–594.
Asphahani, F., Zhang, M., 2007. Analyst 132, 835–841.Asphahani, F., Thein, M., Veiseh, O., Edmondson, D., Kosai, R., Veiseh, M., Xu, J., Zhang,
M.Q., 2008. Biosens. Bioelectron. 23, 1307–1313.Bautista, D.M., Jordt, S.E., Nikai, T., Tsuruda, P.R., Read, A.J., Poblete, J., Yamoah, E.N.,
Basbaum, A.I., Julius, D., 2006. Cell 124, 1269–1282.Campbell, C.E., Laane, M.M., Haugarvoll, E., Giaever, I., 2007. Biosens. Bioelectron.
23, 536–542.Curran, M.E., 1998. Curr. Opin. Biotechnol. 9, 565–572.Giaever, I., Keese, C.R., 1984. Proc. Natl. Acad. Sci. U.S.A. 81, 3761–3764.Giaever, I., Keese, C.R., 1991. Proc. Natl. Acad. Sci. U.S.A. 88, 7896–7900.Giaever, I., Keese, C.R., 1993. Nature 366, 591–592.Han, A., Frazier, A.B., 2006. Lab Chip 6, 1412–1414.Hill, K., Schaefer, M., 2007. J. Biol. Chem. 282, 7145–7153.Hill, K., Schaefer, M., 2009. Cell Calcium 45, 155–164.Hinman, A., Chuang, H.H., Bautista, D.M., Julius, D., 2006. Proc. Natl. Acad. Sci. U.S.A.
103, 19564–19568.Inoue, R., Jensen, L.J., Shi, J., Morita, H., Nishida, M., Honda, A., Ito, Y., 2006. Circ. Res.
99, 119–131.Jahnke, H.G., Rothermel, A., Sternberger, I., Mack, T.G.A., Kurz, R.G., Pänke, O., Strig-
gow, F., Robitzki, A.A., 2009. Lab Chip 9, 1422–1428.Jegla, T.J., Zmasek, C.M., Batalov, S., Nayak, S.K., 2009. Comb. Chem. High Throughput
Screen 12, 2–23.Jordt, S.E., Bautista, D.M., Chuang, H.H., Mckemy, D.D., Zygmunt, P.M., Hogestatt, E.D.,
Meng, I.D., Julius, D., 2004. Nature 427, 260–265.Karashima, Y., Prenen, J., Meseguer, V., Owsianik, G., Voets, T., Nilius, B., 2008.
Pflugers Arch. 457, 77–89.Keese, C.R., Giaever, I., 1994. IEEE Eng. Med. Biol. Mag. 13, 402–408.Keese, C.R., Karra, N., Dillon, B., Goldberg, A.M., Giaever, I., 1998. Toxicology 11,
183–192.Kiselyov, K., Soyombo, A., Muallem, S., 2007. J. Physiol. Lond. 578, 641–653.Kloss, D., Kurz, R., Jahnke, H.G., Fischer, M., Rothermel, A., Anderegg, U., Simon, J.C.,
Robitzki, A.A., 2008. Biosens. Bioelectron. 23, 1473–1480.Ko, K.S.C., Lo, C.M., Ferrier, J., Hannam, P., Tamura, M., McBride, B.C., Ellen, R.P., 1998.
J. Microbiol. Methods 34, 125–132.Krinke, D., Jahnke, H.G., Pänke, O., Robitzki, A.A., 2009. Biosens. Bioelectron. 24,
2798–2803.Kwan, K.Y., Allchorne, A.J., Vollrath, M.A., Christensen, A.P., Zhang, D.S., Woolf, C.J.,
Corey, D.P., 2006. Neuron 50, 277–289.Lo, C.M., Keese, C.R., Giaever, I., 1993. Exp. Cell Res. 204, 102–109.Macpherson, L.J., Dubin, A.E., Evans, M.J., Marr, F., Schultz, P.G., Cravatt, B.F., Pat-
apoutian, A., 2007. Nature 445, 541–545.Mathie, A., 2010. J. Pharm. Pharmacol. 62, 1089–1095.Mccoy, M.H., Wang, E., 2005. J. Virol. Methods 130, 157–161.Mitra, P., Keese, C.R., Giaever, I., 1991. Biotechniques 11, 504–510.Nagata, K., Duggan, A., Kumar, G., Garcia-Anoveros, J., 2005. J. Neurosci. 25,
4052–4061.Nassini, R., Materazzi, S., De Siena, G., De Cesaris, F., Geppetti, P., 2010. Curr. Opin.
Investig. Drugs 11, 535–542.Niikura, M., Maeda, A., Ikegami, T., Saijo, M., Kurane, I., Morikawa, S., 2004. Arch.
Virol. 149, 1279–1292.Nilius, B., 2007. Biochim. Biophys. Acta-Mol. Basis Dis. 1772, 805–812.Okuhara, D.Y., Hsia, A.Y., Xie, M., 2007. Expert Opin. Ther. Targets 11, 391–401.Opp, D., Wafula, B., Lim, J., Huang, E., Lo, J.C., Lo, C.M., 2009. Biosens. Bioelectron. 24,
2625–2629.Patapoutian, A., Tate, S., Woolf, C.J., 2009. Nat. Rev. Drug Disc. 8, 55–68.Pedersen, S.F., Owsianik, G., Nilius, B., 2005. Cell Calcium 38, 233–252.Rahman, A.R.A., Lo, C.M., Bhansali, S., 2006. Sens. Actuators B: Chem. 118, 115–120.Rahman, A.R.A., Register, J., Vuppala, G., Bhansali, S., 2008. Physiol. Meas. 29,
S227–S239.Ramsey, I.S., Delling, M., Clapham, D.E., 2006. Annu. Rev. Physiol. 68, 619–647.Sadik, O.A., Aluoch, A.O., Zhou, A.L., 2009. Biosens. Bioelectron. 24, 2749–2765.Story, G.M., Peier, A.M., Reeve, A.J., Eid, S.R., Mosbacher, J., Hricik, T.R., Earley, T.J.,
Hergarden, A.C., Andersson, D.A., Hwang, S.W., McIntyre, P., Jegla, T., Bevan, S.,
Patapoutian, A., 2003. Cell 112, 819–829.Tiruppathi, C., Malik, A.B., Delvecchio, P.J., Keese, C.R., Giaever, I., 1992. Proc. Natl.Acad. Sci. U.S.A. 89, 7919–7923.
Venkatachalam, K., Montell, C., 2007. Annu. Rev. Biochem. 76, 387–417.Watanabe, H., Murakami, M., Ohba, T., Takahashi, Y., Ito, H., 2008. Pharmacol. Ther.
118, 337–351.
2 Bioele
W
WW
382 O. Pänke et al. / Biosensors and
egener, J., Keese, C.R., Giaever, I., 2000. Exp. Cell Res. 259, 158–166.
issenbach, U., Niemeyer, B.A., Flockerzi, V., 2004. Biol. Cell 96, 47–54.olf, P., Rothermel, A., Beck-Sickinger, A.G., Robitzki, A.A., 2008. Biosens. Bioelec-
tron. 24, 253–259.
ctronics 26 (2011) 2376–2382
Xing, J.Z., Zhu, L.J., Jackson, J.A., Gabos, S., Sun, X.J., Wang, X.B., Xu, X., 2005. Chem.Res. Toxicol. 18, 154–161.
Yamamoto, S., Wajima, T., Hara, Y., Nishida, M., Mori, Y., 2007. Biochim. Biophys.Acta-Mol. Basis Dis. 1772, 958–967.
Yang, L., Bashir, R., 2008. Biotechnol. Adv. 26, 135–150.
