Andrea Mazzatenta et al- Interfacing Neurons with Carbon Nanotubes: Electrical Signal Transfer and...

8/3/2019 Andrea Mazzatenta et al- Interfacing Neurons with Carbon Nanotubes: Electrical Signal Transfer and Synaptic Stimul

1/6

Brief Communications

Interfacing Neurons with Carbon Nanotubes: Electrical

Signal Transfer and Synaptic Stimulation in CulturedBrain Circuits

AndreaMazzatenta,1 *Michele Giugliano,3* Stephane Campidelli,2 LucaGambazzi,3 LucaBusinaro,4 HenryMarkram,3

Maurizio Prato,2 andLauraBallerini11Physiology and Pathology Department, B.R.A.I.N., University of Trieste, I-34127, Trieste, Italy, 2Department of Pharmaceutical Sciences, University of

Trieste, I-34127, Trieste, Italy, 3Laboratory of Neural Microcircuitry, Brain Mind Institute, Ecole Polytechnique Federale de Lausanne, Station 15CH-1015

Lausanne, Switzerland, and 4LILIT BeamLine, TASC-INFM National Laboratory, Area Science Park, I-34012, Basovizza, Trieste, Italy

The unique properties of single-wall carbon nanotubes (SWNTs) and the application of nanotechnology to the nervous system may have

a tremendous impactin thefuture developmentsof microsystemsfor neuralprostheticsas well as immediatebenefits forbasicresearch.Despite increasinginterest in neurosciencenanotechnologies, little is knownabout theelectricalinteractionsbetween nanomaterials and

neurons. We developed an integrated SWNTneuron system to test whether electrical stimulation delivered via SWNT can induceneuronal signaling. To that aim, hippocampal cells were grown on pure SWNT substrates and patch clamped. We compared neuronal

responsesto voltage steps delivered either viaconductive SWNT substrates or via the patch pipette. Ourexperimentalresults,supportedby mathematical models to describe the electrical interactions occurring in SWNTneuron hybrid systems, clearly indicate that SWNTs

can directly stimulate brain circuit activity.

Key words: nanotechnology; hippocampal neurons; neuronal stimulation; voltage clamp; biophysical modeling; seal resistance

IntroductionDifferent biomedical devices implantedin the CNS,better knownas neural interfaces, have been developed to control motor disor-ders (Benabid et al., 2005) or to translate willful brain processesinto specific actions by the control of external devices (Mussa-Ivaldi and Miller, 2003; Lebedev and Nicolelis, 2006). Clinicalprogresses of the numerous interface concepts proposed in basicresearch are limited by common recurrent problems, rangingfrom improper neuronal adhesion to inadequate signal stability.The success in overcoming these problems for the design of fu-ture interfaces mainly relies on the application of emergingtechnologies (but see also Fromherz, 2002; Stieglitz et al.,2005). Examples of current research include technologies,

such as nanotechnology, that are designed to improve materi-al/neuron junctions to be applied as implantable interfaces(Patolsky et al., 2006).

The potential of nanotechnology application in neuroscienceis widely accepted (Silva, 2006). In particular, SWNTs have re-ceivedgreat attention because of their uniquephysicaland chem-ical features, which allow the development of devices with out-standing electrical properties(Krishnanet al., 1998). Recent workhas focused on the feasibility of using high-capacitance, low-resistance electrodes (Gabay et al., 2007), with the goal of large-scale integration with CNS interfaces (Patolsky et al., 2006). TheSWNT biocompatibility has been shown previously (Mattson etal., 2000; Hu et al., 2005; Lovat et al., 2005), and neuronal adhe-sion, survival, and growth can be modulated by SWNTpolymerconjugates (Bekyarova et al., 2005). Nanofibers have been re-ported to minimize astrocyte reactions, leading to decreased glialscar tissue (McKenzie et al., 2004) in addition, the nano-scaledimension of SWNTs allows molecular interactions with neu-rons (Silva, 2006) that will determine the electrical interfacing ofSWNTs to neural cells.

We recently demonstrated that purified carbon nanotubes aregood growth surfaces for neurons and promote an increase in theefficacy of neural signal transmission (Lovat et al., 2005). Al-though recent reports indicate the possibility of stimulating iso-lated neurons in culture via SWNT (Liopo et al., 2006), the un-derstanding of such electrical coupling is very poor. In particular,the possibility to evoke, via purified SWNT substrates, synapticactivity in long-term neural circuits is unknown.

Wereport here, byscanningelectronmicroscopy,the presenceofintimate contacts between cultured neurons and purified SWNT,which might provide the physical substrate for neuronSWNT

Received March 8, 2007; revised May 15, 2007; accepted May 15, 2007.

ThisworkwassupportedbytheMinistryofUniversityandResearch(COFINandMUR-FIRBtoM.P.andtoL.B.)and

EC-NEURONANO-NMP4-CT-2006-031847. We thank Dr. C. Petersen for helpful discussions, Dr. M. Hines for assis-

tancewithNEURON,Dr. L.Vaccarifor assistancewithscanningmicroscopy,Dr. S.Graziosifor computerand graphic

assistance, and Drs. B. Pastore and C. Zacchigna for assistance with tissue cultures.

*A.M. and M.G. contributed equally to this work.

Correspondence should be addressed to either of the following: Dr. Laura Ballerini, Physiology and Pathology

Department, Center for Neuroscience, B.R.A.I.N., University of Trieste, via Fleming 22, 34127, Trieste, Italy, E-mail:

[email protected]; or Maurizio Prato, Department of Pharmaceutical Sciences, University of Trieste, Piazzale

Europa 1, I-34127, Trieste, Italy, E-mail: [email protected]:10.1523/JNEUROSCI.1051-07.2007

Copyright 2007 Society for Neuroscience 0270-6474/07/276931-06$15.00/0

The Journal of Neuroscience, June 27, 2007 27(26):69316936 6931


2/6

neuron electrical coupling. In voltage-clamped neurons, we characterize responsesevokedviaSWNT stimulation.Usingneuro-nal circuits chronically grown on purifiedSWNT layer, we demonstrate how SWNTstimulations reliably evoke postsynaptic re-sponses. Finally, a mathematical model in-

corporating SWNT and neuronal propertiesreproducesthemain features of theresponseto SWNT stimulations. This model providesfor the first time the basis for understandingthe electrical coupling between neurons andSWNT.

Materials andMethodsPurification of SWNTs and deposition onto glasscoverslips. The nanotubes (HiPCO nanotubes;Carbon Nanotechnology, Houston, TX) werefunctionalized via 1,3-dipolar cycloaddition re-action (Georgakilas et al., 2002), in dimethyl-formamide (DMF), between sarcosine, hep-

tanal, and purified SWNT. After reaction, thenanotubes were filtered on a polytetrafluoro-ethylene membrane and extensively washedwith DMF and dichloromethane. For deposi-tion on the coverslips, functionalized SWNTswere resuspended in fresh DMF (using a ultra-sonic bath), and few drops of the solution weredeposited on the glass slides. After slow evapo-ration of DMF (100C), the glass coverslipswere annealed at 350C under N

2for 20 min.

This process permits to defunctionalize thenanotubes and fix them on the surface.

Resistivity of the SWNT deposition was esti-mated to be 1.01.2 mm, by the four-wiremeasurement technique and a Precision Im-

pedance Analyzer (4294A; Agilent Technology,Santa Clara, CA). Capacitive reactance was negligible, being the phase of

measured impedance 0.1 in thefrequencyrange2001000Hz. SWNT

deposition with a uniform cross-section thickness of 5060 nm was

assumed, based on scanning electron microscopy investigations using

focused ion beam technique (data not shown).

Tissue cultures. Standard dissociated hippocampal cultures (32 series)

were prepared according to Lovat et al. (2005). Hippocampi were dis-

sected from 0/3-d-old Sprague Dawley rats killed by decapitation. This

procedureis in accordancewith therelevantEuropeanUnionlegislation.

Hippocampal slices were digested, and cells were plated on peptide-free

or on SWNT-treated glass coverslips and cultured for 8 14 d (Lovat et

al., 2005).

Electrophysiological recordings. Recording solution contained the fol-

lowing (in mM): 156 NaCl, 4 KCl, 1 MgCl2, 2 CaCl2, 10 HEPES, and 10glucose (Lovat et al., 2005). Whole-cell patch-clamp recordings were

obtained at room temperature using patch pipettes (47 M), under

G patch sealing, using an Axopatch 1-D (Molecular Devices, Foster

City, CA). Intracellular solution contained the following (mM): 120

K-gluconate, 20 KCl, 10 HEPES, 10 EGTA, 2 MgCl2, and 2 Na

2ATP (pH

7.35 adding KOH). In voltage clamp, holding potential was set at 58

mV. The uncompensated value for series resistance was 8 10 M. In

current-clamp recordings, bridge balancing was continuously moni-

tored and adjusted. On a sample of 40 neurons (18 culture series), the

resting membrane potential did not significantly differ between controls

(47 2 mV) and SWNT (49 3 mV) neurons, as well as the input

resistance (RIN

) andthe cell-capacitance values(544 50 M; 69 7 pF

and 555 54 M; 62 9 pF, glass and SWNT, respectively).

Details on electrical SWNT stimulation, electron microscopy, immu-nocytochemistry, and modeling are provided as supplemental Methods

(available at www.jneurosci.org as supplemental material).

ResultsCarbon nanotubes and neuronal morphologiesStandard conductivity measurements revealed that the SWNTlayer acts as a purely resistive homogeneous network, character-ized by a resistance between any two points of the order of 110k. This is consistent with scanning electron microscopy imagesshowing that SWNTs form a dense meshwork characterized by alarge surface roughness (Fig. 1A), suggesting a very large SWNTelectrolyte capacitance (Gabay et al., 2007).

We achieved direct SWNTneuron interactions by culturingrat hippocampal cells on a film of purified SWNTs for 8 14 d, toallow for neuronal growth. Neuronal cell morphologies were an-

alyzed by scanning electron microscopy (n 20 cultures) andquantified by immunocytochemistry (see supplemental Fig. 1,available at www.jneurosci.org as supplemental material) (seealso Lovat et al., 2005). In Figure 1 BD, scanning electron mi-croscopy micrographs show that healthy hippocampal neuronsadhered and grew on SWNT substrates displaying typical cellbody morphology and size. As shown in Figure 1 BD, neuronalgrowth was accompanied by variable degree of neurite extensionon the SWNT mat, similar to those observed for neurons grownon control, peptide-free glass surfaces and confirmed by immu-nocytochemistry analysis (supplemental Fig. 1, available at www.jneurosci.org as supplemental material) (Lovat et al., 2005). De-tailed scanning electron microscopy analysis at higher magnifi-

cations of this and additional cultures suggested the presence oftight interactions between cell membranes and SWNTs at thelevel of neuronal processes and cell surfaces. In Fig. 1 EF, high-

A B

C D

E F

Figure 1. Scanning electron microscopy images of cultured hippocampal neurons on SWNTs. A, High-magnification micro-graphshowing SWNTdetails.BD, Subsequent micrographs athigher magnificationsof neurons grownon SWNTs(10 d).Samesampleas inA is shown.Notethe healthy morphology of neurons andthe outgrowthof neuritesattaching tothe SWNT surface.

E, F, Details of the framed area in D. At higher magnifications, the intimate contacts between bundles of SWNT and neuronalmembrane are clearly shown. Scale bar (in E):A, 1m;B, 200m; C, 25m; D, 10m; E, 2m; F, 450 nm.

6932 J. Neurosci., June 27, 2007 27(26):6931 6936 Mazzatenta et al. Neural Stimulation through Carbon Nanotubes


3/6

resolution micrographs were taken at subsequent magnifications

to obtain details of a region in which a presumed afferent processcontacted the neuronal cell body (Fig. 1 DF). By scanning theSWNT surface topography, we observed the stable retention of

SWNT bundles (Fig. 1 F, arrow). SuchSWNT bundles appeared intermingledwith neuronal membranes, and such atight contact is potentially helpful in stabi-lizing electrical interfacing. The picture inFigure1 Freports a typical example of suchproximity between substrate and neuronal

membranes. Although our specimenswere not gold-coated to enhance EM res-olution, we could resolve such structureswithout distortion by beam irradiation(Fig. 1 F).

Distributed electrical stimulation ofcultured hippocampal neuronsHippocampal neurons, grown on SWNTsor on control glass substrates, displayedprominent spontaneous electrical activityafter the first days in culture (Lovat et al.,2005). In each experiment, baseline re-cordings were performed from neuronsdisplaying similar passive properties (seeMaterials and Methods), grown onSWNTs or on control glass substrates.Subsequently, in each SWNT neuron (n 82), the responses to electrical stimulationvia SWNT substrates were studied byvoltage-clampand, in some cases, current-clamp recordings.

We characterized cell behavior in bothglass and SWNT growth conditions. Un-der current clamp, a typical strong in-crease in the average frequency of sponta-neous action potentials (APs) (12 3 Hz

and44 1 Hz,control andSWNT,respec-tively, n 40 cells, each group) was ob-served in neurons grown on SWNTs (Fig.2A, top) (Lovat et al., 2005) when com-pared with controls. In Figure 2A (bot-tom), under voltage-clamp recordings,such high firing activity is reflected bythe occurrence of spontaneous PSCs,which were fully blocked by coapplica-tion of SR-95531 (gabazine; 5 M) andCNQX (5 M).

We did not further characterize suchchanges in circuit activity, because the in-

crease in APs and frequency of PCSs aretypical landmarks of neurons grown onSWNT substrates, as reported previously(Lovat et al., 2005).

To investigate neuronal responses toSWNT stimulations, we used a modifiedchamber with a small dry area, isolatedfrom the recording chamber, where an Agelectrode was positioned in electrical con-tact with theSWNTfilm to deliver voltage-controlled stimulation (Fig. 2 B, sketch).In a first set of experiments, individual

current responses were elicited in voltage-clamped neurons via

the Ag wireSWNT stimulation by delivering square-pulse volt-age steps (20 ms duration; 00.3 V; n 31) (Fig. 2 B, left) ofdistinct positive and negative amplitudes. As shown in Figure 2 B

A

500 ms300

pA

20mV

B+

10 ms200pA

C

200 ms200pA

-60 -30 30

-400

400

800

I (pA)

Ve (mV)

-1,5 1,5E (V)

-400

-200

200

400

I (pA)

5 ms

2nA

E

1 s

30 ms300pA

D

10mV

300pA

3 ms

VeE

Figure 2. Electricalinteractions between SWNTs andneurons.A, Spontaneous activityrecordedfrom a 12 d culturedneuron,undercurrent-clamp (top trace) or voltage-clamp (bottom trace) configurations.B, Sketch of the recording chamber. Tracings, Current stepselicited by SWNTstimulation recorded froma patch-clamped neuron (left) or in control-glass preparation (middle) or in SWNTs beforesealingtoacell(right). C,Top,Currentresponsesto400-ms-longSWNTstimuliinavoltage-clampedneuron,summarizedintheplot(0.2

Vbins;n5).Bottom,Currentresponsesin avoltage-clampedneuronelicitedviathepatchpipettebya stepprotocol,summarizedintheplot (20 mV bins; n 5). D, Postsynaptic events evoked via SWNT stimulation (*) and recorded under voltage-clamp (bottom) orcurrent-clamp (top) configurations. E, Spontaneous patterned activity and SWNT stimulation. Note that similar waveforms could beevoked viaSWNT stimuli(*). Bottom,Singleevoked andspontaneousevents,as indicatedby bars.

Mazzatenta et al. Neural Stimulation through Carbon Nanotubes J. Neurosci., June 27, 2007 27(26):69316936 6933


4/6

(middle), analogous voltage steps deliv-ered in control-glass preparation resultedonly in the appearance of fast capacitivetransients on both polarities. In contrast,voltage steps delivered via SWNT inducedcurrent changes, on both polarities, whenthe tip of the patch electrode was simply

submerged by saline solution (Fig. 2 B,right). Thus, voltage steps applied to theSWNT layer were sensedas steadycurrentsby the patch electrode under voltageclamp, suggesting the possible generationof redox current flows (Girault, 2004).

In Figure 2C(top row, left), longer cur-rent steps with similar amplitude and po-larity were obtained by delivering 400 msvoltage steps viaSWNT to measure steady-state current values (Fig. 2C, plot, toprow). Larger SWNT voltage steps inducedthe appearance of a fast inward current(range 15 nA, n 12) (Fig. 2C, top row,right), which was abolished by TTX (0.1M; 5 min) applications. When switchingto current clamp, delivering supra-threshold SWNT stimulations induced theappearance of repetitive APs on a depolar-izing step (n 4) (supplemental Fig. 2,available at www.jneurosci.org as supple-mental material).

Voltage steps of similar duration (from30 mV to 110 mV in 20 mV steps; n 15), applied directlyvia the patch pipette,generated larger current responses (Fig.2C, bottom, left tracings, 10, 10, 70

mV steps;see also I/Vplot, middle). Stron-ger depolarizing voltage steps deliveredthrough the patch pipette (30mV) (Fig. 2C, bottom, right trac-ings) resulted again in fast inward TTX-sensitive currents (range,13.5 nA; n 6).

Under both stimulating conditions, large inward currentsprobably represent unclampedAPs in distal processes (Arancio etal., 1995).

The similarity among extrinsic (via SWNT) and intrinsic (viapatch pipette) stimulations is intriguing and opens several ques-tions about the nature of the coupling between SWNT and neu-rons; in particular, we asked whether stimulating theSWNT layereffectively generates a steady current flow in the closely attachedneurons and whether the voltage-clamp configuration is theproper experimental setting to evaluate such an issue (Liopo etal., 2006).

To assess the applicability of the SWNT layer to perform localnetwork stimulations, we delivered SWNT brief voltage stepsstrong enough to induce the appearance of Na fast inward cur-rent in the recorded neuron. Such stimulations should also in-duce APs in neighboring neurons. Induced APs will triggermonosynaptic responses to the connected neurons, including therecorded one. In fact, short (2 ms)SWNTvoltage pulsesdeliveredin brief trains of five consecutive steps induced, in 65% of therecorded neurons, PSCs (Fig. 2 D, bottom) (n 38) or the cor-

responding PSPs (top) when switching to current-clamp config-uration (n 10). Under both recording conditions, reversingstimulus polarity failed to evoke postsynaptic events. Evoked

events were completely abolished by coapplication of SR-95531(gabazine) and CNQX.

We then selected hippocampal networks grown on SWNTswhere neurons showed the occurrence of bursts of synchronizedPSCs, superimposed on a background activity, and separated byquiescent periods of variable length (Fig. 2 E, top tracings). Abrief stimulation pulse (2 ms) delivered via the SWNT layer reli-ably evoked a reproducible summated PSC. Such evoked epi-sodes displayed current waveforms similar to those of spontane-ously generated ones (Fig. 2 E, bottom tracings), suggesting theinduction of similar firing patterns in the network.

Modeling the neuronSWNT junctionOur recordings suggest that the coupling between neurons andSWNTs might be in part resistive (Liopo et al., 2006). Persistent(DC) lateral electrical stimulation delivered through the SWNTlayer could (hyperpolarize) depolarize the membrane and (in-hibit) elicit sustained firing in neurons intimately growing onSWNT.

To elucidate such a DC coupling, we defined, analyticallystudied, and numerically simulated simple resistive electricalequivalent models for the SWNTelectrolyte and the SWNTneuron junctions (Fig. 3). These models are reduced versionsof detailed capacitiveresistive descriptions incorporatingHodgkinHuxley-like currents, numerically simulated in Fig. 4.

For the SWNTelectrolyte junction, we privileged a phenome-nological description to a biophysicalelectrochemical investiga-tion as by Grattarola and Martinoia (1993) for metal/semicon-

+ Rs ve

ieRc

E

CRe

D+ Rs ve

ieRe

ERm

vrest

+ Rs ve

ieRe

ERm

vrest

RcE

+ Rs1 ve

ieRe

ERm

vrest

Rc

Rs2

F

ve

ie

E+

Ave

ie

E+

B

Figure3. ModelingelectricalstimulationdeliveredviatheSWNTlayer.A,B,Sketchofthebath(A)orofthewhole-cellpipetteconfigurations(B).TheDCequivalentcircuitofA isrepresentedinC.ForB,thehypothesisontheresistivecapacitivenatureoftheSWNTcellcoupling,aswellasthequalityofthewhole-cellconfiguration,mustbecomeexplicit.Theseareidealwhole-cellpatchclamp with (E) or without (D) a resistive electrical coupling between the cytoplasm and SWNT. Finally, Fequivalently captures

bothoftheprevioussituations(DF),underthehypothesisofanonidealwhole-cellconfiguration.Thevalueof Rcinonecasehas

the same meaning as in Cbut in the other, has the same meaning as in E.



5/6

ductorneuron interface. Nevertheless, we note that the DCproperties modeled, observed, and discussed here might be at-tributable to oxygen chemabsorption at the SWNTelectrolyteinterface (Perkins and Robinson, 2006). These reactions, not be-inga feature of pristine nanotubes, are probably unavoidable sideeffects of a variety of technological fabrication processes.

Figure 3 sketches two situations: the pipette in the bath (A) orestablishing a whole-cell patch ( B). For the first configuration,thecircuit of Figure3Caccounts forthe DC properties of both theSWNT and pipetteelectrolyte interfaces, as confirmed experi-mentally (Fig. 2 B). In fact, simulating single-electrode voltageclamp (SEVC), during step voltage-command, ve, through thepipette or applying a voltage step E through the SWNT (Fig. 4C),one measures a current, ie, through the pipette:

ie ve/Re//Rc E/Rc. (1)

Re//Rc indicates theresistancesparallelof thepipette (Re 5 M)and of the resistive component of the SWNTelectrolyte inter-face (Rc 100400 M, estimated through Eq.1 and Fig. 2 B),and E is the voltage applied to SWNT.

In the second configuration, an ideal seal is obtained. Thecircuit of Figure 3D models the conditions of no resistive cou-pling (Rc ) between SWNT and the intracellular potential.Thus, E has no effect on ie as

ie ye vrest/ReRm. (2)

Rm (500600 M) is the input resistance, and vrest its restingpotential. This situation can be rejected on the basis of Figure 2 B(Liopo et al., 2006): when a step voltage E is applied throughSWNT, ie indeed undergoes a steady change.

Letus nowassume that a resistivepathway between theSWNTsubstrate and thecytoplasm exists modeled as in Figure 3E. Then,E or ve should induce similar perturbations (Fig. 4A, D,E). Undersuch hypotheses,

ie ve Rc//Rm Vrest/Rm E/Rc Rc//Rm Re1. (3)

If resistive coupling occurs from a nano-scale portion of theSWNT layer, effectively exposed to the cytoplasm, the corre-sponding value ofRc is certainly much larger than those in Equa-

tion 1. IfRc is larger than the membrane resistance Rm (i.e., (Rc//Rm)Rm), Equations 2 and 3 become similar. Thus, (passive)

membrane currents can be evoked by acting on ve or byE, re-

corded under SEVC as ie. Both ve and Eappear on the right side of Equation 3,with E weighted by a small factor R

m/

(Rm Rc) (Fig. 4A). This was confirmedby the large values ofE (1001000 mV)experimentally required to elicit currentsie thatweresimilartothoseevokedbysweeps

10100 times smaller in ve (see Fig. 2B).What happens if we challenge the hy-pothesis of resistive coupling, consideringthe realistic case of an excellent but non-ideal membrane-patch sealing? This iscaptured by the model circuit of Figure 3F,and ie takes the following form:

ie ve Rs2//Rc//Rm vrest/Rm

E/Rcs2//Rc//Rm Re)1. (4)

Interestingly, Equations 3 and 4 are iden-tical, apart from the additional parallelresistor Rs2, which accounts for the (gi-

gaohm-large) seal resistance of the patch pipette.However, if we had previously assumed that no resistive cou-

pling exists (i.e., Rc M) and that the observed DC couplingoccurs through the patch-seal resistance connecting the pipettetipto theextracellular SWNT, we would have derived an identicalequation. In fact, although weak, the seal resistance provides apath to ground and to SWNT. In such a scenario, one could notimmediately discriminate between (1) the gigaohm-large valuesofRc, because of the very small cytosolic domain of SWNT and(2) the series between a smaller Rc (100400 M) and agigaohm-large seal-resistance path to the pipette tip.

Evoked PSCs in the patched cell represent the only, althoughindirect, evidence of SWNT induction of APs in neighboring

neurons (see Fig. 2 D, E). In fact, although our current-clamprecordings of sustained firing evoked by steady SWNT stimuli(supplemental Fig. 2, available at www.jneurosci.org as supple-mental material) are another suggestive evidence of charge trans-fer between SWNT and neurons, they do not completely rule outthe same caveats as for the voltage clamp. Notably, the nonide-alities of series and seal resistance under SEVC can therefore hidemany features of SWNTneuron coupling (Liopo et al., 2006).Excellent but nonideal series resistance can lead to unclampedaction potentials after stimulation of E or ve as experimentallyobserved (Figs. 2C) and replicated in the model simulations (Fig.4). In addition, because of the intrinsic ambiguity of Equation 4,the pipette voltage can be affected equivalently byve and E, apart

from appropriate scaling factors (Rc).

DiscussionThe main result of the present study is the new observation thatneuronal circuits chronically grown on SWNT substrates can beeffectively stimulated via the SWNT layers. Here, we propose forthe first time a model of neuronal/SWNT coupling that shouldimprove the understanding of bionanomaterial interactions. Wespecifically addressed the issue of the electrical coupling betweenSWNT and neurons and outlined possible deceiving resultsemerging from the applications of SEVC configurations in suchexperiments (Liopo et al., 2006). In addition, we show here the

possibility to stimulate single and multiple synaptic pathways incultured networks via SWNTplatforms. Elucidating the electricalcoupling between nanomaterial and neurons should also hint at

D E

A B C

10 ms

2.5 ms 2.5 ms

300pA

2.5nA

0.1mV

2.5nA

0.7mV

Figure4. Simulatedstimulations viathe SWNTlayer. Intracellular currents elicited by SWNTstimulation simulatedin a modelneuroninthepresence(A)orabsence(B)ofSWNTandbeforesealingtoacell(C),as inFigure2. Theemissionof singleunclamped

APs simulated in the models through holding voltage steps (D) or by SWNT-delivered stimulation (E), when incorporating thedetails of SEVC.

Mazzatenta et al. Neural Stimulation through Carbon Nanotubes J. Neurosci., June 27, 2007 27(26):69316936 6935


6/6

highly targeted design of SWNT-based interfaces, minimizingunwanted interactions (Silva, 2006).

Cultured brain circuits provide an in vitro simple model of aneuronal network. Hippocampal neurons grew and developedfunctional circuits on SWNT surfaces, indicating, as detailed pre-viously, the general biocompatibility of purified SWNT (Hu etal., 2005;Lovatet al., 2005). When compared with control abiotic

surfaces, SWNT boosted neuronal network activity underchronic growth conditions (Lovat et al., 2005). This effect hasbeen described previously and is not attributable to differences inneuronal survival, morphology, or passive membrane propertiesbut possibly represents a consequence of the properties of theSWNT substrate (Lovat et al., 2005). Our scanning electron mi-croscopy observations of neuron/SWNT contacts, comparedwith the current state-of-the-art (Gheith et al., 2006; Liopo et al.,2006), suggested superior neuronal cell adhesion. In fact, in theabsence of specific coating, the distance between electrode sur-faces and cells is most likely rather large (Lu et al., 2006). How-ever, it is not known whether our reported intimate adhesionbetween hippocampal neurons and SWNT affects normal cellfunction. Our recordings seem to exclude a long-term neurotox-icity of such interactions, because the detected spontaneous pat-terned activity is omnipresent in cultured neuronal circuits andhas been shown to primarily result from synaptic interactions(Marom and Shahaf, 2002). This indicates that neurons, withinsynaptically interacting networks formed on SWNT substrates,engage in population firing. This is further strengthened by theappearance of fast Na current, taken to constitute an early signof axonal differentiation (Alessandri-Haber et al., 1999).

Thus far, very few investigators succeeded in growing primaryneuronal cultures on SWNT substrates, none takingadvantage ofa purified and mechanically stable substrate, whose electricalproperties could be related to SWNT properties by traditionaland electrochemical methods (Gabay et al., 2007). The suggested

adhesion properties between neurons and SWNTs might alsosustain unconventional electrical coupling, thus unveiling newapproaches to basic understanding of the CNS electrophysiology(Silva, 2006).

The electrical coupling between neurons and SWNTs is a rel-evant matter (Llinas et al., 2005). Our experimental results,strengthened by themodeling, supports theidea that any resistivecoupling between biomembranes and SWNT(Gheith et al., 2006;Liopo et al., 2006) is qualitatively undistinguishable from a cou-pling between SWNT and the patch pipette through the patch-seal path to ground. Because of the nonidealities of the SEVC,eliciting Na currents through SWNT stimulation does not con-clusively prove a resistive coupling between SWNT and neurons.

It is indeed through detecting synaptic responses, evoked by APselicited in not-clamped neurons, that the effectiveness of SWNTneuron interaction can be assessed. On a first approximation,regardless of the interpretative complications of the voltage-clamp data, such a phenomenological observation suggests thatSWNT-coated substratesmight provide the best mechanicalcou-pling between artificial devices and neural tissue.

ReferencesAlessandri-Haber N, Paillart C, Arsac C, Gola M, Couraud F, Crest M (1999)

Specific distribution of sodium channels in axons of rat embryo spinalmotoneurones. J Physiol (Lond) 518:203214.

Arancio O, Kandel ER, Hawkins RD (1995) Activity-dependent long-term

enhancement of transmitter release by presynaptic 3,5-cyclic-GMP in

cultured hippocampal neurons. Nature 376:7480.

Bekyarova E, Ni Y, Malarkey EB, Montana V, McWilliams JL, Haddon RC,

Parpura V (2005) Applications of carbon nanotubes in biotechnology

biomedicine. J Biomed Nanotech 1:317.

Benabid AL, Wallace B, Mitrofanis J, Xia C, Piallat B, Fraix V, Batir A, Krack

P, Pollak P, Berger F (2005) Therapeutic electrical stimulation of the

central nervous system. C R Biol 328:177186.Fromherz P (2002) Electrical interfacing of nerve cells and semiconductor

chips. Chemphyschem 3:276284.

Gabay T, Ben-David M, Kalifa I, Sorkin R, Abrams ZR, Ben-Jacob E, Hanein

Y (2007) Electro-chemical biological properties of carbon nanotube

based multi-electrode arrays. Nanotechnology: 035201 (6 pp).

Georgakilas V, Kordatos K, Prato M, Guldi DM, Holzinger M, Hirsch A

(2002) Organic functionalization of carbon nanotubes. J Am Chem Soc

124:760761.

Gheith MK, Pappas TC, Liopo AV, Sinani V, Shim BS, Motamedi M, Wick-

sted, JP Kotov NA (2006) Stimulation of neural cells by lateral currents

in conductive layer-by-layerfilms of single-walledcarbon nanotubes.Adv

Mater Deerfield 18:29752979.

Girault HH (2004) Analytical and chemical electrochemistry. New York:

Marcel Dekker.

Grattarola M, Martinoia S (1993) Modeling the neuron-microtransducerjunction: from extracellular to patch recording. IEEE Trans Biomed Eng

40:3541.

HuH, NiY, MandalSK,Montana V,ZhaoB, HaddonRC,Parpura V (2005)

Polyethyleneimine functionalized single-walled carbon nanotubes as a

substrate for neuronal growth. J Phys Chem B 109:42854289.

Krishnan A, Dujardin E, Ebbesen TW, Yianilos PN, Treacy MMJ (1998)

Youngs modulus of single-walled nanotubes. Phys Rev B

58:1401314019.

Lebedev MA, Nicolelis MA (2006) Brain-machine interfaces: past, present

future. Trends Neurosci 29:536546.

Liopo AV, Stewart MP, Hudson J, Tour JM, Pappas TC (2006) Biocompat-

ibility of native functionalized single-walled carbon nanotubes for neuro-

nal interface. J Nanosci Nanotechnol 6:13651374.

Llinas RR, Walton KD, Nakao M, Hunter I, Anquetil PA (2005) Neuro-

vascular central nervous recording/stimulating system: using nanotech-nology probes. J Nanoparticle Res 7:111127.

Lovat V, Pantarotto D, Lagostena L, Cacciari B, Grandolfo M, Righi M, Spal-

luto G, Prato M, Ballerini L (2005) Carbon nanotube substrates boost

neuronal electrical signaling. Nano Lett 5:11071110.

Lu S, Bansal A, Soussou W, Berger TW, Madhukar A (2006) Receptor-

ligand-based specific cell adhesion on solid surfaces: hippocampal neuro-

nal cells on bilinker functionalized glass. Nano Lett 6:19771981.

MaromS, Shahaf G (2002) Development,learningmemoryin largerandom

networks of cortical neurons: lessons beyond anatomy. Q Rev Biophys

35:6387.Mattson MP, Haddon RC, Rao AM (2000) Molecular functionalization of

carbon nanotubes use as substrates for neuronal growth. J Mol Neurosci

14:175182.

McKenzie JL, Waid MC, Shi R, Webster TJ (2004) Decreased functions of

astrocytes on carbon nanofiber materials. Biomaterials 25:13091317.Mussa-Ivaldi FA, Miller LE (2003) Brain-machine interfaces: computa-

tional demands clinical needs meet basic neuroscience. Trends Neurosci

26:329334.

Patolsky F, Timko BP, Yu G, Fang Y, Greytak AB, Zheng G, Lieber CM

(2006) Detection, stimulation, inhibition of neuronal signals with high-density nanowire transistor arrays. Science 313:11001104.

Perkins FK, Robinson JA (2006) Chemical vapor deposition using single-

walled carbon nanotubes. Chem Soc Rev 35:790798.

Silva GA (2006) Neuroscience nanotechnology:progress,opportunitiesand

challenges. Nat Rev Neurosci 7:6574.

Stieglitz T, Schuettler M, Koch KP (2005) Implantable biomedical micro-

systems for neural prostheses. IEEE Eng Med Biol Magn 24:58 65.


Andrea Mazzatenta et al- Interfacing Neurons with Carbon Nanotubes: Electrical Signal Transfer and...

Documents

Transcript of Andrea Mazzatenta et al- Interfacing Neurons with Carbon Nanotubes: Electrical Signal Transfer and...