Label free sensing platform for amyloid fibrils effect on living cells

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

http://www.elsevier.com/authorsrights

Author's personal copy

Label free sensing platform for amyloid fibrils effect on living cells

Mihaela Gheorghiu a, Sorin David a, Cristina Polonschii a, Andreea Olaru a,Szilveszter Gaspar a, Ovidiu Bajenaru b,c, Bogdan O. Popescu d,e, Eugen Gheorghiu a,n

a International Centre of Biodynamics, 1 B Intrarea Portocalelor, 060101 Bucharest 6, Romaniab Department of Neurology, Neurosurgery and Psychiatry, University of Medicine and Pharmacy “Carol Davila”, Bucharest, Romaniac Clinical Department of Neurology, University Emergency Hospital, 169 Splaiul Independentei, RO-050098, Bucharest 5, Romaniad Laboratory of Molecular Medicine, ‘Victor Babeş’ National Institute of Pathology, 99-101 Spl. Independentei, 050096 Bucharest 5, Romaniae Department of Neurology, Colentina Clinical Hospital – CDPC, School of Medicine, “Carol Davila” University of Medicine and Pharmacy, 19-21 Sos.Stefan cel Mare, 020125 Bucharest 2, Romania

a r t i c l e i n f o

Article history:Received 22 July 2013Accepted 17 August 2013Available online 26 August 2013

Keywords:Surface plasmon resonanceElectrical impedance spectroscopyLabel-free multiparametric real-timemonitoringAmyloid beta fibrilsCellular dynamicsCellular platform

a b s t r a c t

This study presents a multiparametric label-free analysis gathering surface plasmon resonance (SPR) andelectrical impedance spectroscopy (EIS) for monitoring the progress of a model epithelial cell culture(Madin Darbey Canine Kidney – MDCK) exposed to a peptide with high bio-medical relevance, amyloid β(Aβ42). The approach surpasses the limitations in using the SPR angle for analyzing confluent cellmonolayers and proposes a novel quantitative analysis of the SPR dip combined with advanced EIS as atool for dynamic cell assessment.

Long, up to 48 h time series of EIS and SPR data reveal a biphasic cellular response upon Aβ42exposure corresponding to changes in cell-substrate adherence, cell–cell tightening or cytoskeletalremodeling.

The equivalent circuit used for fitting the EIS spectra provided substantiation of SPR analysis on theprogress of cell adhesion as well as insight on dynamics of cell–cell junction.

Complementary endpoint assays: western blot analysis and atomic force microscopy experimentshave been performed for validation.

The proposed label free sensing of nonlethal effect of model amyloid protein at cellular level providesenhanced resolution on cell-surface and cell–cell interactions modulated by membrane related proteinapparatus, applicable as well to other adherent cell types and amyloid compounds.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Multi-parametric label free cellular platforms hold substantialpotential in dynamic assessment of subtle, nonlethal effects atcellular level, in a substantial step forward against endpointanalyses. An application area of major analytical and bio-medicalrelevance concerns evaluation of the direct interaction of specificcompounds with different cell compartments, as an avenue toexplore cellular processes in conjunction with toxicity effects.

One such example is provided by amyloid β (Aβ) peptides,compounds considered pivotal in Alzheimer's disease (AD) patho-genesis that showed a complex repertoire of effects at variouslevels within cellular models. Depending on the concentration,molecular structure and fibrilar state, Aβ peptides have beenproved to exert various toxic cellular effects via disturbance ofthe structure and function of cell membranes (Williams andSerpell, 2011; Capone et al., 2009; Verdier et al., 2004), stress

fiber formation, disruption and aggregation of actin filaments andcellular gap formation (Nagababu et al., 2009), increased trans-epithelial transport of large molecules (Deli et al., 2010; Gonzalez-Velasquez et al., 2008; Marco and Skaper, 2006) in conjunctionwith altered tight junctions (TJ) protein expression. This complexrepertoire of, sometimes contradictory effects has been assessedusing mostly end-point, individual analyses on various cell typesor barrier models, avoiding standardization.

In this context, we report on a multiparametric biosensingplatform combining Surface Plasmon Resonance (SPR) and elec-trical impedance spectroscopy (EIS) label-free assays to revealcellular changes within a fully developed cell monolayer uponexposure to amyloid β fibrils. While the combination of these twotechniques is not new (Terrettaz et al., 1993), this is, to the best ofour knowledge, the first study to correlate the cell-surfacedynamics assessed by SPR with electrical cellular and intercellularparameters provided by multifrequency EIS analysis for fullydeveloped cell layers.

SPR matured into a powerful, well established analytic tool inbiosensing, whereas SPR exploitation in biological cell monitoringhas been only recently acknowledged. Reported applications range

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/bios

Biosensors and Bioelectronics

0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.bios.2013.08.028

n Corresponding author. Tel.: þ402 131 043 54; fax: þ402 131 043 61.E-mail address: [email protected] (E. Gheorghiu).

Biosensors and Bioelectronics 52 (2014) 89–97


from SPR-based measurements of volume changes in adherentcells (Vala et al., 2012), cell activation (Cuerrier et al., 2008) tovirus mimetic interactions (Chabot et al., 2009). The measure-ments are label-free, noninvasive and provide a relevant timeresolution cell dynamics and rely on sensitive detection of changesin refractive index of cell monolayer due to cell-surface contactsand protein/lipid changes. Related optical biosensor technologies(Fang et al., 2007) are used to record integrated cellular responsesrather than individual components of signaling pathways via labelfree monitoring of the redistribution of intracellular constituentstriggered upon receptor activation (Schroder et al., 2010) as well ascell–cell and cell-surface interactions (Yashunsky et al., 2010).

SPR imaging approaches on whole cells (Yanase et al., 2010;Yanase et al., 2012) showed exquisite sensitivity of SPR assay onchanges associated with plasma membrane, cell spreading, proteintranslocation and membrane potential, further confirmed throughplasmon based impedance microscopy technique, pioneered byTao group (Foley et al., 2008; Wang et al., 2011).

Complementarily, electrical impedance spectroscopy, a wellestablished technique in cell assessment, provides noninvasive,real time monitoring capabilities of the electrical and morpholo-gical parameters of cell monolayers. Key contributors to theimpedance measurements are changes in cell-substrate adher-ence, changes in cell shape and volume (Gheorghiu, 1996) andchanges in cell–cell interactions (Gheorghiu et al., 2002; Sanduet al., 2010). These factors individually or collectively affect theflow of extracellular and trans-cellular current, influencing themagnitude and characteristics of the signal measured. As such,Electrode Cell substrate Impedance Sensing (ECIS) (Giaever andKeese, 1991), a validated marker for barrier integrity in real time(Jepson, 2003), enables label free monitoring of the interactionsbetween cells and the substrate, study of cell properties such asattachment, spreading, motility, growth and proliferation(Asphahani et al., 2008; Ghenim et al., 2010; Giaever and Keese,1991; Han et al., 2007; Hong et al., 2011) and cellular state (Arndtet al., 2004; Gheorghiu et al., 1999). The method enables real-timeanalyses to monitor cellular responses to chemical, physical, andbiological stimuli (Hong et al., 2011). Interestingly, ECIS typeassays, based on cellular index, have also been used to followcellular response to G-protein coupled receptors activation in realtime, revealing compound (Abassi et al., 2009), cell line (Petersand Scott, 2009) and signaling cascade (Kammermann et al., 2011)characteristic impedance-based time-dependent cell responseprofiles.

While a certain degree of convergence is evident for SPR andEIS data in particular for cell-surface interaction, as both techni-ques are sensitive to surface coverage, the combination of the twotechniques holds potential in providing additional insight intocellular processes at various levels within cell structure: SPRregarding intracellular refractive index shifts, whereas EIS willprovide data on cell–cell interactions with special focus on cellbarrier properties. As such this combined platform is particularlyapplicable for assessment of the effect of target compounds withcomplex biological interactions at various levels within cell struc-ture (such as Aβ) to yield insight into the fibril associated effect ona model cell culture.

As model compound, we choose the Aβ1–42 fragment, one ofthe predominant form of Aβ found in brains of AD patients, basedon its well established cytotoxic profile (Klein et al., 2004) anddocumented self-association propensity (Jarrett et al., 1993). Inter-action between this amyloid protein and cellular membranes, aswell as the associated cellular effect is assumed to be importantboth in disease onset and propagation.

MDCK (Madin Darbey Canine Kidney cells), provide a wellcharacterized model of polarized epithelial cells, a suitable androbust cell model for investigation of Aβ effect on both “naive” and

specialized (e.g. brain) barriers, easy to grow and highly relevantfor elucidating whether Aβ related alterations are univocallyachieved in the in situ microenvironment of the Blood BrainBarrier (BBB) or can be exerted in an externalized environment aswell. MDCK cells were applied to the study of polarization ofamyloid precursor peptide trafficking and sorting, as models forsimilar events that go on in neuronal cells (Haass et al., 1994,1995); (Capell et al., 2002) and as in vitro model for the transportof Aβ across the BBB, (Nazer et al., 2008). Being of renal origin,MDCK cells provide an added bonus given the documented:(i) greater renal impairment for AD patients than for cognitivelynormal controls (Kerr et al., 2009), (ii) the influence on amyloidhomeostasis of reduced renal clearance of peripheral Aβ, as well as(iii) the recently drawn epithelial parallels in neuronal adhesioncontrol (Famulski and Solecki, 2012).

The proposed label free sensing of nonlethal effect of modelamyloid protein at cellular level enables enhanced resolution oncell-surface and cell–cell interactions modulated by membranerelated protein apparatus, applicable as well to other adherent celltypes. To this purpose, the following advancements have beenpursued:

– combined EIS/SPR platform for real time, noininvasive assess-ment of the multiparametric changes of fully developed cellmonolayers exposed to Aβ fibrils;

– monitoring biological effects at the level of cell-substrateadherence, cell interior and cell–cell junctions, using a novelSPR dip analysis and multichannel, multifrequency EIS analysisof cell monolayers based on a realistic equivalent circuit;

– cross validation of cell substrate interaction dynamics providedby SPR and EIS data.

Endpoint Western Blot (WB) and Atomic Force Microscopy(AFM) analyses complete, with molecular and high resolutionmorphological snapshots, the multi-parametric insight of the Aβfibrils effect on cell monolayers.

2. Materials and methods

2.1. Chemicals

2.1.1. Aβ42 fibril formationSynthetic Aβ42 (Sigma, A9810) was reconstituted in cell media

as previously reported (Cedazo-Minguez et al. 2003) to form(proto)fibrils. Briefly, peptides were suspended in serum free cellmedium (Dulbecco's modified Eagle's medium, DMEM pH 7.4) at aconcentration of 10 μM and incubated at 37 1C, 5% CO2 for 48 h,with occasional stirring. Cell treatment, was performed on cells at90% confluence, at a concentration of 5 μM.

2.1.1.1. Cell cultures. MDCK I cells (ECACC, cat. no. 00062106), weregrown on in DMEM/F12 medium (Invitrogen) supplemented with10% fetal calf serum, 2 mM L-glutamine, 100 U/mL penicillin, and100 μg/mL streptomycin at 37 1C in a 5% CO2 humidified incubator.Passages were performed at 80% confluence and cells were seededat 105 cells/cm2 on the measurement chips. Environmental controlis ensured during subsequent cellular assays within Sanyo MCO-20AIC CO2 Cell Culture Incubator.

2.1.2. Biomimetic lipid membranesIt was shown that zwitterionic supported phospholipid bilayers

selectively bind Aβ aggregates but not the monomeric form(Kotarek and Moss, 2010). To check for the effective Aβ concentra-tion that would potentially bind to the cell membrane and alter

M. Gheorghiu et al. / Biosensors and Bioelectronics 52 (2014) 89–9790


the lipid structure, we used the zwitterionic phospholipid 1-pal-mitoil-2-oleyl-sn-glycero-3-phosphocholine (POPC) (Avanti Lipids,Alabaster, USA) in conjunction with commercial L1 SPR chips(GE Healthcare Life Sciences, Biacore Division, Uppsala, Sweden)to form supported lipid membrane mimics of the cellular counter-parts (Gheorghiu et al., 2009; Olaru et al., 2009).

2.1.3. Other test compoundsTriton X 100, a zwitterionic detergent (Sigma-Aldrich) is chosen

as test compound mimicking possible nonspecific membranedestabilization and acute cellular responses.

2.2. Methods

2.2.1. Surface plasmon resonance, SPR chips, data measurement andprocessing for cell analysis

For whole cell analysis, gold coated BK7 glass chips (2 nmchromium adherence layer, 50 nm gold evaporated by Physical VaporDeposition, Kurt J. Lesker) are used in conjunction with refractiveindex matching oil (518F, Zeiss, Germany) and polydimethylsiloxane(PDMS, Sylgard) cell cultivation chambers. 105 cells/cm2 MDCK cellsare seeded in the PDMS growth chambers, cured on top of the goldcoated chips, and cultivated for 2 days prior to SPR analysis. Theensemble is affixed to the SPR detector chip via a refractive indexmatching oil film and monitored for extended intervals, with asampling rate of 1 Hz.

For monitoring cell response to Aβ42 we used a home builtevaluation kit based on TSPR2K23 sensing components developedby Texas Instruments with optimized readout and data analysis.This SPR approach is based on a Kretschmann configuration inwhich individual surface elements within the sensing channel areilluminated with an infrared light-emitting diode (LED), withλ¼834 nm, at different angles and analyzed via an array ofphotodetectors (schematics of the set-up is presented in Fig. 1A).The LED emits a diverging beam that passes through a polarizerand strikes the sensor surface, within a range of angles above thecritical one. The angle at which light is incident upon this surfacevaries with the location on the surface (Chinowsky et al., 2003)and accordingly, each photodetector (pixel) corresponds to aspecific incidence angle.

The system developed in house enables that the whole reflec-tivity spectrum (reflectivity values corresponding to individualincidence angles as pixels intensities) is recorded at each timepoint and analyzed by a dedicated LabView routine, providing realtime access to both the position of the SPR dip minimum and thewhole reflectivity spectrum.

SPR monitoring of the interaction of Aβ42 fibrils with the POPCmembranes during long injections, up to 60 min, is achieved in aBiacores 3000 instrument (GE Life Sciences, Biacore Division).Detergent free 10 mM HEPES running buffer is used in conjunctionwith specific concentrations of aged Aβ42 in DMEM cell growthmedium. The L1 SPR chip (GE Healthcare Life Sciences AB, Uppsala,Sweden) allows for capturing of liposomes or subcellular prepara-tions via lipophilic groups partially adorning the dextran matrix.The protocol for lipid matrix formation, (Gheorghiu et al., 2009),renders stable lipid membranes that are further used to recordduring long injection times the interaction of Aβ42 fibrils with thebiomembrane mimics.

2.2.2. Impedance spectroscopy analysis2.2.2.1. Impedance chips. For parallel control and test experiments,we used a measuring chamber whose base consists of a glass coverslip (22 mm�22 mm�0.17 mm). Thin film technology was usedto pattern four sets of Au working and counter electrodes foronchip impedance spectroscopy onto such cover slips. The circularcounter electrodes surround the 1 mm diameter disk workingelectrodes for a uniform distribution of the electric field. Twocomponent poly(dimethylsiloxane) (PDMS, Sylgard 184 from DowCorning Corporation, USA) and an appropriate mold were used tofabricate two wells of 250 mL each, Fig. 1B, as previously reported(Gaspar et al., 2012). Cells were seeded into each compartment ofthe measuring chamber and left to grow into a confluent layer(2 days).

2.2.2.2. Impedance data measurement and analysis. A 4294A Pre-cision Impedance Analyzer from Agilent Technologies Ltd., Japan,interfaced with in house multiplexing module was used for recordingthe four individual electrode sets. An AC signal of 50 mV pp amplitude,zero DC bias, within 40 Hz–100 kHz frequency range (100 frequencypoints with logarithmic distribution) was chosen. Multifrequencyimpedance monitoring, as opposed to single frequency assays is aprerequisite for discerning between the various processes involved incell response and enable data deconvolution by modeling. Upontreatment, the culture medium was removed from both com-partments and replaced with fresh, serum free medium (control)and 5 mM aged Aβ42 in serum free medium (Aβ42), respectively.

Cell growth (attachment, spreading on the electrode surface,formation of cell–cell junctions) leads to a corresponding changein the geometry of the current pathways and, thus shapes therecorded impedance. The interplay between resistive and capaci-tive pathways makes the current flow from the electrode into thebulk medium frequency dependent. Real and imaginary compo-nents of the complex impedance at each individual frequency are

Fig. 1. (A) Experimental set-up for SPR monitoring of living cells adhered to the sensing surface and (B) experimental set-up for 4 channels impedance analysis of living cellsand proposed equivalent circuit.

M. Gheorghiu et al. / Biosensors and Bioelectronics 52 (2014) 89–97 91


analyzed based on an equivalent electrical circuit to derive thedielectric parameters of the cell monolayer.

Coping with the whole measured frequency domain (40 Hz–100 kHz), the experimental data are fitted (Mathematica, WolframResearch Europe Ltd, U.K) with equivalent electrical circuits, thatassociate distinct functional entities in the biological probe (e.g.cell monolayer, para-cellular space, cell-surface and cell-to-cellcontacts) with abstract representations (e.g. constant phase ele-ments – CPE, resistors and capacitors).

One can derive highly detailed equivalent circuits for impe-dance data. Nevertheless, such a circuit has to be optimized forincreased robustness and rendering capability to fit large timeseries of impedance data. Therefore, we considered the circuitdepicted in Fig. 1B, comprising a resistor for the bulk media (Rb),a capacitor (C1) in series with a resistor (R1) for the cell–cell gapregion and a constant phase element (CPE) for the cells withincellular layer. Note that this constant phase element incorporatesas well the electrical contribution of cell-electrode interfacebeneath the cells. The impedance of CPE is characterized by2 independent parameters: CPE_T and the exponent CPE_P,according to the following equation:

ZCPE ¼1

CPE_T iωð ÞCPE_Pð1Þ

The phase angle of this element is independent of frequency andhas a value of �(90�CPE_P)1. When the value of distributionparameter, CPE_P, equals 1, the CPE behaves as an ideal capacitor (cellmembrane), whereas when 0, the CPE behaves as a pure resistor.When CPE_P equals 0.5, ZCPE corresponds to a series RC circuit, oftentermed Warburg impedance. The frequency dependence of themagnitude of this element has been ascribed to a wide variety ofelectrode effects such as surface roughness and heterogeneity, nonuni-form current distribution, and distribution of relaxation times(Macdonalds, 1987). In our case, the expected effects at cell-electrode interface do not allow clear separation between electrodeimpedance and the impedance of the adjacent heterogeneous celllayer, thus the common element CPEcell has been used.

Given the fact that cell monolayer is fully developed, data analysisis best performed based on an equivalent electrical circuit and not onindividual frequencies values. Nevertheless, the details of analysisbased on a modified cell index VN (with V being either of the compleximpedance components e.g. Real or Imaginary part at individualfrequencies) are given for completion and reference to the existingapproaches (e.g. xCELLIgence), in Supplementary material.

2.2.3. Atomic force microscopyCells grown to confluence on coverslips after 24 and 48 h of

Aβ42 treatment were briefly washed 3 times with Ringer, incu-bated at 37 1C for 15 min in a 2.5% solution of Glutaraldehyde(Sigma-Aldrich), in Ringer, washed 3 times with Ringer andmounted for AFM analysis. A NanoWizzard II instrument (JPK,Germany) was used in contact mode (VEECO SNL probe, 0.06 N/mnominal spring constant), in air, to record the height and lateraldeflection profiles of the cells within the monolayer.

3. Results and discussions

3.1. SPR monitoring of MDCK cells attached on SPR surface

Upon seeding, MDCK I cells attach, spread, grow and interactwith each other to form polarized cell monolayers on top of goldsurfaces.

Cell presence on top of the sensing surface is accompanied bysignificant refractive index changes (the lipid membrane enclosethe cytoplasm with high protein content providing a high refrac-tive index contrast with measurement media) that will be visiblein characteristic reflectivity curves (presenting the dependence ofthe reflectivity on the angle of incidence).

Classical SPR sensorgrams reveal time evolutions of position ofreflectivity dip minimum (sensitive to the overall, equivalentrefractive index) not accounting for the inherent inhomogeneitypertaining to biological cells, or are based on reflectivity values atparticular, fixed angles. In contrast to these approaches, we haverecently demonstrated (Olaru et al., 2013) that quantitative eva-luation of the changes in particular areas in the reflectivity curves(shape of the SPR dip) provides a rapid and effective procedure forsurface quality assessment.

Whole dip analysis is all the more needed when studying fullydeveloped cell monolayers. Biological inhomogeneity (differentcell morphology, variable cell density) adds further complexity tothe classical SPR dip analysis and limits the applicability of SPRsensorgrams based on SPR minimum. Depending on cell growth,surface coverage, cell-substrate adherence and tight junctionformation, cell shape and volume, and cell–cell interactions, thewhole SPR spectrum is changed. The structure of the intracellularspace close to the membrane as well as the variable thickness ofthe cell/substrate distance affects the spatial distribution of therefractive index values sensed by the evanescent field. This isrevealed by the fringes exhibited by the SPR spectra, where eachpixel corresponds to both a different angle of incidence and adistinct region on the sensor surface. Thus, the whole reflectivitycurve provides a wealth of information about molecular changesoccurring at the surface.

Representative dips for cell covered and bare SPR surfaces arepresented in Fig. 2.

Unlike for bulk refractive index changes seen as translationalshifts of the SPR dip, changes of the cell monolayer attached on thesensor surface determine specific patterns within different regionsof the SPR dip.

Noisier and shallower dips correspond to attached cells, with asubsequent shift to larger SPR angles. While this shift has beenrecently reported (Robelek and Wegener, 2010; Vala et al., 2012),the intricate pattern (spikes and split dips) in connection to fullydeveloped cell layers on the surface has not been reported before.

A certain degree of inhomogeneity in the cell layer developedon the surface explains to some extent the occurrence of multiple

Fig. 2. SPR dips with, without cells and upon cell detachment.



individual dips, or noisier spectra. Nevertheless, it is the specificconfiguration of the SPR set-up, where the light emitted from apoint LED source strikes individual areas on the sensor surface, atparticular angles within a range above the critical angle thatallows separation of distinct resonances corresponding to variablevolume effective refractive indexes. In this configuration, indivi-dual pixels record particular incidence angles and the reflectivityvalues are related further on, via transfer matrix approaches, tovolume refractive index changes.

The structure of the intracellular space close to the membraneas well as the variable thickness of the cell/substrate distanceaffects the spatial distribution of the refractive index values sensedby the evanescent field and is revealed by the fringes exhibited bythe SPR spectra. Each pixel corresponds to both a different angle ofincidence and a distinct region on the sensor surface. The interplaybetween the effective refractive indices and the angles of inci-dence determines the shape of the SPR dip. Since the penetrationdepth of the evanescent wave depends on the angle of incidence,integrating the reflectivity over selected angle intervals (i.e.selected pixels) provides quantitative insights on the overall cell/substrate adherence, and cell structure, close to the measurementsurface.

In contrast to a typical sensorgram that displays only theposition of the SPR minimum (a function of equivalent refractiveindex), access to and analysis of time series of SPR dips is able toprovide specific focus on spatially distributed effective refractiveindices and their change in time.

Details on the dependency of the reflectivity changes on thepenetration depth, evaluated based on transfer matrix approach,are included in supplementary materials (Figs. SM1, SM2) for anillustrative case of an uniform dielectric slab shifting from thesurface while a theoretical and experimental study, in the moregeneral context of volume refractive index contrasts at micro-scopic scale, explaining this effect, is in progress.

To confirm that the whole reflectivity curve (as compared tosingle point measurements (Robelek and Wegener, 2010)) iscapable to reveal the nonlinear processes undergone by the fullydeveloped cell monolayer until cell detachment from the surface,we chose Triton X 100 exposure (0.02%), as test compoundmimicking possible nonspecific membrane destabilization andacute cellular responses: cell burst and cell detachment. Fig. 3shows a multiphase process undergone by the fully developed cell

monolayer until complete cell detachment from the surface thatcan be accurately monitored using SPR. Notable, confirming thatindeed dip alterations are related to cell presence on the surface,there are still visible after extended detergent exposure some dipalterations related to possible cell residues or incomplete celldetachment and can be eliminated upon thorough washing stepsor trypsin addition (that provides enzymatic attack of cell-surfaceattachment sites).

This extreme test reveals SPR dip changing patterns: when cellsare detaching from the surface, the central dip corresponding tothe refractive index of the growth medium is gradually revealed,simultaneously with the decrease in the SPR dip region corre-sponding to large angles (insets Fig. 3). Accordingly, the analysis ofdistinct spectral regions is able to reveal dynamics of cell-surfaceinteraction and will be used in the following in the analysis of SPRdata for Aβ42 exposure. This approach is all the more needed as weare interested in analyzing mild Aβ42 effects for which cell deathor complete cell detachment are not expected. Given the largeconcentration span reported in the literature as eliciting biologicaleffects, we first “calibrated” the concentration of our test com-pound using biomimetic lipid membranes.

3.2. Aβ42 fibrils effect on biomimetic lipid membrane

Aiming to confirm membrane attachment and mild lipiddestabilization for a given concentration range of the Aβ42(proto)fibrils at membrane level, we checked the effect anddynamics of interaction with a stable zwitterionic lipid (POPC)matrix formed on L1 chips. It was shown, (Kotarek and Moss,2010), that zwitterionic supported phospholipid bilayers selec-tively bind Aβ aggregates but not the monomeric form and thesebound aggregates are capable of supporting nonsaturable rever-sible growth via monomer addition.

Synthetic Aβ42 is reconstituted in cell media and aged for 48 h.AFM analysis has been deployed to characterize the nature ofaggregated Aβ during the “aging” process since literature reportsemphasize different biological effects depending on the oligomer/fibrilar state. According to Fig. SM3, Supplementary material,where individual fibrils, with characteristic interfibrilar distances(inset) are shown, we confirm that the chosen aging processdetermines Aβ fibrils formation. Fibrillar Aβ42 was shown to elicita concentration dependent cellular effect, with Aβ42 concentra-tions up to 5 mM reported to only slightly affect cell viability (Solitoet al., 2009).

Our results, presented in Supplementary materials confirmedthat at 5 mM concentration, the Aβ42 (proto)fibrils injected onnewly formed, supported lipid membrane loosely interact with thelipid, with a nonlinear binding kinetics (Fig. SM4 SupplementaryMaterial).

Based on similarly long injection times and the same lipidmembrane, oligomers of melittin at similar concentrations showed(Gheorghiu et al., 2009) a stronger propensity to fully destabilizethe membranes upon attachment and insertion. In contrast, foraged Aβ42 fibrils, one can note a rapid (up to 30 min), saturable,mild lipid destabilization, evident starting with the lowest testedconcentration (1.25 mM), suggesting a direct pore formation(Ambroggio et al., 2005; Capone et al., 2009; Lashuel andLansbury, 2006; Mobley et al., 2004). These data confirm that1.25–5 mM concentration range determines a direct Aβ42 (proto)fibrils membrane interaction, yet with a reduced accompanyingcell membrane destabilizing effect. The still mild membranedestabilization achieved for 5 mM concentration of aged Aβ42,combined with the consistent membrane attachment (Fig. SM4Supplementary Material) further supports the use of this concen-tration for in vitro studies. Full characterization of the interaction

Fig. 3. Normalized time evolutions of the SPR angle for two cell cultures exposed totriton X 100 detergent; insets evolutions of the SPR dips corresponding to arrowindicated time windows during Triton X exposure. The SPR angle at the specifictime point of 10 min is subtracted from subsequent time series.



kinetics on supported membranes is out of the scope of thepresent study.

In the following, the 5 mM concentration of aged Aβ42 (proto)fibrils is applied on MDCK cells grown, at sub-confluence, for all(SPR, impedance, AFM) experiments.

3.3. Real time, multiparametric monitoring of Aβ42 interaction withcell layers attached on SPR/electrode surface

Upon Aβ42 exposure of fully developed cell layers on top ofSPR/EIS sensing surfaces, the SPR and impedance spectra change,revealing dynamic alterations of cell–cell and cell-surface compo-nents. When high refractive index structures (cell membrane,organelles) become closer to the surface (e.g cell-surface tighten-ing, protein amassment), an increase in associated SPR values isexpected. Alternatively, when either cell detaches from the surfaceor cell volume increases (accompanied by a dilution of cellcontent), a decrease in SPR values is estimated. Since the SPR dipcomprises the reflectivity sensed at different angles of incidence towhich different penetration depths correspond, analysis of thewhole SPR spectrum is able to reveal the distribution of theeffective refractive index pertaining to cell-surface neighborhood.Therefore, due to the wide refractive index span corresponding tolipid or protein (i.e. 1.45 or 1.67, respectively, (Olaru et al., 2009))cell dynamics upon Aβ42 exposure is revealed by analyzingdifferent incidence angle domains within the SPR spectrum.

As a new way of analyzing the dynamic changes of the SPRspectra, coping with the complexity of cell related SPR dips, timevariations of individual pixel values (Rθ(t)) against the moment ofinjection (Rθ(t0)) are integrated over selected angle intervalswithin the reflectivity spectrum, according to Eq. (2), where θiand θj provide the limits of incidence angle range used in theintegration and correspond to specific detectors (pixels) in thearray.

IR¼Z θj

θi

ðRθðtÞ�Rθðt0ÞÞdθ ð2Þ

In agreement with SPR theory (Schasfoort and Tudos, 2008)and supplementary material Figs. SM1 and SM2, stating thatdepending on the incidence angle used, different penetrationdepths of the evanescent field are achieved, our proposedapproach to analyze the changes of the SPR spectra by integratingthe reflectivity over selected angle intervals (i.e. selected pixels)reveals distinct dynamics for different incidence angles, (see Fig. 4and inset where the plateau obtained in the reflectivity valuescorresponding to 69.51C721 domain – indicated with a dashedline, precedes the peak in reflectivity values corresponding to65.01C691 domain; the evolution in this domain mirrors exactlythe SPR angle evolution as seen in Fig. 4 inset) provides quanti-tative insights on the overall cell/substrate adherence, and cellstructure, close to the measurement surface, enabling a moredetailed scrutiny of potential cell responses.

Given the decrease of the penetration depth towards themargins of the domain of incidence angles used in analysis (Fig.SM2 supplementary materials), the observed dynamics upon Aβ42addition imply interfacial processes (i.e. cell-surface interactions)preceding intracellular protein reorganization and suggest a rea-listic succession of cellular effects, in line with literature reportsand our results on biomimetic lipid membranes (supplementarymaterial, Fig. SM4). In summary, possible Aβ42 effects include:rapid, mild cell membrane destabilization (slight lipid loss, fol-lowed by an overall cell swelling due to pore formation andsubsequent cell–cell tightening); for longer incubation times, cellsurface contacts improve while protein expression changes at bothintracellular level as well as at the level of tight junctions.

Accordingly, the multiphasic dynamics exhibited by cell adhe-sion, evidenced by analyzing time series of SPR dips as in Fig. 4,upon Aβ42 addition can be related to looser cell-surface contacts(decrease of SPR values for the first 1.5 h) and subsequent reattach-ment (increase of the values up to 5 h) due to cell membranedestabilization (inducing cell swelling) followed by processesdetermining better refractive index contrast such as cell–celltightening (more membrane area within the evanescent field), cellthinning (cell content concentration). Protein mobilization towardthe membrane (above 20 h) is suggested by the occurrence of theplateau for small penetration depths while for larger penetrationdepths the increase is unperturbed; the consistent decrease forlonger cultivation times (432 h) can be related to initiation of celldetachment in batch cultures or loss of actin content. Nevertheless,discrimination between these processes cannot be made merelybased on SPR measurements (Vala et al., 2012).

Based on these findings, we stress on procedure's applicability,especially for high cell coverage where SPR minimum can nolonger be derived, to gain insights into cell-substrate dynamicsbased on SPR assays.

Aiming for increased information on cell dynamics, moreoversince due to the morphological features of MDCK cells (height4500 nm), and the limited depth resolution of SPR measurements(o300 nm), the tight junction area is usually non-visible by SPR,we used impedance spectroscopy in a wide frequency range tomonitor evolution of cells exposed to Aβ42. Changes in surfacecoverage, cell-substrate adherence, tight junction formation andcell–cell interactions affect the flow of extracellular and trans-cellular current influencing the magnitude and characteristics ofthe measured signal. In particular, for fully developed cell mono-layers, at low to medium frequencies, the current is mainlyconcentrated in paracellular pathways and is therefore highlyinfluenced by the space between adjacent cells and the barrier-forming cell–cell junctions positioned at the apical side of the cellmonolayer (modeled as capacitor C1 in series with resistor R1).As such, EIS is a powerful complement to SPR assays that is ratherinsensitive to processes taking place at the apical side of the cellmonolayer.

Fig. 4. Evolution of the SPR angle (red) and of the integral (over 69.51C721incidence angle domain) of the reflectivity changes revealed by cell culturesresponse upon Aβ42 exposure. Aβ42 addition is indicated by arrow while theapparent plateau in IR data is indicated through a dashed line. The match betweenthe time evolutions of SPR angle (red) and the integral of the reflectivity changesover the lower, i.e. 66.51C691 incidence angles domain is revealed in the inset. (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this article.)



In contrast with single frequency analyses (Primiceri et al.,2010), the wide frequency range measured with time resolutionbelow 1 min, enables a more precise tuning of the focus level ofanalysis by appropriate selection of frequency points and datadeconvolution by modeling. The interplay between impedancedata at various frequencies is capable to provide dynamic informa-tion on specific cellular pathways likely to be affected during theinteraction between Aβ42 fibrils and MDCK cells, therefore experi-mental data covering the 40 Hz–100 kHz frequency domain, arefitted (Mathematica, Wolfram) with the equivalent electricalcircuit depicted in Fig. 1.

Comparison between control and Aβ42 exposed cell cultures,with focus on time intervals up to 24 h, reveals for impedancedata, distinct dynamics of the cell–substrate interaction underAβ42 exposure and in controlled conditions, Fig. 5, at both the levelof cell surface /cell monolayer (Fig. 5A – the evolution of theparameters connected to ZCPE) as well as at the level of gapjunctions (Fig. 5B – normalized R1 evolution R1N¼(R1(t)�R1(t0))/R1(t0)).

Both control and Aβ42 exposed cell cultures undergo full mediaexchange at the moment indicated by arrow therefore slight

monolayer perturbation independent of Aβ42 effect might benoticed immediately after media exchange, thus the first 10–20 min are not considered in the analysis.

Using appropriate equivalent model, one can easily relatebiological processes with individual parameters dynamics.

Increase of R1 values revealed in Fig. 5B corresponds to anincrease in monolayer tightness, more rapid for Aβ42 exposed cellsin comparison with control ones for the first 5 h. A possible effecton the tight junction protein components can be inferred based onthe evolution of R1 values up to 24 h: barrier tightness is achievedfaster and at higher values for control versus Aβ42 exposed cells.

The first rapid increase in R1 values (up to 5 h), is corroboratedwith an increase in CPEcellT values suggesting looser cell surfaceattachment and increased cell layer homogeneity and possiblecell–cell tightening. A possible interaction mechanism explainingthis behavior might concern membrane perturbation upon Aβ42exposure, nonspecific entry of water from the extracellular med-ium that determines swelling of the cells in the monolayer andconsequently monolayer tightening.

The appropriate match between the evolutions of SPR data andthe amplitude of cell monolayer impedance revealed in Fig. 6 isnotable. As such, impedance values confirmed the SPR data addingfurther details concerning cell junction status. Corroborated withthe recent report, (Robelek and Wegener, 2010) emphasizing thecapabilities of SPR measurements in following cell volumechanges, this match can be regarded as a confirmation of theappropriateness of the proposed equivalent circuit that enablesthe separation between dynamics of cell-substrate, and cell–celljunctions, respectively.

Further supporting the ability of SPR reflectivity dip analysis inselected angle domains to monitor cellular processes undergoneduring Aβ42 exposure at different depths within the cell layer, it isworth noting the match between SPR and EIS data following Aβ42exposure (Fig. SM5 Supplementary material).

Accordingly, EIS SPR analysis shows an initial cell–surfacetightening upon Aβ42 exposure that can be related to membranepermeabilization, as proved by complementary SPR assays onsupported lipid membranes (presented in Fig. SM4 SupplementaryMaterial) and an accompanying increase in cell volume (visible aswell upon early Triton X exposure, Fig. 3), followed by a morepronounced, transitory overall increase of cell monolayer impe-dance and homogeneity.

Importantly, the SPR and impedance measurements wereperformed in the same time and conditions. As such, EIS is avaluable complement to SPR measurements, the combination ofboth techniques enabling cross validation.

Fig. 5. Fitted parameters of the equivalent circuit of the measured impedancerelated to (A) cell layer (CPEcellT) and (B) cell–cell contacts (R1), for control (black)and Aβ42 exposed cell cultures (red). (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.) Fig. 6. Evolution of the amplitude parameter of CPE correlated with SPR dynamics.



The analysis confirms a multipronged effect of Aβ42 exposure atboth membrane/interface as well as of cell interior and cell–cellcontacts, capable to be revealed by the proposed multiparametric,label free cell platform.

Notably, the barrier properties are lower for Aβ42 exposed cellsthan for control (Fig. 5B) suggesting a direct Aβ42 effect on cell–celljunctions in agreement with literature reports relating Aβ42 fibrilsexposure with modified expression of barrier proteins (Deli et al.,2010; Gonzalez-Velasquez et al., 2008; Marco and Skaper, 2006).

As cytoskeletal actin reorganization is often linked with delo-calization or degradation of TJ proteins and barrier propertiesdisruption, the actin content was checked with western blotanalysis. Actin down-regulation (Table 1, supplementary material)in MDCK I cells is notable, suggesting as well cytoskeletal remo-deling in response to Aβ42 exposure.

This effect is further confirmed by AFM analysis, Fig. SM6Supplementary Material, revealing, after 48 h of Aβ42 exposure,collapsed cells (i.e. better exposure of internal organelles due to adecrease in actin content) with ruffled membranes in contrastwith the fuzzy, microvilled appearance after 24 h (see comparativeanalysis in Table 2 supplementary material) and enhanced mem-brane inhomogeneity as function of Aβ exposure duration.

4. Conclusions

Barrier permeability changes accompany a plethora of diseasesincluding Alzheimer's. Aiming for a label free, noninvasive plat-form for dynamic assessment of cell monolayers, as models for cellbarriers, exposed to compounds with medical relevance, wepropose a multi-parametric analytical approach enabling dynamicanalysis of the related, subtle, nonlethal effects at cellular levelinduced by Aβ42 exposure. Improved SPR and impedance assayswere combined to pinpoint the intricate cell response to stimula-tion. We provide not only continuous (as opposed to selected timepoints) monitoring of cell monolayer in control and under Aβ42exposure, but also quantitative interpretation of SPR and impe-dance spectra capable to provide a new perspective on thedynamic processes at various levels within the cellular system.

SPR dip analysis recently reported for chip quality assessment(Olaru et al., 2013) is advanced for living cells monitoring andreveals a previously unreported intricate pattern (spikes and splitdips) in connection to cell attachment and cell-surface interaction.In contrast to a typical sensorgram that displays only the positionof the SPR minimum (a function of equivalent refractive index),this study proposes access to and analysis of time series of thewhole SPR dips, as reflectivity changes integrated over selectedangle intervals, demonstrating specific focus on spatially distrib-uted effective refractive indices and their change in time andhence provides insights on the cell/substrate adherence and cellstructure, close to the measurement surface.

Moreover, we stress on the capability of our approach (analyz-ing the whole SPR dip) to provide quantitative information basedon SPR data evenwhen, due to cell growth, the SPR angle providedby standard SPR assays can no longer be derived.

Multi-frequency Electric Cell-substrate Impedance Sensing cor-relates with and complements the SPR data and enables throughappropriate tuning of the frequency domain and use of anequivalent model, characterization of electrical parameters of intraand inter cellular structures.

Based on the SPR/EIS combination we were able to derivedynamic data on cell - cell tightening, membrane permeabilizationand accompanying increase in cell volume, followed by tightjunction formation. In conjunction with endpoint literature reportsconnecting Aβ42 fibrils with modified expression of barrier proteins(occludin and claudin 2) on a time dependent manner, the

functional characterization of cell layer properties as detected usingSPR dip analysis and multi-frequency Electric Cell-substrate Impe-dance Sensing confirms this evolution and reveal additionally atemporary cell surface attachment tightening, with subsequentchanges in cytosol composition, signaling processes or cytoskeletonremodeling.

Aβ42 fibrils effect revealed by the combined SPR and impe-dance assay has been further confirmed with AFM and westernblot analyses and, using a procedure previously reported(Gheorghiu et al., 2009), on a model lipid membrane as well.

This study is, to the best of our knowledge, the first label freeEIS/SPR multiparametric approach to reveal the dynamic effect ofAβ fibrils on cell layers and provides the basis of a more generallyapplicable label free cellular sensing platform.

Acknowledgments

The financial support of the National Project “Bioscope” PN-II-ID-PCCE-2011-2-0075 Contract no. 11/2012 is acknowledged.

Appendix A. Supporting information

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.bios.2013.08.028.

References

Abassi, Y.A., Xi, B., Zhang, W., Ye, P., Kirstein, S.L., Gaylord, M.R., Feinstein, S.C.,Wang, X., Xu, X., 2009. Chemistry and Biology 16 (7), 712–723.

Ambroggio, E.E., Kim, D.H., Separovic, F., Barrow, C.J., Barnham, K.J., Bagatolli, L.A.,Fidelio, G.D., 2005. Biophysical Journal 88 (4), 2706–2713.

Arndt, S., Seebach, J., Psathaki, K., Galla, H.J., Wegener, J., 2004. Biosensors andBioelectronics 19 (6), 583–594.

Asphahani, F., Thein, M., Veiseh, O., Edmondson, D., Kosai, R., Veiseh, M., Xu, J.,Zhang, M., 2008. Biosensors and Bioelectronics 23 (8), 1307–1313.

Capell, A., Meyn, L., Fluhrer, R., Teplow, D.B., Walter, J., Haass, C., 2002. Journal ofBiological Chemistry 277 (7), 5637–5643.

Capone, R., Quiroz, F.G., Prangkio, P., Saluja, I., Sauer, A.M., Bautista, M.R., Turner, R.S., Yang, J., Mayer, M., 2009. Neurotoxicity Research 16 (1), 1–13.

Cedazo-Minguez, A., Popescu, B.O., Blanco-Millan, J.M., Akterin, S., Pei, J.J., Winblad,B., Cowburn, R.F., 2003. Journal of Neurochemistry 87 (5), 1152–1164.

Chabot, V., Cuerrier, C.M., Escher, E., Aimez, V., Grandbois, M., Charette, P.G., 2009.Biosensors and Bioelectronics 24 (6), 1667–1673.

Chinowsky, T.M., Quinn, J.G., Bartholomew, D.U., Kaiser, R., Elkind, J.L., 2003.Sensors and Actuators B-Chemical 91 (1–3), 266–274.

Cuerrier, C.M., Chabot, V., Vigneux, S., Aimez, V., Escher, E., Gobeil, F., Charette, P.G.,Grandbois, M., 2008. Cellular and Molecular Bioengineering 1 (4), 229–239.

Deli, M.A., Veszelka, S., Csiszar, B., Toth, A., Kittel, A., Csete, M., Sipos, A., Szalai, A.,Fulop, L., Penke, B., Abraham, C.S., Niwa, M., 2010. Journal of Alzheimer'sDisease 22 (3), 777–794.

Famulski, J.K., Solecki, D.J., 2012. Trends in Neurosciences 36 (3), 163–173.Fang, Y., Li, G., Ferrie, A.M., 2007. Journal of Pharmacological and Toxicological

Methods 55 (3), 314–322.Foley, K.J., Shan, X., Tao, N.J., 2008. Analytical Chemistry 80 (13), 5146–5151.Gaspar, S., David, S., Polonschii, C., Marcu, I., Gheorghiu, M., Gheorghiu, E., 2012.

Analytica Chimica Acta 713, 115–120.Ghenim, L., Kaji, H., Hoshino, Y., Ishibashi, T., Haguet, V., Gidrol, X., Nishizawa, M.,

2010. Lab on a Chip 10 (19), 2546–2550.Gheorghiu, E., 1996. Bioelectromagnetics 17 (6), 475–482.Gheorghiu, E., Balut, C., Gheorghiu, M., 2002. Physics in Medicine and Biology 47

(2), 341–348.Gheorghiu, M., Gersing, E., Gheorghiu, E., 1999. Annals of the New York Academy of

Sciences 873, 65–71.Gheorghiu, M., Olaru, A., Tar, A., Polonschii, C., Gheorghiu, E., 2009. Biosensors and

Bioelectronics 24 (12), 3517–3523.Giaever, I., Keese, C.R., 1991. Proceedings of the National Academy of Sciences U S A.

88 (17), 7896–7900.Gonzalez-Velasquez, F.J., Kotarek, J.A., Moss, M.A., 2008. Journal of Neurochemistry

107 (2), 466–477.Haass, C., Koo, E.H., Capell, A., Teplow, D.B., Selkoe, D.J., 1995. Journal of Cell Biology

128 (4), 537–547.Haass, C., Koo, E.H., Teplow, D.B., Selkoe, D.J., 1994. Proceedings of the National

Academy of Sciences U S A. 91 (4), 1564–1568.Han, A., Yang, L., Frazier, A.B., 2007. Clinical Cancer Research 13 (1), 139–143.



Hong, J., Kandasamy, K., Marimuthu, M., Choi, C.S., Kim, S., 2011. Analyst 136 (2),237–245.

Jarrett, J.T., Berger, E.P., Lansbury Jr., P.T., 1993. Annals of the New York Academy ofSciences 695, 144–148.

Jepson, M.A., 2003. Cellular and Molecular Biology (Noisy-le-grand) 49 (1),101–112.

Kammermann, M., Denelavas, A., Imbach, A., Grether, U., Dehmlow, H., Apfel, C.M.,Hertel, C., 2011. Biochemical and Biophysical Research Communications 412 (3),419–424.

Kerr, E., Craig, D., McGuinness, B., Dynan, K.B., Fogarty, D., Johnston, J.A., Passmore,A.P., 2009. International Journal of Geriatric Psychiatry 24 (9), 927–932.

Klein, W.L., Stine Jr., W.B., Teplow, D.B., 2004. Neurobiology of Aging 25 (5),569–580.

Kotarek, J.A., Moss, M.A., 2010. Analytical Biochemistry 399 (1), 30–38.Lashuel, H.A., Lansbury Jr., P.T., 2006. Quarterly Reviews of Biophysics 39 (2),

167–201.Macdonalds, J.R., 1987. Impedance Spectroscopy: Emphasizing Solid Materials and

Systems. Wiley, lnterscience, New York.Marco, S., Skaper, S.D., 2006. Neuroscience Letters 401 (3), 219–224.Mobley, D.L., Cox, D.L., Singh, R.R., Maddox, M.W., Longo, M.L., 2004. Biophysical

Journal 86 (6), 3585–3597.Nagababu, E., Usatyuk, P.V., Enika, D., Natarajan, V., Rifkind, J.M., 2009. Journal of

Alzheimer's Disease 17 (4), 845–854.Nazer, B., Hong, S., Selkoe, D.J., 2008. Neurobiology of Disease 30 (1), 94–102.Olaru, A., Gheorghiu, M., David, S., Polonschii, C., Gheorghiu, E., 2013. Biosensors

and Bioelectronics 45C, 77–81.Olaru, A., Gheorghiu, M., David, S., Wohland, T., Gheorghiu, E., 2009. Journal of

Physical Chemistry B 113 (43), 14369–14380.

Peters, M.F., Scott, C.W., 2009. Journal of Biomolecular Screening 14 (3), 246–255.Primiceri, E., Chiriaco, M.S., D’Amone, E., Urso, E., Ionescu, R.E., Rizzello, A., Maffia,

M., Cingolani, R., Rinaldi, R., Maruccio, G., 2010. Biosensors and Bioelectronics25 (12), 2711–2716.

Robelek, R., Wegener, J., 2010. Biosensors and Bioelectronics 25 (5), 1221–1224.Sandu, T., Vrinceanu, D., Gheorghiu, E., 2010. Physical Review E: Statistical,

Nonlinear, and Soft Matter Physics 81 (2 Pt 1), 021913.Schasfoort, B.M.R, Tudos, J.A. (Eds.), 2008. Handbook of Surface Plasmon Resonance.

RSC Publishing, Cambridge U.K.Schroder, R., Janssen, N., Schmidt, J., Kebig, A., Merten, N., Hennen, S., Muller, A.,

Blattermann, S., Mohr-Andra, M., Zahn, S., Wenzel, J., Smith, N.J., Gomeza, J.,Drewke, C., Milligan, G., Mohr, K., Kostenis, E., 2010. Nature Biotechnology 28(9), 943–949.

Solito, R., Corti, F., Fossati, S., Mezhericher, E., Donnini, S., Ghiso, J., Giachetti, A.,Rostagno, A., Ziche, M., 2009. Experimental Cell Research, 315 (3), 385–395.

Terrettaz, S., Stora, T., Duschl, C., Vogel, H., 1993. Langmuir 9 (5), 9.Vala, M., Robelek, R., Bockova, M., Wegener, J., Homola, J., 2012. Biosensors and

Bioelectronics 40 (1), 417–421.Verdier, Y., Zarandi, M., Penke, B., 2004. Journal of Peptide Science 10 (5), 229–248.Wang, W., Foley, K., Shan, X., Wang, S., Eaton, S., Nagaraj, V.J., Wiktor, P., Patel, U.,

Tao, N., 2011. Nature Chemistry 3 (3), 249–255.Williams, T.L., Serpell, L.C., 2011. FEBS Journal 278 (20), 3905–3917.Yanase, Y., Hiragun, T., Kaneko, S., Gould, H.J., Greaves, M.W., Hide, M., 2010.

Biosensors and Bioelectronics 26 (2), 674–681.Yanase, Y., Hiragun, T., Yanase, T., Kawaguchi, T., Ishii, K., Hide, M., 2012. Biosensors

and Bioelectronics 32 (1), 62–68.Yashunsky, V., Lirtsman, V., Golosovsky, M., Davidov, D., Aroeti, B., 2010. Biophysical

Journal 99 (12), 4028–4036.


Label free sensing platform for amyloid fibrils effect on living cells

Documents