Mc

Pa

Db

a

1

cbttn2rapra(Ga

0d

Biosensors and Bioelectronics 24 (2008) 253–259

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journa l homepage: www.e lsev ier .com/ locate /b ios

icroelectrode chip based real time monitoring of vital MCF-7 mammaarcinoma cells by impedance spectroscopy

eter Wolfa, Andree Rothermela, Annette G. Beck-Sickingerb, Andrea A. Robitzkia,∗

University of Leipzig, Centre for Biotechnology and Biomedicine, Faculty of Biosciences, Pharmacy, and Psychology, Institute of Biochemistry,ivision of Molecular Biological-Biochemical Processing Technology, Deutscher Platz 5, 04103 Leipzig, Germany

University of Leipzig, Faculty of Biosciences, Pharmacy, and Psychology, Institute of Biochemistry, Division of Biochemistrynd Bioorganic Chemistry, Bruderstr. 34, 04103 Leipzig, Germany

nce spvasiveeoplaressiorese

adioand Gteinscell lyshed atancedetecnt forprein

hed o

a r t i c l e i n f o

Article history:Received 14 January 2008Received in revised form 3 March 2008Accepted 28 March 2008Available online 6 April 2008

Keywords:MonitoringBiosensorNeuropeptide YImpedance spectroscopyMicroelectrode array

a b s t r a c t

Sensorchip based impedaon living cells in a non-inSince the discovery that nthe receptor subtype expcome to the fore of cancertions like fluorescent and rmonophosphate (cAMP) ato either label related proor the need of producingthese problems we establithe complex electric resisrange. Cell alterations arecyclase-stimulating reageeffect can be inhibited byeffect of hNPY can be was

. Introduction

Recently, it has been shown that 85% of primary human breastarcinomas and 100% of the affected lymph nodes are characterisedy an increased expression of human neuropeptide Y (hNPY) recep-or subtype Y1, the main responding receptor of neuropeptide Y. Inhis context, the expression of Y2, that is normally high in non-eoplastic breast tissue, is dramatically decreased (Reubi et al.,001). Based on this shifted expression pattern the role of the hNPY-eceptor system in the development of breast cancer becomes morend more important. hNPY, the 36 amino acid member of theancreatic polypeptide family which acts over at least 5 differenthodopsin-like receptor subtypes (Y1, Y2, Y4, Y5, Y6), can inhibitdenylyl cyclase (AC)-activity via inhibitory G protein-activationLarhammar et al., 1992; Herzog et al., 1992; Lundell et al., 1995;erald et al., 1996; Mullins et al., 2000). In this group Y1, Y2 and Y5re the main responding receptors of hNPY. But in contrast to Y1

∗ Corresponding author. Tel.: +49 341 97 31 241; fax: +49 341 97 31 249.E-mail address: andrea.robitzki@bbz.uni-leipzig.de (A.A. Robitzki).

956-5663/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.bios.2008.03.040

ectroscopy can detect inhibitory effects of human neuropeptide Y (hNPY)labelling free way in real time without the need of supporting reagents.smatic transformations in breast cancer are correlated with a change ofn of hNPY in the affected tissue, the hNPY receptor–ligand system has

arch. Today there are different methods detecting hNPY receptor interac-ctive labelling or detecting hNPY-pathway activation like cyclic adenosine

protein-coupled receptor (GPCR)-assays. For all these assays it is necessarywith additional substances, which can affect the nature state of the cell,sate which allows only a snapshot of the investigated cells. To overcomenew method to detect hNPY–receptor interactions. Therefore, we monitor(impedance) of cells attached to a microelectrode over a wide frequencyted as changes in the impedance spectra. After application of the adenylylskolin, impedance is decreased at 5 kHz frequency within minutes. Thiscubating the cells with hNPY for a time range of 20 min. The inhibitory

ut and the same cells can be stimulated by forskolin again.© 2008 Elsevier B.V. All rights reserved.

and Y2, a possible change in expression during breast cancer devel-opment is not described for Y5. The inhibitory effect of Y-receptorson the intracellular cAMP level can influence major downstreampathways of cAMP-like protein kinase A activation (Walsh & vanPatten, 1994) or extracellular-signal related kinase (ERK) inhibi-tion (Nakamura et al., 1995) by cAMP. Furthermore an activationof intracellular calcium mobilisation is described depending on thecell type specific pathways (Herzog et al., 1992). Today there aredifferent molecular biological methods detecting hNPY receptorinteractions like fluorescent and radioactive labelling (Walsh et al.,1993) or detecting hNPY-pathway activation like cAMP (Herzog etal., 1992) and GPCR-assays (Dautzenberg, 2005). For all these assaysit is necessary to either mark related proteins with additional sub-stances, which can affect the nature state of the cell, or the needof producing cell lysate which allows only a snapshot of the inves-tigated cells. To overcome these problems we established a newmethod to detect hNPY–receptor interactions.

Sensorchip-based impedance spectroscopy is a combination ofmultielectrode arrays (MEA) with a highly sensitive impedanceanalyser and allows the detection of cellular effects on living cellsin a non-invasive, destruction-free way, without the need of assay

Bioele

254 P. Wolf et al. / Biosensors and

supporting substances. Thereby, the complex electrical resistance,so-called impedance (Z), of cells attached to a recording electrodeis measured at 10 mV voltage and alternating current over a widefrequency range (102–106 Hz). Cells in the electric circuit repre-sent a complex network of electrical resistance which is influencedby cellular elements like ion channels and lipid layers (Robitzki& Rothermel, 2006). The data is processed by a computer-aidedcontrol unit in real time. Cells can be cultured and manipulateddirectly on the MEA electrodes, which were originally developedto detect the electrical activity of neurons and exclusively buildof non-cytotoxic, biocompatible materials (Stett et al., 2003; Grosset al., 1977). Agent induced cell alterations are shown as a changein the impedimetric spectrum. As we are interested in the furtherdevelopment of the detection of agent-mediated cell status vari-ances by real time impedance measuring systems, we investigatedif the forskolin mediated stimulation and the hNPY mediated inhi-bition of the eukaryotic AC in MCF-7 mamma carcinoma cells isdetectable by this method in real time. MCF-7 cells express themain hNPY-receptors Y1 an Y5.

2. Materials and methods

2.1. Reverse transcription-PCR

For RT-PCR complete RNA of MCF-7 mamma carcinomacells cultured in a 35 mm cell culture dish until subconfluentstate was isolated, RNA concentration was determent by lightabsorption at 260 nm wave length in a UV–vis-spectrometerand 1 �g RNA was transcripted into cDNA using the PromegaReverse-Transcription-System. cDNA was screened of hNPY-receptor transcripts using specific primers. cDNA of humanneuroblastoma cell line SH-SY5Y was used for Y2 receptormRNA positive control. PCR was performed with parametersof 94 ◦C for 1 min, 53 ◦C for 1 min and 72 ◦C for 1 min for 35cycles. Used Primers for PCR were 5′-GCCCTTGGCCATGATATTTA-3′

and 5′-AAGGCAAAGAAGAAGCCACA-3′ for Y1, 5′-GATCATCCCG-GACTTTGAGA-3′ and 5′-TGCTGTTCATCCAGCCATAG-3′ for Y2,and 5′-CTGGCAGCCATAAATGGAGT-3′ and 5′-ACATCATGCCCAA-CAAATGA-3′ for Y5.

2.2. Multielectrode array on chip

Multielectrode array based cell culture sensorchips (Fig. 1A)

were designed with 60 planar round microelectrodes arranged in an8×8 array (Fig. 1B) on a glass substrate. Each electrode consists ofan Iridium surface (Fig. 1C), provided by an isolated platinum con-ductor, and has a diameter of 30 �m with a clearance of 200 �mamong each other. The 5000 times larger counter electrode islocated at the edge of the culture area. Based on this proportion,the measuring electrode with the adherent cells is the dominantfactor in the electric circuit. An affixed ring of polysulfone forms aculture chamber with about 1 ml capacity which allows cultivatingof cells directly on the MEA. During the measurement the sensor-chip is connected to an MEA adapter and placed into a standardCO2-incubator to assure optimal culture conditions.

2.3. Culturing of the mamma carcinoma cell line MCF-7 on chip

To culture cells directly on a sensorchip, MEAs were laminatedby incubating with 5 �g/ml laminin in PBS for 2 h at 37 ◦C. Lamininimproves the adhesion of cells to the electrodes surface and there-fore avoids leakage current (Rothermel et al., 2005; Slaughter etal., 2004). 1×105 of harvested MCF-7 cells were resuspendedin 1 ml standard culture medium [DMEM (Gibco Invitrogen, UK),

ctronics 24 (2008) 253–259

1% penicillin/streptomycin (Invitrogen Corporation, USA), 1% glu-tamax (Invitrogen, Germany), 10% fetal calf serum (FCS; GibcoInvitrogen, UK)], transferred to the chip culture chamber and keptunder standard culture conditions (37 ◦C, 5% CO2, 95% air).

For repeated use, MEAs were cleared from cells by 30 min trypticdigestion and sonicated for 5 min in 2% ultrasonol solution. Afterrinsing with distilled water, MEAs were disinfected in 70% ethanolfor 30 min and finally washed with sterile PBS. Instead of sterilisingby ethanol they can be autoclaved alternatively.

2.4. Impedance measurement

For impedance spectroscopy the impedance analyzer Solartron1260A, the biomedical interface Solartron 1294 (both fromSolartron Analytical, Hamshire, UK), and the MEA adapter board(self-modified electronic board originally purchased from MultiChannel System, Reutlingen, Germany) were used. The adapterboard allows connecting every electrode to the biomedical inter-face (Fig. 1). For impedance analysis of MCF-7 cells, an alternatingcurrent of 10 mV was applied to the electrodes. The impedancebetween the detection electrode and the counter electrode wasmeasured at 10 points per decade over a frequency of 100 Hz to1 MHz. For experiments applying agents, approximately 16 h beforemeasurement, the culture medium was replaced by 990 �l freshculture medium (DMEM, 1% glutamax, 1% penicillin/streptomycin,10% FCS). The next day, impedance of confluent MCF-7 cells wasmonitored at least for 1 h to control stable impedance before drugapplication. Thereafter, different agents were added to culturesand impedance was measured every minute for different periods,according to experimental requirements. Scanning of one completefrequency spectrum took about 49 s. For data acquisition and dataprocessing the impedance analysis software SMaRT 2.3 (SolartronAnalytical) was used. To point out alterations after drug applicationthe measured impedance |Z|was normalised and expressed as rel-ative impedance |Z|rel where |Z|b is the impedance value before and|Z|a after the stimulation: |Z|rel (%) = ((|Z|a− |Z|b)/|Z|b)×100.

2.5. Serum-stimulated ERK activation studies

For ERK-activation studies, 105 cells per well of MCF-7 cells wereseeded out on silicone chamber subdivided microscope slides andcultured to subconfluence under standard conditions. 24 h beforeserum stimulation, standard growth medium was changed toserum free medium [DMEM, 1% penicillin/streptomycin, 1% gluta-

max, 0.1% bovine serum albumin (BSA; Sigma–Aldrich, Germany)].Before serum stimulation (20% FCS without medium change, 4 min,according to Capiati et al., 2004) cells were treated either with 5 �Mforskolin (Sigma–Aldrich, Germany) for 10 min or 20 min with 1 �MhNPY (Sigma–Aldrich, Germany) followed by 10 min coincubationwith 5 �M forskolin. Control cells were only serum stimulated.After stimulation, cells were washed twice with cold PBS and fixedwith 4% formaldehyde for 15 min on ice. After formaldehyde treat-ment cells were washed 2 times with PBS and one time withdistilled water to prevent salt crystallisation. The dried slides werefrozen at −20 ◦C until immunocytochemical staining.

2.6. Immunocytochemistry

For immunochemistry staining fixed cells on microscopic slideswere treated 90 min with 5% (w/v) BSA in PBST [0.1% (v/v) TritonX-100 (Merck, Germany) in PBS, pH 7.4] to block unspecific proteinbinding sides. Primary staining was done by p(Tyr 204)-ERK (SantaCruz Biotechnology, USA) mouse monoclonal antibody (4 �g/ml)diluted in TBST [0.1% (v/v) Tween-20 (Carl Roth GmbH, Germany) inTris-buffered saline] for 90 min at room temperature and secondary

P. Wolf et al. / Biosensors and Bioelectronics 24 (2008) 253–259 255

de array with 60 iridium electrodes (B and C) is fixed into a MEA adapter board whichO2 cell incubator (D). The analysis software SMaRT is triggering the sample of a frequencyed according to Rothermel and Nieber et al., 2006).

7) indicates that Y2 specific primers are functioning correctly.Non-template negative controls (Fig. 2, Lanes 4, 6, 9) show nounspecific PCR products. All PCR products show the expected frag-ment length with 413 bp for Y1, 413 bp for Y2, and 435 bp forY5.

3.2. The location of phosphorylated ERK in MCF-7 mammacarcinoma cells can be influenced by forskolin and hNPYapplication

Fig. 1. Measuring setup for impedance spectroscopy: a chip based (A) multielectroallows the connection of every electrode to the dielectric interface into a standard Crange from 102 to 106 Hz and the capture of frequency depend voltage drop (modifi

staining with Cy2-conjugated goat-anti-mouse (Dianova, Germany,1 h at room temperature, 7.5 �g/ml in TBST). After primary and sec-ondary staining, slides were washed two times with TBST for 5 min.Nuclear staining was done by Sytox orange (Invitrogen, Germany)for 30 s. Staining procedure was completed by washing slides indistilled water to prevent salt crystallisation.

2.7. Fluorescent microscopy

Immunocytochemically stained cells were documented using

a Nikon Eclipse TE2000 inverted microscope with an EC-1 laser-scanning unit (Nikon, Dusseldorf, Germany). Photographs wereprocessed with EZ-C1 3.4 Software (Nikon) and arranged by AdobePhotoshop® 7.0.

3. Results

3.1. Expression of hNPY-receptors Y1, Y2, and Y5 in MCF-7

To clear, if the mamma carcinoma cell line MCF-7, which rep-resents the biological sensory part in our measuring system, issuitable for our kind of detection, we investigated if the main hNPYreceptors Y1, Y2, and Y5 are expressed in the cell system. This wasdone by RT-PCR with specific primer. For Y2 positive control, cDNAof human neuroblastoma cell line SH-SY5Y was used.

MCF-7 cells show product bands for hNPY-receptors Y1 andY5 (Fig. 2, Lane 3 and 8) which indicates for particular mRNAtranscripts. Expression of Y2 mRNA (Fig. 2, Lane 5) is lackingor under detection limit, which conforms to the fact that hNPY-receptor expression shifts from Y2 to Y1 in neoplasmatic breasttissue (Reubi et al., 2001). The Y2 positive control (Fig. 2, Lane

To clear if our cell system is sensitive to extracellular hNPY andforskolin, we investigated the extracellular signal-related kinaseactivation of serum stimulated MCF-7 cells after application of

Fig. 2. Expression of hNPY-receptors in MCF-7 cells: cDNA of complete mRNAof MCF-7 mamma carcinoma cells was tested checked for transcripts of thehNPY-receptors Y1, Y2 and Y5 by PCR with specific primers. PCR products were elec-trophoretically separated in a 2% agarose gel. Product bands for Y1 and Y5 (Lanes3 and 8) show the appearance of the particular mRNA-transcripts in MCF-7 cells.mRNA transcript for Y2 (Lane 5) is lacking. Positive (pos) control for Y2 specificprimer (Lane 7) indicates the correct function. Additional non-template controls(NTC) in Lanes 4, 6, 9 show no unspecific products. Labelled marker band (←)represents a size of 500 base pairs.

256 P. Wolf et al. / Biosensors and Bioelectronics 24 (2008) 253–259

in and after application of forskolin and hNYP preincubation: cells were cultured untilh before serum stimulation. Before serum stimulation (20% FKS, 4 min) cells were treated

owed by 10 min coincubation with 5 �M forskolin (G–I). Control cells were only serumnucleus (A and C). This effect can be suppressed by forskolin-induced cAMP accumulation

via inhibitory G-proteins, leads to an increased nuclei-location of phosphorylated ERKmonoclonal antibody which detects Tyr-204 phosphorylated p44 and p42 MAP kinase.

nal antibody. Cell nuclei were stained with Sytox orange (B, E and H).

Fig. 3. Location of phosphorylated ERK in MCF-7 cells after application of forskolsubconfluence and standard growth medium was changed to serum free medium 24either with 5 �M forskolin (FSK) for 10 min (D–F) or 20 min with 1 �M hNPY follstimulated (A–C). After serum stimulation phosphorylated ERK is located in the cell(D and F). Preincubation with hNPY, which can in turn inhibit cAMP accumulation(G and I). Phosphorylated ERK was stained with Santa Cruz p-ERK (Tyr 204) mouseSecondary antibody staining was done by Cy2-conjugated goat-anti-mouse polyclo

either forskolin or hNPY and forskolin. Activation of ERK can beinhibited by intracellular cAMP (Nakamura et al., 1995). hNPY hasalready been described as inhibitor of forskolin-stimulated cAMPaccumulation in different cell types including MCF-7 cells (Zhu etal., 1992; Amlal et al., 2006).

For ERK activation studies, MCF-7 mamma carcinoma cells were

cultured to a subconfluent state and standard culture mediumwas changed to serum free medium 24 h before serum stimula-tion. Before stimulation (20% FKS, 4 min) cells were treated eitherwith 5 �M forskolin (Fig. 3 D–F) for 10 min or 20 min with 1 �MhNPY followed by 10 min coincubation with 5 �M forskolin (Fig. 3G–I). For control experiment cells were treated 10 min with 0.05%DMSO (vehicle) only (Fig. 3A–C). After serum stimulation cellswere fixed and stained by immunocytochemistry for phospho-rylated ERK (Tyr-204). To show if ERK is activated, cells wereanalysed for the cytoplasmatic and nuclear distribution of phos-phorylated ERK in the cell. Active ERK is translocated to the cellnucleus (Chen et al., 1992; Lenormand et al., 1993; Paul et al.,2003). Control cells showed a clear concentration of phospho-rylated ERK in the nucleus (Fig. 3A–C). Forskolin treated cellsshowed no accumulation of phosphorylated ERK in the nucleus,the only signal was evenly spread in the cytosol (Fig. 3D–F).Pretreatment with hNPY leads to an increased concentration ofphosphorylated ERK in the nucleus with only a slight signalin the cytosol (Fig. 3G–I). The difference between cytosolic andnuclear signal was clear, but not as strong as in control cells(Fig. 3A–C).

Fig. 4. Alteration of an impedance spectrum after application of 5 �M forskolin:time course is shown from 10 min before to 60 min after application. The applicationof 5 �M forskolin on MCF-7 cells leads to a steady decrease in the impedimetricspectrum within 15 min at a range between 103 and 105 Hz with a variation peakat 5 kHz. The peak frequency is the most sensitive point to measure impedancealteration after forskolin stimulation. For a better overview not for every minute aspectrum is shown.

Bioele

P. Wolf et al. / Biosensors and

3.3. Impedance measurement

105 cells of MCF-7 mamma carcinoma cell line were seededout in 1 ml standard culture medium in MEA culture chamber undkept under standard culture conditions (37 ◦C, 5% CO2, 95% air). Cellgrowth was observed with inverse cell culture microscope until thecells reached the desired confluence on the measuring electrodes.After determining of the measuring electrode for the present anal-ysis cell culture medium was changed and cells were transferredto the measurement station in the incubator (standard cell cultureconditions). The next day the chosen electrodes were measuredwith the given setup (Fig. 1D) over a frequency range from 102 to106 Hz at standard culture conditions. The spectra show the rela-tive impedance between the counter electrode and the cell coveredmeasuring electrode (Fig. 4). Before drug application impedance ismeasured over several hours to make sure that basal impedancespectrum is stable (variations <5%). Thereafter, different agentswere added, medium was carefully mixed, and impedimetric spec-tra were measured dependent on the experimental setup. Due tothe circumstance that forskolin induced effects in the impedimet-ric spectrum can be most sensitively measured at 5 kHz frequency(Fig. 4), the following results are rated at this frequency.

3.4. Forskolin application leads to a significant decrease at 5 kHz

The variations of the impedimetric spectrum were measured10 min before to 60 min after application of DMSO-diluted forskolin(final conc. 5 �M). Related to the spectrum before application, theimpedance is decreasing within 15 min with a peak about 5 kHz(Figs. 4 and 5A). At that frequency the impedance is decreasingsteadily with an average minimum delta Z nearly−15% after 20 min(Fig. 5A).

3.5. hNPY can inhibit forskolin induced impedance effects

To investigate the inhibitory effects of hNPY on forskolin inducedcAMP accumulation hNPY (final conc. 1 �M) was added 20 minbefore application of 5 �M forskolin. Impedance was measured5 min before hNPY application, during 20 min hNPY preincubationand up to 60 min after application of forskolin. After forskolin appli-cation the impedance at 5 kHz frequency is decreasing to an averageminimum delta Z around −5% after 50 min (Fig. 5B).

3.6. hNPY inhibited impedance effects can be washed out

To show that cells in which the forskolin induced impedancedrop was inhibited by hNPY were originally sensitive to forskolinstimulation, the complete chip medium was replaced by stan-dard medium and the cells were cultivated for at least 3 h. Then5 �M forskolin were added to the cells, medium was mixed andimpedance was measured for 60 min. The same electrode whichwas used for the previous inhibitory experiment was measured.After forskolin application the impedance at 5 kHz frequency isdecreasing to an average minimum of the relative impedancearound −15% after 30 min (Fig. 5C).

3.7. Application of an equal volume DMSO shows no significanteffect

To control that there are no impedance effects caused by vehicleinfluence an equal volume of DMSO (0.05%) was added to the cells,medium was mixed and impedance was measured for 60 min afterapplication. After DMSO application the impedance at 5 kHz fre-quency is slightly increasing but shows no significant change overthe measuring period (Fig. 5D).

ctronics 24 (2008) 253–259 257

4. Discussion

The aim of this study was to detect hNPY-mediated effectson vital MCF-7 mamma carcinoma cells in a labelling free, non-invasive way by real-time impedance spectroscopy. Here weutilized a sensorchip based measuring setup in which MCF-7mamma carcinoma cells, adherently attached to measuring elec-trodes, were investigated by impedance spectroscopy during drugapplication. First we had to clarify if main hNPY-receptors areexpressed in the MCF-7 cell line which was used as biological partof the sensor system. This cell line was originally established frommetastatic lobular mammary carcinoma. Utilizing RT-PCR with spe-cific primers for human hNPY-receptor Y1, Y2, and Y5 we coulddemonstrate the occurrence of mRNA transcripts of receptor Y1 andY5. Expression of hNPY-receptor Y2 was lacking. This expressionpattern is corresponding with former conclusions that breast can-cer tissue shows an increased expression of hNPY-receptor Y1 whileexpression of receptor subtype Y2, which is highly expressed innormal breast tissue, is dramatically decreased (Reubi et al., 2001).

Next goal was to show that hNPY can really influence intra-cellular pathways in our cell system. Therefore, we showed theinhibitory effect of forskolin on serum-stimulated ERK-activation inour cell system and achieved to inhibit this effect in turn by applica-tion of hNPY. Serum-stimulated MCF-7 show a clear translocation ofphosphorylated ERK to the nucleus. After preincubating with 5 �Mforskolin for 10 min, serum-stimulation fails to locate phosphory-lated ERK in the MCF-7 nucleus. Application of 1 �M hNPY beforeforskolin addition leads again to translocation of phosphorylatedERK in the cell nucleus. These results denote that extracellular sig-nals by forskolin and hNPY can influence the intracellular cAMPlevel and major downstream pathways like ERK inhibition in ourcell system, which is an important condition for the bioelectricalsensor system.

For impedimetric analysis MCF-7 cells were cultured on thesensorchip until they reach a confluent state and then measuredbefore and after application of 5 �M forskolin into the media.This application leads to a decreased impedance within minuteswith a minimum peak at 5 kHz frequency indicating that forskolin-mediated impedance alterations can be detected most sensitive atthis frequency. So, for further experiments impedance data wasappraised and compared at that frequency. Statistical data pro-cessing indicates that the impedance drop was significant from5 min after forskolin application. Preincubation with hNPY beforeforskolin stimulation can inhibit the forskolin-induced impedance

drop at 5 kHz frequency which correlates to the inhibitory effectof hNPY on forskolin-stimulated intracellular cAMP accumulation(Amlal et al., 2006). The inhibitory effect of hNPY can be washed outby changing application medium to fresh standard culture mediumand measuring impedance drop at same electrode as in inhibitoryexperiment 3 h later. A delay of measurement is necessary to stabi-lize cells and the associated impedance signal to change of completemedium, which is indeed a change in ionic and ingredient compo-sition. After stabilization of the impedimetric signal, application offorskolin is leading again to a significant impedance drop at 5 kHzfrequency. There are different studies measuring impedance onadherent cells after application of cAMP and forskolin, respectively(Reddy et al., 1998), or stimulating of GPCRs (Ciambrone et al., 2004;Yu et al., 2006). Thereby, changes in the cellular shape are oftenidentified as major impedance changing effect. However, carriedout staining of the tight junction protein occludin as well as micro-scopic monitoring of MCF-7 cells with and without the applicationof forskolin do not indicate remarkable changes in the cellular mor-phology (data not shown). Nevertheless, an impedance drop of 15%can be the result of minor morphological changes, which cannot beclearly detected by light microscopy.

258 P. Wolf et al. / Biosensors and Bioelectronics 24 (2008) 253–259

Fig. 5. Alteration of impedance at 5 kHz after drug application on MCF-7 mamma carcindecrease at 5 kHz within 15 min. Time course is shown from 10 min before to 60 min afterafter 20 min preincubation with 1 �M hNPY and application of 5 �M forskolin. Time coursto 60 min after application of 5 �M forskolin on MCF-7 cells. The impedance shows a sligh5 kHz after 3 h washout of experiment (B) and 5 �M forskolin-stimulation measured on tafter application of 5 �M forskolin on MCF-7 cells. The impedance is decreasing to ≤−10effect on impedance at 5 kHz from 10 min before to 60 min after adding 0.05% DMSO (smedium without cells. Here, cell free controls demonstrate that there is no effect of theapplication.

5. Conclusions

The present study shows, that impedance monitoring of MCF-7 mamma carcinoma cells on a multielectrode sensorchip is afunctional method to detect the stimulating effect of the diter-pene forskolin and the inhibitory effect of hNPY on intracellularcAMP-accumulation. This is done in real-time without additionalutilization of any assay supporting substances. At present themethod is limited by detecting a general activation of hNPY associ-ated receptors. Therefore, future work is focussed on the screeningof receptor specific ligands to distinguish the effects of several

oma cells: (A) after application of 5 �M forskolin the impedance shows a steadyapplication of 5 �M forskolin on MCF-7 cells. (B) Alteration of impedance at 5 kHz

e is shown from 24 min before application of 1 �M hNPY over 20 min preincubationt decrease at 5 kHz within 45 min after application. (C) Alteration of impedance at

he same electrode. Time course is shown from 10 min before application to 60 min% within 20 min after application. (D) Control measurements show no significant

olvent) on MCF-7 cells, 5 �M forskolin to medium without cells or 1 �M hNPY toadded agents on the electrolyte impedance. n = 3; *P < 0.05 compared to point of

receptor subtypes. The method has the potential to become a highthroughput system for screening for hNPY-sensitive cells whichis in particular interesting against the background of main hNPY-receptor Y1 as assumed tumour marker in breast cancer diagnostic(Reubi et al., 2001).

Acknowledgement

This work was supported by the Deutsche Forschungsge-meinschaft (Grant SFB 610, project Z5, A.A. Robitzki, A1 A.G.Beck-Sickinger).

P. Wolf et al. / Biosensors and Bioele

References

Amlal, H., Faroqui, S., Balasubramaniam, A., Sheriff, S., 2006. Cancer Res. 66,3706–3714.

Capiati, D.A., Rossi, A.M., Picotto, G., Benassati, S., Boland, R., 2004. J. Cell Biochem.93, 384–397.

Chen, R.H., Sarnecki, C., Blenis, J., 1992. Mol. Cell Biol. 12, 915–927.Ciambrone, G.J., Liu, V.F., Lin, D.C., McGuinness, R.P., Leung, G.K., Pitchford, S., 2004.

J. Biomol. Screen. 9, 467–480.Dautzenberg, F.M., 2005. Biochem. Pharmacol. 62, 1493–1499.Gerald, C., Walker, M.W., Criscione, L., Gustafson, E.L., Batzl-Hartmann, C., Smith,

K.E., Vaysse, P., Durkin, M.M., Laz, T.M., Linemeyer, D.L., Schaffhauser, A.O., White-bread, S., Hofbauer, K.G., Taber, R.I., Branchek, T.A., Weinshank, R.L., 1996. Nature382, 168–171.

Gross, G.W., Rieske, W., Kreutzberg, G.W., Meyer, A.A., 1977. Neurosci. Lett. 6,101–105.

Herzog, H., Hort, Y.J., Ball, H.J., Hayes, G., Shine, J., Selbie, L.A., 1992. Proc. Natl. Acad.Sci. U.S.A. 89, 5794–5798.

Larhammar, D., Blomqvist, A.G., Yee, F., Jazin, E., Yoo, H., Wahlested, C., 1992. J. Biol.Chem. 267, 10935–10938.

Lenormand, P., Sardet, C., Pages, G., L’Allemain, G., Brunet, A., Pouyssegur, J., 1993. J.Cell Biol. 122, 1079–1088.

Lundell, I., Blomqvist, A.G., Berglund, M.M., Schober, D.A., Johnson, D., Statnick, M.A.,Gadski, R.A., Gehlert, D.R., Larhammar, D., 1995. J. Biol. Chem. 270, 29123–29128.

ctronics 24 (2008) 253–259 259

Mullins, D.E., Guzzi, M., Xia, L., Parker, E.M., 2000. Eur. J. Pharmacol. 395, 87–93.

Nakamura, M., Sakanaka, C., Aoki, Y., Ogasawara, H., Tsuji, T., Kodama, H., Matsumoto,T., Shimizu, T., Noma, M., 1995. J. Biol. Chem. 270, 30102–30110.

Paul, S., Nairn, A.C., Wang, P., Lambrose, P.J., 2003. Nat. Neurosci. 6, 34–42.Reddy, L., Whang, H., Keese, C.R., Giaever, I., Smith, T.J., 1998. Exp. Cell Res. 245,

360–367.Reubi, J.C., Gugger, M., Waser, B., Schaer, J.C., 2001. Cancer Res. 61, 4636–

4641.Robitzki, A.A., Rothermel, A., 2006. In: Marx, U., Sanding, V. (Eds.), Drug Testing In

Vitro. Wiley, Weinheim, pp. 79–101.Rothermel, A., Kurz, R., Ruffer, M., Weigel, W., Jahnke, H.G., Sedello, A.K., Stepan,

H., Faber, R., Schulze-Forster, K., Robitzki, A.A., 2005. Cell Physiol. Biochem. 16,51–58.

Slaughter, G.E., Bieberich, E., Wnek, G.E., Wynne, K.J., Guiseppi-Elie, A., 2004. Lang-muir 20, 7189–7200.

Stett, A., Egert, U., Guenther, E., Hofmann, F., Meyer, T., Nisch, W., Haemmerle, H.,2003. Anal. Bioanal. Chem. 377, 486–495.

Walsh, D., Wharton, J., Blake, D., Polak, J., 1993. Br. J. Pharmacol. 108, 304–311.Walsh, D.A., van Patten, S.M., 1994. FASEB J. 8, 1227–1236.Yu, N., Atienza, J.M., Bernard, J., Blanc, S., Zhu, J., Wang, X., Xu, X., Abassi, Y.A., 2006.

Anal. Chem. 78, 35–43.Zhu, J., Li, W., Toews, M.L., Hexum, T.D., 1992. J. Pharmacol. Exp. Ther. 263, 1479–

1486.