at SciVerse ScienceDirect

Neuropharmacology 73 (2013) 301e310

Contents lists available

Neuropharmacology

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

Adenosine A2A receptors inhibit delayed rectifier potassium currentsand cell differentiation in primary purified oligodendrocyte cultures

Elisabetta Coppi 1, Lucrezia Cellai, Giovanna Maraula, Anna Maria Pugliese, Felicita Pedata*

Department of Neuroscience, Psychology, Drug Research and Child Health (NEUROFARBA), Division of Pharmacology and Toxicology, University of Florence,Viale Pieraccini 6, 50139 Florence, Italy

a r t i c l e i n f o

Article history:Received 6 December 2012Received in revised form7 May 2013Accepted 19 May 2013

Keywords:Adenosine A2A receptorsKþ currentsOligodendrocyte differentiationMyelination

Abbreviations: OPC, oligodendrocyte progenitorcurrent; DMSO, dimethyl sulphoxide; IA, transient Kþ

Amino-9-(N-ethyl-b-D-ribofuranuronamidosyl)-9H-puzene propanoic acid; SCH58261, 2-(2-Furanyl)-7-(2-pe][1,2,4]triazolo[1,5-c]pyrimidin-5-amine; TEA, ttetrodotoxin.* Corresponding author. Tel.: þ39 055 4271262; fax

E-mail address: felicita.pedata@unifi.it (F. Pedata).1 Present address: Department of Health Sciences,

Pieraccini 6, 50139 Florence, Italy.

0028-3908/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.neuropharm.2013.05.035

a b s t r a c t

Oligodendrocyte progenitor cells (OPCs) are a population of cycling cells which persist in the adultcentral nervous system (CNS) where, under opportune stimuli, they differentiate into mature myeli-nating oligodendrocytes. Adenosine A2A receptors are Gs-coupled P1 purinergic receptors which arewidely distributed throughout the CNS. It has been demonstrated that OPCs express A2A receptors, buttheir functional role in these cells remains elusive. Oligodendrocytes express distinct voltage-gated ionchannels depending on their maturation. Here, by electrophysiological recordings coupled with immu-nocytochemical labeling, we studied the effects of adenosine A2A receptors on membrane currents anddifferentiation of purified primary OPCs isolated from the rat cortex. We found that the selective A2A

agonist, CGS21680, inhibits sustained, delayed rectifier, Kþ currents (IK) without modifying transient (IA)conductances. The effect was observed in all cells tested, independently from time in culture. CGS21680inhibition of IK current was concentration-dependent (10e200 nM) and blocked in the presence of theselective A2A antagonist SCH58261 (100 nM).

It is known that IK currents play an important role during OPC development since their blockdecreases cell proliferation and differentiation. In light of these data, our further aim was to investigatewhether A2A receptors modulate these processes. CGS21680, applied at 100 nM in the culture medium ofoligodendrocyte cultures, inhibits OPC differentiation (an effect prevented by SCH58261) withoutaffecting cell proliferation.

Data demonstrate that cultured OPCs express functional A2A receptors whose activation negativelymodulate IK currents. We propose that, by this mechanism, A2A adenosine receptors inhibit OPCdifferentiation.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The developing and mature central nervous system (CNS) con-tains NG2þ stem-like cells that serve as oligodendrocyte progenitorcells (OPCs) which are the primary source of myelinating cells.Myelination of neuronal axons by oligodendrocytes is essential forsaltatory conduction of electric impulses thus allowing fast

cell; IK, delayed rectifier Kþ

current; CGS21680, 4-[2-[[6-rin-2-yl]amino]ethyl]ben-henylethyl)-7H-pyrazolo[4,3-etraethylammonium; TTX,

: þ39 055 4271280.

University of Florence, Viale

All rights reserved.

synaptic transmission in the vertebrate CNS. In the adult brain andspinal cord, OPCs are the majority of proliferating cells (Dawsonet al., 2003; Horner et al., 2000) and present multifunctional fea-tures since, under specific conditions, they can also give rise toneurons and astrocytes (Nishiyama et al., 2009). Furthermore, someNG2þ cells might also generate action potentials (De Biase et al.,2010; Karadottir et al., 2008) and physically interact with axonterminals, leading to the novel concept of neuron-glia synapses.

Cultured and in situ OPCs express distinct voltage-gated ionchannels depending on their maturation (Sontheimer et al., 1989),including both inward and outward rectifying Kþ channels(Sontheimer and Kettenmann, 1988; Williamson et al., 1997), Naþ

channels (Berger et al., 1992; Barres et al., 1990) and differentsubtypes of Ca2þ channels (Berger et al., 1992; Verkhratsky et al.,1990). In undifferentiated OPCs, outward rectifying, sustained, IKcurrents represent the major component of outward Kþ currentsthat also comprise transient, rapidly inactivating, currents (IA)

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310302

(Soliven et al., 1988; Attali et al., 1997; Knutson et al., 1997). It isknown that IK currents are involved in the regulation of oligoden-drocyte differentiation, since their block inhibits this process (Galloet al., 1996; Ghiani et al., 1999; Attali et al., 1997; Chittajallu et al.,2002; Vautier et al., 2004).

Purinergic signaling has recently emerged as a most pervasivemechanism for intercellular communication in the nervous system,affecting communication between many types of neurons and alltypes of glial cells (Fields and Burnstock, 2006). Concerning oligo-dendrocytes, it has been demonstrated that purines exert multipleeffects including increased motility, proliferation and differentia-tion of cultured OPCs (Othman et al., 2003; Agresti et al., 2005).Among purines, adenosine is a neuromodulator of the CNS, where itacts on four subtypes of P1 purinergic receptors (A1, A2A, A2B, A3)(Fredholm et al., 2011). It is known that OPCs express all foursubtypes of adenosine receptors, including A2A receptors (Stevenset al., 2002). However, to date, a functional role has been attrib-uted only to A1 receptors. Adenosine, through the activation of A1receptors, seems to be a primary activity-dependent signal inhib-iting the proliferation and promoting differentiation of premyeli-nating progenitors into myelinating oligodendrocytes (Stevenset al., 2002) and stimulating OPC migration (Othman et al., 2003).In the present work we investigated the functional role of adeno-sine A2A receptors in purified primary OPC cultures. In particular,we examined the effect of A2A receptor activation on in vitro OPCproliferation and differentiation and their role in modulatingmembrane currents recorded in cultured OPCs.

Identifying the molecular mechanisms underlying oligoden-drocyte development is not only critical to furthering our knowl-edge of oligodendrocyte biology, but also has implications forunderstanding the pathogenesis of demyelinating diseases such asmultiple sclerosis (MS). For example, the observation that OPCs arepresent in MS lesions but fail to differentiate into mature oligo-dendrocytes (Levine and Reynolds, 1999; Chang et al., 2000) sug-gests that the remyelination process is blocked at a premyelinatingstage in demyelinating lesions. Thus, identification of critical reg-ulators that inhibit myelination/remyelination could facilitate thedevelopment of therapeutic targets for myelin repair in CNSdemyelinating diseases. Our results demonstrate that the selectiveactivation of A2A receptors decreases outward rectifying, sustained,IK currents and inhibits in vitro OPC differentiation towards mature,myelinating, oligodendrocytes, without affecting cell division.

2. Materials and methods

2.1. Cell cultures

All animal procedures were conducted according to the Italian Guidelines forAnimal Care, DL 116/92, application of the European Communities Council Directive(86/609/EEC). Purified cortical OPC cultures were prepared as described elsewhere(Fumagalli et al., 2011). Wistar rat pups (postnatal day 1e2) were killed and corticesremoved, mechanically and enzymatically dissociated, suspended in DMEM me-dium containing 20% fetal bovine serum (FBS), 4 mM L-glutamine, 1 mM Na-pyruvate, 100 U/ml penicillin, 100 U/ml streptomycin (all products are from Euro-Clone, Milan, Italy), and plated in poly-D-lysin coated T75 flasks (1 flask per animal).After 2e3 days in culture, OPCs growing on top of a confluent monolayer of astro-cytes were detached by overnight horizontal shaking. Contaminating microglialcells were eliminated by a 1 h pre-shake and by further plating detached cells onplastic culture dishes for 1 h. OPCs, which do not attach to plastic, were collected bygently washing the dishes and replated onto poly-DL-ornithine-coated (final con-centration: 50 mg/ml, SigmaeAldrich, Italy) 13 mm-diameter glass coverslips placedin 24 multiwell chambers (l04 cells/well). OPC cultures were maintained in Neuro-basal medium (Invitrogen, S. Giuliano Milanese, Milan, Italy) containing 2% B27,4 mM L-glutamine, 1 mM Na-pyruvate, 100 U/ml penicillin, 100 U/ml streptomycin,10 ng/ml platelet derived growth factor (PDGF-BB) and 10 ng/ml basic fibroblastgrowth factor (bFGF; both growth factors were from PeproTech EC Ltd, London, UK)to promote cell proliferation (proliferating medium: PM). After 2e3 days in PM, cellswere switched to a Neurobasal medium lacking growth factors in order to allow celldifferentiation (differentiating medium: DM).

As bipolar NG2þ progenitors undergo differentiation under these conditions,they becomemultipolar and acquire O4þ immunoreactivity before the expression ofmyelin basic protein (MBP) and myelin associated glycoprotein (MAG) (Mann et al.,2008). No contaminating microglial cells and a very low percentage (<1%) of as-trocytes were detected in OPC cultures when Iba1 or glial fibrillary acid protein(GFAP) staining, respectively, was performed (data not shown).

Experiments were performed on cells at different times in culture: just beforeswitching to DM (t0), or at different times after switching to DM (e.g. after 1 day ofculture in DM: t1). In the whole text, the term “OL cultures” (oligodendrocyte cul-tures) indicates primary purified OPC cultures at different times of maturation.

2.2. Immunofluorescence analysis

Purified primary OPCs at different times in culture were fixed with 4% para-formaldehyde in 0.1 M phosphate buffered saline (PBS) for 20 min at room tem-perature (RT). The following primary antibodies were diluted in bovine serumdilution buffer (BSDB: 450 mM NaCl, 20 mM sodium phosphate buffer, pH 7.4, 15%bovine serum, 0.3% Triton X-100) and incubated for 2.5 h at RT: mouse anti-NG2(Millipore, Temecula, CA, USA; 1:500), mouse anti-O4 (Millipore, Temecula, CA,USA; 1:100), mouse anti-MAG (Millipore, Temecula, CA, USA; 1:250), rabbit anti-Iba1 (Wako, Osaka, Japan; 1:300) and mouse anti-GFAP (BD Biosciences, FranklinLakes, NJ USA; 1:500).

Cells were then washed three times with PBS and incubated for 1 h at RT withthe following secondary antibodies (diluted 1:500 in BSDB): donkey anti-mouse andgoat anti-rabbit conjugated to AlexaFluor 488 and to AlexaFluor 555, respectively(Molecular Probes, Invitrogen, Milan, Italy). Coverslips were mounted with Vecta-shield mounting medium (Vector Laboratories, Burlingame, CA, US) containing 40 ,6-diamidino-2-phenylindole (DAPI) to visualize cell nuclei, and analyzed by using anOlympus BX40 microscope coupled to analySIS’B Imaging Software (Olympus,Milan, Italy). Control experiments were performed in each group by incubating fixedcells with the secondary antibody alone in order to exclude non-specific binding.The number of cells labeled by a specific antibody in each coverslip was quantifiedby counting the number of fluorescent cells in 10 random microscopic fields (20�)and expressed as a percentage over the total cell number (DAPIþ nuclei) in the samefield. Two coverslips for any given experimental condition were evaluated. Experi-ments were run at least in triplicate. Cell density in each coverslip was quantified byaveraging the total number of cells (DAPIþ nuclei) counted in 10 microscopic fields.All quantification analyses were performed blind by two different experimenter andresults were averaged.

BrdU Incorporation. The degree of OPC proliferation was determined byanalyzing 5-Bromo-20-deoxyuridine (BrdU) incorporation in cell cultures. After 3days of cell culture in PM, OPCs were switched to DM for 3 days, either in theabsence (control group) or in the presence of 100 nM CGS21680. Brdu (10 mM) wasadded to both experimental groups 24 h before cell fixing. After fixing, cells wereincubated in 2 NHCl for 30min at RT to denaturate nuclear DNA, followed bywasheswith 0.1M sodium borate, pH 8.5, to neutralize HCl. Immunostaining was performedusing mouse anti-BrdU (1:400; Abcam, Cambridge, UK) antibody. The number ofBrdUþ cells in each coverslip was quantified by counting the number of fluorescentcells in 10 random microscopic fields (20�) and expressed as the percentage overthe total cell number (DAPIþ nuclei) in the same field. Two coverslips for any givenexperimental condition were evaluated. Experiments were run at least in triplicate.

2.3. Electrophysiology

Whole-cell patch-clamp recordings were performed on purified primary OPCs atdifferent stages of maturation: undifferentiated OPCs were recorded from t0-t2cultures with careful attention to select cells with a strict bipolar morphology;multipolar pre-OLGs were chosen from t5et6 cultures whereas highly ramifiedmature OLGs were recorded from t10et13 cell cultures. Cells grown on a poly-DL-ornithine-coated 13 mm-diameter glass coverslip were transferred to a smallchamber (1 ml volume) mounted on the platform of an inverted microscope(Olympus CKX41, Milan, Italy) and superfused at a flow rate of 2 ml/min with astandard extracellular solution containing (mM): HEPES 5, D-glucose 10, NaCl 140,KCl 5.4, MgCl2 1.2, and CaCl2 1.8 (pH adjusted to 7.3 with NaOH). Borosilicate glasselectrodes (Harvard Apparatus, Holliston, MA USA) were pulled with a Sutter In-struments puller (model P-87) to a final tip resistance of 4e7 MU. Standard pipettesolution contained (in mM): K-gluconate 130, NaCl 6, MgCl2 2, Na2-ATP 2, Na2-GTP0.3, EGTA 0.6, HEPES 10 (pH adjusted to 7.4 with KOH). For Kþ-replacement exper-iments a Csþ-based pipette solution was used (in mM): CsCl 130, NaCl 6, MgCl2 2,Na2-ATP 2, Na2-GTP 0.3, EGTA 0.6, HEPES 10 (pH adjusted to 7.4 with CsOH). Datawere acquired with an Axopatch 200B amplifier (Axon Instruments, CA, USA), low-pass filtered at 10 kHz, stored and analyzed with a pClamp 9.2 software (Axon In-struments, CA, USA). All the experiments were carried out at RT (20e22 �C). Pro-tocols. Unless otherwise stated, cells were voltage-clamped at �70 mV. Capacitivetransients generated by the electrode and by cell membrane were digitally sub-tracted by the amplified circuit. Series resistance (Rs), membrane resistance (Rm) andmembrane capacitance (Cm) were routinely measured by fast hyperpolarizingvoltage pulses (from�70 to�75mV, 40ms duration). Only cells showing a stable Cm

and Rs before, during, and after drug application were included in the analysis.

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310 303

Immediately after breakthrough into whole-cell configuration, cell resting mem-brane potential (Vrest) was determined by switching to the current-clamp mode. Avoltage ramp protocol (800 ms depolarization from �140 to þ80 mV) was used toevoke a wide range of overall voltage-dependent membrane currents before, duringand after drug treatments. Currents evoked by the voltage ramp in OL cultures at alltimes of maturation are Kþ currents since theywere absent when cells were patchedwith a Csþ-based pipette solution (data not shown). Representative ramp tracesshown in graphs are the average of 5 consecutive episodes (5 s interval). Variationsof membrane potential (Vm) induced by drug treatments were measured by calcu-lating the reversal potential (the “zero current” potential) of ramp-evoked currentsbefore, during and after drug application. Outward Kþ currents were evoked by twodifferent depolarizing voltage-step protocols in order to separate delayed rectifieroutward Kþ currents (IK) from transient outward (IA) conductances. A first protocolwas applied in which 10 depolarizing voltage steps (10 mV steps from �40to þ80 mV, 200 ms each) were preceded by a 60 ms pre-step potential (Vpre) at�80 mV. This protocol activates a mixture of outward IK and IA currents in culturedOPCs. Since transient IA currents present a voltage-dependent inactivation at po-tential positive to�80mV, a second protocol was applied in the same cell with a Vpre

at�40 mVwhich selectively inactivates IA leaving the IK component unchanged. NetIA current expressed in each cell was then obtained by digital subtraction of the twotraces. Current-to-voltage relationships (IeV plots) of IK or IA currents were obtainedbymeasuring current amplitude at the steady state (200e250ms after step onset) oras a peak (1e20 ms after step onset), respectively. Inward rectifying Kþ (Kir) currentswere activated by hyperpolarizing voltage steps (10 mV steps from �10 mV to�130 mV, 100 ms each) with a Vpre of 0 mV (100 ms duration) in order to inactivateeventual inward Naþ currents (INa). INa were recorded by using a Csþ-based pipettesolution andwere activated by a depolarizing voltage step protocol (5 or 10mV stepsfrom �55 mV to þ40 mV, 15 ms each) starting from a holding potential (Vh) of�90 mV in order to remove Naþ channel inactivation.

Current amplitude (measured as pA) was normalized to respective cell capaci-tance (measured in pF) and expressed as current density (pA/pF) in averaged results.All drugs were applied by superfusion with a three-way perfusion valve controller(Harvard Apparatus, Holliston, Massachusetts US) after a stable baseline of ramp-evoked currents was obtained. A complete exchange of bath solution in therecording chamber was achieved within 30 s.

Immunofluorescence and electrophysiological experiments were run in parallelin order to correlate the current profile with maturation stages of cultured cells.

2.4. Drugs

4-[2-[[6-Amino-9-(N-ethyl-b-D-ribofuranuronamidosyl)-9H-purin-2-yl]amino]ethyl]benzene propanoic acid (CGS21680) and 2-(2-Furanyl)-7-(2-phenylethyl)-7H-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-5-amine (SCH58261) were purchasedfrom Tocris (Bristol, UK; http://www.tocris.com). Tetraethylammonium (TEA) andBrdU were obtained from SIGMA (SigmaeAldrich, Italy). Tetrodotoxin (TTX) waspurchased from Ascent Scientific (Cambridge, UK; http://www.ascentscientific.com). TEA, BrdU and TTX were dissolved in distilled water. CGS21680 andSCH58261 were dissolved in dimethyl sulfoxide (DMSO). Stock solutions of 103e104

times the desired final concentration were stored at �20 �C, daily diluted in theextracellular solution to the final concentration and applied by bath superfusion. Thefinal concentration of DMSO (0.1% and 0.2%) used in our experiments did not affectthe amplitude of currents evoked by ramp or voltage-step protocols in culturedOPCs, nor oligodendrocyte in vitro maturation.

2.5. Data analysis

Data are expressed as mean � SEM (standard error of the mean). Student’spaired or unpaired t-tests or One-way ANOVA followed by NewmaneKeuls post-testanalysis were performed, as appropriate, in order to determine statistical signifi-cance (set at P < 0.05). Data were analyzed using software package GRAPHPAD PRISM

(GraphPad Software, San Diego, CA, USA).

3. Results

3.1. Immunocytochemical and electrophysiological characterizationof purified primary OL cultures at different steps of maturation

The correlation between cellular markers of differentiation andthe expression of distinct voltage-dependent currents in purifiedprimary OPCs has been deeply studied by various investigators(Soliven et al., 1989; Williamson et al., 1997; Attali et al., 1997;Knutson et al., 1997; Sontheimer et al., 1989; Barres et al., 1990).In accordance with the literature, we found that when OL cultureswere grown in the presence of mitogens (t0: before switching toDM, i.e. after 2e3 days in PM) the vast majority (>95%) of cells werebipolar or tripolar, immature, NG2þ OPCs (Fig. 1A, left panel).

When investigated by electrophysiological means, currentsactivated by a voltage-ramp protocol in these cells were dominatedby outward rectifying Kþ conductances (Fig. 1B, left panel), asconfirmed by Csþ-replacement experiments (see Methods). Aspreviously described (Blankenfeld et al., 1992; Mann et al., 2008), apopulation of OPCs (17 out of 27 cells tested: 63%) also expressedfast inward Na currents (INa) blocked by 1 mM TTX (Fig. 1C). After 6days of mitogens withdrawal (6 days in DM: t6), cells differentiatedinto O4þ multipolar pre-oligodendrocytes (pe-OLGs) (Fig. 1A, cen-tral panel: 42.4 � 3.2% of O4þ cells at t6) characterized by areduction of outward rectifying Kþ conductances (Fig. 1B, centralpanel: note the different y scale versus Fig. 1B, left panel) and by anincrease in inward Kþ currents. After 10 days in DM (t10) fullydifferentiated MAGþ oligodendrocytes (OLGs) were observed(Fig. 1A, right panel: 56.2 � 5.7% of MAGþ cells at t10). A significantcomponent of ramp-evoked currents in mature OLGs was carriedby inwardly rectifying Kþ conductances (Fig. 1D) with a voltage-dependence typical of Kir channels (Hibino et al., 2010). INa wasnever detected in OL cultures after t3.

It has been demonstrated that Kþ currents expressed by OPCs arerepresented by delayed rectified, sustained, Kþ (IK) currents andtransient “Atype” (IA) Kþ conductances (SontheimerandKettenmann,1988; Knutson et al., 1997). We applied two different voltage-stepprotocols in order to separate these two components. When depola-rizing voltage pulses (from �40 to þ80 mV) were applied startingfrom a pre-step potential (Vpre) of �80 mV, a rapidly activating andinactivating IA current was followed by a “sustained” IK conductance(Fig. 1E, left panel). When the same pulses were preceded by a Vpre of�80mV, only the transient component was observed (Fig. 1E, centralpanel). Digital subtraction of the two traces allow for the isolation of IA(Fig. 1E, right panel). All currents evoked by both step protocols arecarried by Kþ ions since they are nullified by intracellular Kþ

replacement with equimolar Csþ (Supplemental Fig. 1). As expectedfor IA, transient currentswereblockedby300mM4-aminopyridine (4-AP: data not shown) whereas the IK blocker TEA inhibited the sus-tained component (see below). In accordance with previous data(Attali et al., 1997), 36 out of 52 cells investigated (about 70% of cellsfrom t0 to t10 OL cultures) presented both IA and IK currents, whereasthe remaining 14 cells only expressed IK conductances.

3.2. The selective stimulation of adenosine A2A receptors inhibit IKcurrents in OL cultures

We applied a voltage ramp protocol in cultured OPCs at differenttimes of maturation (from t0 to t10) in the absence and in thepresence of CGS21680, a selective A2A agonist, in order to investi-gate whether this receptor modulates Kþ currents in these cells.Fig. 2A, left panel, shows that the application of 100 nM CGS21680in a typical OPC (tripolar cell at t2) reversibly inhibited ramp-evoked outward Kþ conductances. CGS-sensitive current, ob-tained in the same cell by subtraction of the ramp recorded in thepresence of the A2A agonist from the control ramp, is a voltage-dependent Kþ current with an activation voltage of about 10 mV(Fig. 2A, right panel). Similar results were obtained in 8 cellsinvestigated, independently from cell morphology or time in cul-ture, so datawere pooled. A significant decrease of total outward Kþ

currents was observed in this experimental group (Fig. 2B, leftpanel: from 287.2 � 42.6 pA/pF in control to 224.9 � 40.8 pA/pF inCGS21680 at þ80 mV: 21.7% inhibition; P < 0.001, paired Student’st-test, n¼ 8). Themaximal effect on current amplitudewas reachedafter 5 min of drug application and was completely reversed after5 min of washout (Fig. 2B, right panel). As expected upon inhibitionof Kþ channels, cell membrane significantly depolarized in thepresence of CGS21680 (from �48.4 � 2.8 mV in control to�42.9 � 2.9 mV after 5 min in CGS21680; P < 0.01, paired Student’s

Fig. 1. Immunocytochemical and electrophysiological characterization of OL cultures at different steps of maturation. A. Immunocytochemical staining of purified primary OPCcultures at different steps of maturation. Left panel: at t0 (before switching to differentiation medium: DM) the vast majority (88.2 � 2.8%) of cells in the culture were NG2þ OPCs withthe typical bipolar morphology. Central panel: after 6 days in DM (t6) 42.4 � 3.2% of cells differentiated into O4þ post-mitotic pre-OLGs with a ramified morphology. Right panel: after10 days in DM (t10) 54.9 � 3.7% of cells were highly ramified MAGþ mature OLGs. Blue staining represents cell nuclei marked with DAPI. Data were obtained from two replicated wellsof three independent experiments. Scale bar: 50 mm. B. Different electrophysiological properties characterize distinct stages of cell maturation. Left panel: representative current traceelicited in a bipolar OPC at t0 by a voltage ramp protocol (inset). Note that, at this stage of maturation, the voltage ramp activated a majority of outward currents. During cell dif-ferentiation (central panel: pre-OLG, t6) outward currents decreased in amplitude (note different scales of y axis in left, central and right panels) and an inward component appeared atvoltages negative to �40 mV, which was predominant in mature oligodendrocytes (right panel: OLG, t10) and presented the typical voltage dependence of Kir currents (grayarrowhead). C. INa traces and respective IeV plot recorded in the OPC. Scale bars: 0.5 nA; 2 ms. D. Kir currents and respective IeV plot recorded in the OLG. Scale bars: 0.2 nA; 20 ms. E.Original current traces elicited by two different voltage step protocols (lower insets) in a representative OPC at t1. A pre-step potential (Vpre) of �80 mV elicited both transient outwardKþ (IA) and delayed rectifier, sustained (IK) currents whereas a Vpre of�40mV selectively inactivated IA and isolated IK currents. Net IA current was obtained by digital subtraction of thetwo traces. Scale bars: 1 nA; 50 ms.

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310304

t-test, n ¼ 8) with a time course parallel to IK inhibition (Fig. 2B,right panel).

We further investigatedwhich typeof Kþ currents ismodulatedbyA2A receptor activation by measuring CGS21680 effects on IK and IAcurrents isolated by the voltage-step protocols described above. AsshowninFig. 2C,whendepolarizingvoltage stepswereappliedbeforeand after CGS21680 application, only IK currents (upper panels) wereinhibited by the A2A agonist, with a significant decrease of currentdensity observed from þ20 mV steps and a maximal reductionatþ80mV(from267.6� 44.6pA/pF incontrol to191.9�39.5pA/pF inCGS21680, P< 0.01, paired Student’s t-test, n¼ 8). On the contrary, IAcurrentswereunchanged in thepresenceof thepurinergic compound

(Fig. 2C, lower panels: from 463.3 � 70.3 pA/pF in control to424.2 � 64.8 pA/pF at þ80 mV in CGS21680, n ¼ 8). Above datademonstrate that A2A receptor activation selectively inhibits IK cur-rents in cultured OPCs. As above mentioned, IK currents decrease inamplitude during OPCs maturation, for this reason further experi-ments were performed on OL cultures from t0 to t6.

The effect of CGS21680 on ramp evoked currents wasconcentration-dependent (Fig. 3A). The agonist was ineffective up to10 nM, when the amplitude of CGS21680-senstitive current(32.2 � 9.9 pA/pF at þ 80 mV in CGS21680, n ¼ 4) was significantlydifferent from control (only DMSO application: 1.7 � 2.6 pA/pF atþ 80 mV, n ¼ 5; P < 0.05 vs. 10 nM CGS21680, unpaired Student’s t-

Fig. 2. CGS21680, a selective agonist of adenosine A2A receptors, inhibits IK currents in cultured OPCs. A. Left panel: original ramp-evoked currents in a representative cell (tripolarOPC at t2) before (ctrl), during and after (5 min washout) 100 nM CGS21680 (CGS) application. Right panel: net CGS21680-sensitive current obtained in the same cell by subtractionof the ramp recorded in CGS21680 from that recorded in control. Note the different scale of y axis. B. Left panel: averaged ramp-evoked currents recorded before and after 5 minapplication of 100 nM CGS21680 in 8 cells investigated. Right panel: time course of ramp-evoked current density at a voltage of þ80 mV (filled circles, left y axis) and of membranepotential (Vm: open circles, right y axis) before and after 5 min application of CGS21680. **P < 0.01, paired Student’s t-test, n ¼ 8. C. Original current traces elicited by two differentvoltage step protocols in control conditions or in the presence of CGS21680 in a representative OPC at t2. Left panels show respective IeV plots of IK and IA currents obtained from 8cells tested. *P < 0.05; **P < 0.01, paired Student’s t-test. Scale bars IK: 1 nA; 50 ms. Scale bars IA: 0.5 nA; 50 ms.

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310 305

test). The maximal effect was observed at 100 nM (Fig. 3A), sincefurther increase inCGS21680 concentration tended toproduce a sub-maximal effect, possibly by loosing selectivity (Fig. 3A, inset in theleft panel). The concentrationeresponse curve of CGS21680 effect onramp currents (Fig. 3A, right panel) gave an EC50 of 9.9 nM (confi-dential limits: 5.3e18.5 nM), which is in line with previous data(Jarvis et al.,1989) obtainedonrat striatalmembranes (Kd¼15.5nM).

In order to confirm a receptor-mediated effect, we appliedCGS21680 in the presence of the selective A2A receptor antagonistSCH58261 (used at a concentration of 100 nM on the basis of aKi ¼ 2.3 nM in the rat brain: Zocchi et al., 1996). As shown in Fig. 3B,both ramp-evoked (Fig. 3B, left panel) and step-evoked (Fig. 3B,right panel) currents were unchanged when the purinergic agonistwas applied in the presence of the antagonist.

In some experiments, where appreciable INa currents wererecorded in cultured OPCs, CGS21680 did not modify their ampli-tude (Supplemental Fig. 2).

3.3. The IK blocker TEA occludes the effect of CGS21680 in culturedOPCs

In the next series of experiments, we investigated whether theeffect of A2A receptors on IK currents was sensitive to the typical IK

blocker TEA. In accordancewith the literature (Gallo et al., 1996), weobserved a massive decrease of total outward Kþ currents elicitedby the ramp protocol in the presence of 3 mM TEA (Fig. 4A, rightpanel, from a: 205.3� 23.1 pA/pF in control to b: 84.4�11.1 pA/pF inTEA atþ80 mV, 58.9% current inhibition, P< 0.001, paired Student’st-test, n ¼ 8). As expected upon massive block of Kþ currents, andsimilarly to what observed in the presence of CGS21680 (see Fig. 2B,right panel), TEA also produced a significant depolarization of cellmembrane potential, as summarized in Fig. 4A, central panel (from�41 � 4.7 mV in control to �37.3� 4.3 mV after 5 min of 3 mM TEAapplication, P< 0.05, paired Student’s t-test, n¼ 8). As shown in thetime course of ramp-evoked currents atþ80mV (Fig. 4A, left panel),TEA effect reached a steady-state after 2 min application andcompletely prevented the IK inhibition induced by a subsequentapplication of CGS21680 (Fig. 4A, right panel, from b: 84.4�11.1 pA/pF in TEA to c: 83.6 � 10.7 pA/pF in TEA þ CGS21680 at þ80 mV,n ¼ 8). We next studied whether TEA differently affected IK and IAcurrents byapplyingdepolarizingvoltage stepprotocols.As shown inthe left panel of Fig. 4B, TEA, at the concentrationused, only inhibitedIK currents (from 205.8� 23.8 pA/pF in control to 111.2� 12.4 pA/pFin TEA at þ80 mV, P < 0.01, paired Student’s t-test, n ¼ 5) withoutaffecting transient IA conductances (from 359.5 � 37.3 pA/pF incontrol to 331.7 � 15.9 pA/pF in TEA at þ80 mV, n ¼ 5).

Fig. 3. The effect of CGS21680 on IK currents is concentration-dependent and is blocked in the presence of the adenosine A2A receptor antagonist SCH58261. A. Left panel: averagedCGS-sensitive currents evoked in OPCs by the voltage ramp in the presence of different agonist concentrations. Inset: column bars summarize the effect of CGS21680 on rampcurrents at þ80 mV alone or in the presence of the A2A antagonist SCH58261 (100 nM). Number of cell tested is indicated in parenthesis. Right panel: Concentration-response curveof CGS21680 effect on ramp currents at þ80 mV (EC50 ¼ 9.9 nM; Confidential Limits: 5.3e18.5 nM). B. Averaged ramp-evoked currents (left panel) and IeV plot of IK and IA currents(right panel) recorded in the presence of SCH58261 alone or in combination with CGS21680.

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310306

In line with TEA occlusion of CGS21680 effect on IK currents, Vmwas unchanged when the A2A receptor agonist was applied incombination with the IK blocker (Fig. 4B, central panel:from �38.1 � 5.4 mV in control to �36.2 � 5.2 mV in CGS21680,

Fig. 4. The effect of CGS21680 is prevented in the presence of the IK blocker TEA. A. Left pCentral panel: column bars summarize the effect of 3 mM TEA on membrane potential (Vm)paired Student’s t-test, n ¼ 8. Right panel: averaged ramp-evoked currents recorded in contcells investigated. B. Right panel: IeV plot of IK and IA currents recorded in the absence or inpanel: no effect on Vm was observed when CGS21680 was applied in the presence of TEA (nalone or during co-application with 100 nM CGS21680.

n ¼ 8). Similarly, neither IK nor IA currents elicited by the voltagestep protocols were influenced by CGS21680 when applied in thepresence of TEA (Fig. 4B, right panel, IK: from 107.7 � 16.6 pA/pF inTEA to 97.5 � 15.8 pA/pF in TEA þ CGS21680 at þ80 mV, n ¼ 5; IA:

anel: time course of TEA effect on ramp-evoked currents at þ80 mV in 8 cells tested.. A significant depolarization was observed in the presence of the IK blocker. *P < 0.05,rol conditions (a: ctrl), in the presence of 3 mM TEA (b) or in TEA þ CGS21680 (c) in 8the presence of 3 mM TEA. *P < 0.05; **P < 0.01, paired Student’s t-test, n ¼ 5. Central¼ 8). Right panel: IeV plot of IK and IA currents recorded in the presence of 3 mM TEA

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310 307

from 350.4 � 24.6 pA/pF in TEA to 315.4 � 27.5 pA/pF inTEA þ CGS21680 at þ80 mV, n ¼ 5).

3.4. Adenosine A2A receptor activation inhibits in vitro OPCdifferentiation

In keepingwith data about thewell recognized role of purines incell differentiation (Fields and Burnstock, 2006), we next examinedthe effects of A2A receptor stimulation on in vitro oligodendrocyteprogression towards mature phenotypes. We performed immu-nocytochemical labeling of OL cultures allowed to differentiate fordifferent days in DM (t0, t3, t6 and t9) in order to characterize thetime course of the expression of specific markers of cell differen-tiation. The percentage of NG2þ, O4þ and MAGþ cells was quanti-fied at each time point. In parallel experiments, OL cultures weredifferentiated in the presence of CGS21680 (100 nM) alone or incombination with the A2A antagonist SCH58261 (100 nM), in orderto verify whether A2A receptor activation influenced OPC differen-tiation. As shown in Fig. 5, we found that A2A receptor activationreduced in vitro oligodendrocyte maturation since it increased thepercentage of NG2þ OPCs and diminished the percentage of O4þ

Fig. 5. The selective stimulation of adenosine A2A receptor inhibits in vitro OPC differentiatio(red) of OL cultures grown for 6 days (t6) in DM under control conditions (ctrl: left panelsbination with the A2A antagonist SCH58261 (CGS þ SCH: right panels). Cell nuclei are markpanel) and MAG (right panel) expression in OL cultures grown in DM under control conSCH58261. All experiments were performed at least in triplicate. Ten microscopic fields wereevaluated.

andMAGþ cells starting from day 3 in DM (t3). A significant effect ofCGS21680 on all immunocytochemical markers was observed fromt6, when the percentage of NG2þ cells increased from 50.5� 2.6% incontrol to 69 � 3.8% in CGS21680 (P < 0.05, One way ANOVA,NewmaneKeuls post test) whereas O4þ and MAGþ cells decreasedfrom 42.4 � 3.2% in control to 24 � 2.6% in CGS21680 (P < 0.001,One way ANOVA, NewmaneKeuls post test) and from 33.6 � 3.7%in control to 16.2 � 2.6% in CGS21680 (P < 0.01, One way ANOVA,Newman-eKeuls post test), respectively.

The effect of CGS21680 on OPC differentiation was completelyblocked in the presence of the selective A2A receptor antagonistSCH58261, as no significant differences were found in the per-centage of NG2þ (Fig. 5B, left panel), O4þ (Fig. 5B, central panel) orMAGþ (Fig. 5B, right panel) cells in OL cultures differentiated inCGS21680 þ SCH58261 versus the control group at any timeinvestigated.

In control experiments OPCs were differentiated in the presenceof DMSO alone at the final concentration used to dissolve CGS21680and SCH59261 (0.2%). In this experimental group we observed aslight increase in cell differentiation, even if statistical significancewas not reached. This is in line with previous data indicating a pro-

n. A. NG2 (upper panels), O4 (central panels) and MAG (lower panels) immunolabeling), in the presence of the A2A agonist CGS21680 (CGS: central panels) alone or in com-ed with DAPI (blue). Scale bar: 50 mm. B. Time course of NG2 (left panel), O4 (centralditions, in the presence of 100 nM CGS21680 alone or in combination with 100 nMcounted for each coverslip. Two coverslips for any given experimental condition were

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310308

differentiating effect of DMSO in other cell types (Woodbury et al.,2002, 2000).

Treatment of OL cultures with CGS21680, CGS21680 þSCH58261 or DMSO alone did not affect cell viability, as total cellnumber per coverslip was not different at any time tested in eachexperimental condition (see Table 1).

Finally, we investigated whether A2A receptor activation alsoinfluences in vitro OPC proliferation. Dividing cells in OL cultures,grown in the absence or in the presence of the selective A2Aadenosine receptor agonist CGS21680, were visualized by BrdUincorporation. We found that A2A receptor activation did not affectOPC cell cycle since the percentage of BrdUþ cells was not differentin OL cultures grown for 3 days in the presence of 100 nMCGS21680 in comparison to control conditions (62.2 � 5.1% BrdUþ

cells in the CGS21680 group vs. 69.5 � 6.1% in control conditions,Supplemental Information Fig. 2).

4. Discussion

The main findings of our work are that adenosine A2A receptorsinhibit IK currents and decrease cell differentiation in purified pri-mary OPC cultures.

To our knowledge, the present work is the first functionalcharacterization of adenosine A2A receptors in oligodendrocytecells.

As already stated (Soliven et al., 1988; Barres et al., 1990;Sontheimer et al., 1989), electrophysiological data in our studyconfirm that undifferentiated OPCs express high densities of out-ward rectifying, sustained (IK) and transient (IA) Kþ currents whoseamplitude decreases along with cell maturation, in parallel with anincrease in inwardly rectifying Kþ (Kir) conductances. In line withprevious data, (Karadottir et al., 2008; Kettenmann et al., 1991), apopulation of undifferentiated OPCs also express fast inward, TTX-sensitive, INa which is lost after a few days of differentiation. In thepresent work we demonstrated that selective activation of adeno-sine A2A receptors by CGS21680 inhibits delayed rectifier, sustained,IK currents in cultured OPCs without affecting transient IA con-ductances. The effect of CGS21680 was observed in all cells tested,independently from cell morphology or time in culture, demon-strating that functional A2A receptors are found throughout all stepsof oligodendrocyte maturation: from bipolar, proliferating, NG2þ

progenitors to mature, myelinating MAGþ cells. CGS21680 inhibi-tion of IK currents is concentration-dependent, with an EC50 in thelow nanomolar range (which is in line with values reported in theliterature: see Jarvis et al., 1989; Fredholm et al., 2011), and iscompletely prevented in the presence of the selective A2A antago-nist SCH58261, thus confirming a receptor-mediated effect.Furthermore, TEA, which blocks IK but not IA currents in ourexperimental conditions, mimics and occludes the effect of the A2Aagonist, confirming a selective modulation of this purinergic

Table 1Treatment of OL cultures with CGS21680, SCH58261 or DMSO does not influence cellviability.

Cell number/microscopicfield

t0 t3 t6 t9

Control 89.8 � 18.6 81.6 � 21 77.6 � 12.6 62.9 � 11.1CGS21680 71.9 � 14.2 88.1 � 7.2 57.4 � 10.2 87.7 � 26.3CGS21680 þ SCH58261 75.1 � 15.7 76.3 � 9 59.4 � 17.1 59.4 � 17.1DMSO 67.5 � 16.2 87.4 � 4.3 64 � 6.2 66.9 � 6.3

Total cell number in each microscopic field (20�) was evaluated by counting DAPIþ

nuclei in OL cultures grown in different experimental conditions. No significantdifferences were found in total cell number per microscopic field among experi-mental groups (One-way ANOVA, NewmaneKeuls post-test). Ten microscopic fieldswere counted for each coverslip. Two coverslips for any given experimental con-dition were evaluated. Experiments were performed at least in triplicate.

receptor subtype on IK currents. Similar effects of A2A receptors onoutward rectifying Kþ channels have already been described inother cell types (Xu and Enyeart, 1999; Duffy et al., 2007), with aninvolvement of either intracellular cAMP rise or a direct action ofthe Gs protein coupled to A2A receptors being hypothesized. Ofnote, we have evidence that the activation of GPR17, a recentlydeorphanized Gi-coupled P2Y-like receptor, elicits opposite effectsin comparison to the Gs coupled A2A subtype, thus increasing theamplitude of IK currents recorded from cultured OPCs (Coppi et al.,2013). Increased amplitude of IK currents is obtained also by se-lective stimulation of adenosine, Gi-coupled, A1 receptors (personalunpublished data). Taken together, these data suggest that theintracellular signaling pathway leading to IK modulation in OLcultures is negatively coupled to cAMP.

It is known that purines, in addition to their functions as neu-rotransmitters and neuromodulators, can also act as growth andtrophic factors, influencing the development of neuronal (Migitaet al., 2008; Mishra et al., 2006) and glial (Stevens et al., 2002;Stevens and Fields, 2000) cells. In particular, a key role of adeno-sine in oligodendrocyte maturation has been recognized(Burnstock, 2011). As already mentioned, data in the literaturereport that adenosine, either exogenously applied or released byelectrical stimulation of co-cultured DRG neurons, reduces in vitroOPC proliferation and promotes their differentiation into myeli-nating OLGs through A1 receptor stimulation (Stevens et al., 2002).In the present work we demonstrate that adenosine A2A receptorsinhibit in vitro OPC differentiation since their stimulation byCGS21680 increases the percentage of NG2þ immature OPCs andreduces O4þ pre-OLGs and MAGþ mature OLGs without affectingcell viability. This effect is completely prevented in the presence ofthe selective A2A receptor blocker SCH58261. In keeping with datademonstrating that the inhibition of IK currents impairs prolifera-tion and maturation of cultured OPCs (Gallo et al., 1996; Chittajalluet al., 2002; Vautier et al., 2004; Ghiani et al., 1999; Attali et al.,1997) and blocks myelin deposition in embryonic spinal cord(Shrager and Novakovic, 1995), we hypothesize that A2A receptorstimulation inhibits OPC differentiation by reducing IK currents. Inline with this assumption is the notion that adenosine A1 receptors,which enhances IK currents in OPCs, exert a pro-differentiatingeffect in OL cultures (Stevens et al., 2002). Thus, it appears thatadenosine acts as a dual modulator of OPC development by stim-ulating, through A1 receptors (Stevens et al., 2002), and inhibiting,by A2A activation (the present work), oligodendrocyte maturation.However, we cannot exclude that other intracellular pathways, inaddition to IK current block, might contribute to the A2A receptormediated inhibition of OPC differentiation.

Differently from TEA which inhibits in vitro OPC proliferation(Gallo et al., 1996), CGS21680 does not influence OPC cycling ratesince the percentage of BrdUþ cells was unchanged when OPCswere grown for 3 days in the presence or in the absence of the A2Aagonist. In considering data demonstrating that IK blockade byCGS21680 does not inhibit OPC proliferation, it could be argued thatselective stimulation of A2A adenosine receptor exerts multipleeffects, such as cAMP increase and CREB phosphorylation (Cabezaet al., 2007; Mori et al., 2004). It is known that CREB activationpromotes OPC proliferation (Lin et al., 2005). So, we could hy-pothesize that the lack of inhibitory effects of CGS21680 on OPCproliferation results from the sum of opposite effects: i.e. IKblockade and CREB activation.

The inhibitory effect of adenosine on OPC maturation might beof relevance during demyelinating pathologies, such as MS,experimental autoimmune encephalomyelitis (EAE: an animalmodel for MS) or ischemia. In a mouse model of EAE, adenosine A2Aantagonists protect from disease development (Mills et al., 2012),suggesting that the activation of adenosine A2A receptors on

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310 309

neuronal and glial cells is necessary for the EAE development inmice. In light of these data, it can be postulated that, underdemyelinating conditions, A2A receptor-mediated inhibition of OPCmaturation is associated with an increased damage, since thestimulation of this receptor subtype prevents myelin deposition.

Of note, such a role of adenosine A2A receptor is in contrast withthe observation that genetic ablation of A2A receptors exacerbatesbrain damage and neuroinflammation in a mouse model of EAE(Yao et al., 2012). In order to explain these data, we might hy-pothesize that adenosine A2A receptors, by inhibiting cell differ-entiation, contribute to amplify the pool of proliferating OPCs in thelesioned areas by maintaining these cells in a predifferentiatedstate available to respond to myelin-inducing signals in repairingmechanisms, accordingly to what was proposed by Stevens andFields about ATP effects on Schwann cells (Stevens and Fields,2000). So, it appears that A2A receptor activation in demyelin-ating conditions can be associated with both protective and detri-mental effects, and the final outcome likely depends on timing ofreceptor stimulation during the development of pathology.

5. Conclusion

The present results, in conclusion, indicate that adenosine A2Areceptor stimulation reduces IK currents and, by this mechanism,inhibits cell differentiation in purified primary OPC cultures, thussuggesting that this receptor subtype plays an important role inoligodendrocyte development and myelin formation. Brain A2Areceptors may therefore represent a newmolecular target for drugsuseful in demyelinating pathologies, such as MS, stroke and braintrauma.

Disclosure

We confirm that we have read the Journal’s position on issuesinvolved in ethical publication and affirm that this report isconsistent with these guidelines. All the authors have no conflictsof interest.

Acknowledgments

This research was supported by the University of Florence andPRIN-COFIN 2008 (FP).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.neuropharm.2013.05.035.

References

Agresti, C., Meomartini, M.E., Amadio, S., Ambrosini, E., Serafini, B., Franchini, L.,Volonte, C., Aloisi, F., Visentin, S., 2005. Metabotropic P2 receptor activationregulates oligodendrocyte progenitor migration and development. Glia 50,132e144.

Attali, B., Wang, N., Kolot, A., Sobko, A., Cherepanov, V., Soliven, B., 1997. Charac-terization of delayed rectifier Kv channels in oligodendrocytes and progenitorcells. J. Neurosci. 17, 8234e8245.

Barres, B.A., Koroshetz, W.J., Swartz, K.J., Chun, L.L., Corey, D.P., 1990. Ion channelexpression by white matter glia: the O-2A glial progenitor cell. Neuron 4, 507e524.

Berger, T., Schnitzer, J., Orkand, P.M., Kettenmann, H., 1992. Sodium and calciumcurrents in glial cells of the mouse corpus callosum slice. Eur. J. Neurosci. 4,1271e1284.

Blankenfeld, G.V., Verkhratsky, A.N., Kettenmann, H., 1992. Ca2þ channel expressionin the oligodendrocyte lineage. Eur. J. Neurosci. 4, 1035e1048.

Burnstock, G., 2011. Introductory overview of purinergic signalling. Front Biosci.(Elite Ed.) 3, 896e900.

Cabeza, d.V., Kannan, P., Pan, Y., Jiang, N., Sun, Y., Carr, K.D., 2007. The adenosine A2Areceptor agonist, CGS-21680, blocks excessive rearing, acquisition of wheel

running, and increases nucleus accumbens CREB phosphorylation in chronicallyfood-restricted rats. Brain Res. 1142, 100e109.

Chang, A., Nishiyama, A., Peterson, J., Prineas, J., Trapp, B.D., 2000. NG2-positiveoligodendrocyte progenitor cells in adult human brain and multiple sclerosislesions. J. Neurosci. 20, 6404e6412.

Chittajallu, R., Chen, Y., Wang, H., Yuan, X., Ghiani, C.A., Heckman, T., McBain, C.J.,Gallo, V., 2002. Regulation of Kv1 subunit expression in oligodendrocyte pro-genitor cells and their role in G1/S phase progression of the cell cycle. Proc. Natl.Acad. Sci. U. S. A. 99, 2350e2355.

Coppi, E., Maraula, G., Fumagalli, M., Failli, P., Cellai, L., Bonfanti, E., Mazzoni, L.,Coppini, R., Abbracchio, M.P., Pedata, F., Pugliese, A.M., 2013. UDP-glucose en-hances outward Kþ currents necessary for cell differentiation and stimulatescell migration by activating the GPR17 receptor in oligodendrocyte precursors.Glia. http://dx.doi.org/10.1002/glia.22506 (in press).

Dawson, M.R., Polito, A., Levine, J.M., Reynolds, R., 2003. NG2-expressing glialprogenitor cells: an abundant and widespread population of cycling cells in theadult rat CNS. Mol. Cell Neurosci. 24, 476e488.

De Biase, L.M., Nishiyama, A., Bergles, D.E., 2010. Excitability and synapticcommunication within the oligodendrocyte lineage. J. Neurosci. 30, 3600e3611.

Duffy, S.M., Cruse, G., Brightling, C.E., Bradding, P., 2007. Adenosine closes the Kþ

channel KCa3.1 in human lung mast cells and inhibits their migration via theadenosine A2A receptor. Eur. J. Immunol. 37, 1653e1662.

Fields, R.D., Burnstock, G., 2006. Purinergic signalling in neuron-glia interactions.Nat. Rev. Neurosci. 7, 423e436.

Fredholm, B.B., IJzerman, A.P., Jacobson, K.A., Linden, J., Muller, C.E., 2011. Interna-tional Union of Basic and Clinical Pharmacology. LXXXI. Nomenclature andclassification of adenosine receptorsean update. Pharmacol. Rev. 63, 1e34.

Fumagalli, M., Daniele, S., Lecca, D., Lee, P.R., Parravicini, C., Fields, R.D., Rosa, P.,Antonucci, F., Verderio, C., Trincavelli, M.L., Bramanti, P., Martini, C.,Abbracchio, M.P., 2011. Phenotypic changes, signaling pathway, and functionalcorrelates of GPR17-expressing neural precursor cells during oligodendrocytedifferentiation. J. Biol. Chem. 286, 10593e10604.

Gallo, V., Zhou, J.M., McBain, C.J., Wright, P., Knutson, P.L., Armstrong, R.C., 1996.Oligodendrocyte progenitor cell proliferation and lineage progression areregulated by glutamate receptor-mediated Kþ channel block. J. Neurosci. 16,2659e2670.

Ghiani, C.A., Yuan, X., Eisen, A.M., Knutson, P.L., DePinho, R.A., McBain, C.J., Gallo, V.,1999. Voltage-activated Kþ channels and membrane depolarization regulateaccumulation of the cyclin-dependent kinase inhibitors p27(Kip1) andp21(CIP1) in glial progenitor cells. J. Neurosci. 19, 5380e5392.

Hibino, H., Inanobe, A., Furutani, K., Murakami, S., Findlay, I., Kurachi, Y., 2010.Inwardly rectifying potassium channels: their structure, function, and physio-logical roles. Physiol. Rev. 90, 291e366.

Horner, P.J., Power, A.E., Kempermann, G., Kuhn, H.G., Palmer, T.D., Winkler, J.,Thal, L.J., Gage, F.H., 2000. Proliferation and differentiation of progenitor cellsthroughout the intact adult rat spinal cord. J. Neurosci. 20, 2218e2228.

Jarvis, M.F., Schulz, R., Hutchison, A.J., Do, U.H., Sills, M.A., Williams, M., 1989. [3H]CGS 21680, a selective A2 adenosine receptor agonist directly labels A2 re-ceptors in rat brain. J. Pharmacol. Exp. Ther. 251, 888e893.

Karadottir, R., Hamilton, N.B., Bakiri, Y., Attwell, D., 2008. Spiking and nonspikingclasses of oligodendrocyte precursor glia in CNS white matter. Nat. Neurosci. 11,450e456.

Kettenmann, H., Blankenfeld, G.V., Trotter, J., 1991. Physiological properties of oli-godendrocytes during development. Ann. N. Y. Acad. Sci. 633, 64e77.

Knutson, P., Ghiani, C.A., Zhou, J.M., Gallo, V., McBain, C.J., 1997. Kþ channelexpression and cell proliferation are regulated by intracellular sodium andmembrane depolarization in oligodendrocyte progenitor cells. J. Neurosci. 17,2669e2682.

Levine, J.M., Reynolds, R., 1999. Activation and proliferation of endogenous oligo-dendrocyte precursor cells during ethidium bromide-induced demyelination.Exp. Neurol. 160, 333e347.

Lin, X., Ramamurthy, S.K., Le Breton, G.C., 2005. Thromboxane A receptor-mediatedcell proliferation, survival and gene expression in oligodendrocytes.J. Neurochem. 93, 257e268.

Mann, S.A., Versmold, B., Marx, R., Stahlhofen, S., Dietzel, I.D., Heumann, R., Berger, R.,2008. Corticosteroids reverse cytokine-induced block of survival and differenti-ation of oligodendrocyte progenitor cells from rats. J. Neuroinflammation 5, 39.

Migita, H., Kominami, K., Higashida, M., Maruyama, R., Tuchida, N., McDonald, F.,Shimada, F., Sakurada, K., 2008. Activation of adenosine A1 receptor-inducedneural stem cell proliferation via MEK/ERK and Akt signaling pathways.J. Neurosci. Res. 86, 2820e2828.

Mills, J.H., Kim, D.G., Krenz, A., Chen, J.F., Bynoe, M.S., 2012. A2A adenosine receptorsignaling in lymphocytes and the central nervous system regulates inflamma-tion during experimental autoimmune encephalomyelitis. J. Immunol. 188,5713e5722.

Mishra, S.K., Braun, N., Shukla, V., Fullgrabe, M., Schomerus, C., Korf, H.W.,Gachet, C., Ikehara, Y., Sevigny, J., Robson, S.C., Zimmermann, H., 2006. Extra-cellular nucleotide signaling in adult neural stem cells: synergism with growthfactor-mediated cellular proliferation. Development 133, 675e684.

Mori, Y., Higuchi, M., Masuyama, N., Gotoh, Y., 2004. Adenosine A2A receptor fa-cilitates calcium-dependent protein secretion through the activation of proteinkinase A and phosphatidylinositol-3 kinase in PC12 cells. Cell Struct. Funct. 29,101e110.

Nishiyama, A., Komitova, M., Suzuki, R., Zhu, X., 2009. Polydendrocytes (NG2 cells):multifunctional cells with lineage plasticity. Nat. Rev. Neurosci. 10, 9e22.

E. Coppi et al. / Neuropharmacology 73 (2013) 301e310310

Othman, T., Yan, H., Rivkees, S.A., 2003. Oligodendrocytes express functional A1adenosine receptors that stimulate cellular migration. Glia 44, 166e172.

Shrager, P., Novakovic, S.D., 1995. Control of myelination, axonal growth, and syn-apse formation in spinal cord explants by ion channels and electrical activity.Brain Res. Dev. Brain Res. 88, 68e78.

Soliven, B., Szuchet, S., Arnason, B.G., Nelson, D.J., 1988. Voltage-gated potassiumcurrents in cultured ovine oligodendrocytes. J. Neurosci. 8, 2131e2141.

Soliven, B., Szuchet, S., Arnason, B.G., Nelson, D.J., 1989. Expression and modulationof Kþ currents in oligodendrocytes: possible role in myelinogenesis. Dev.Neurosci. 11, 118e131.

Sontheimer, H., Kettenmann, H., 1988. Heterogeneity of potassium currents incultured oligodendrocytes. Glia 1, 415e420.

Sontheimer, H., Trotter, J., Schachner, M., Kettenmann, H., 1989. Channel expressioncorrelates with differentiation stage during the development of oligodendro-cytes from their precursor cells in culture. Neuron 2, 1135e1145.

Stevens, B., Fields, R.D., 2000. Response of Schwann cells to action potentials indevelopment. Science 287, 2267e2271.

Stevens, B., Porta, S., Haak, L.L., Gallo, V., Fields, R.D., 2002. Adenosine: a neuron-glial transmitter promoting myelination in the CNS in response to action po-tentials. Neuron 36, 855e868.

Vautier, F., Belachew, S., Chittajallu, R., Gallo, V., 2004. Shaker-type potassiumchannel subunits differentially control oligodendrocyte progenitor prolifera-tion. Glia 48, 337e345.

Verkhratsky, A.N., Trotter, J., Kettenmann, H., 1990. Cultured glial precursor cellsfrom mouse cortex express two types of calcium currents. Neurosci. Lett. 112,194e198.

Williamson, A.V., Compston, D.A., Randall, A.D., 1997. Analysis of the ion channelcomplement of the rat oligodendrocyte progenitor in a commonly studiedin vitro preparation. Eur. J. Neurosci. 9, 706e720.

Woodbury, D., Schwarz, E.J., Prockop, D.J., Black, I.B., 2000. Adult rat and humanbone marrow stromal cells differentiate into neurons. J. Neurosci. Res. 61, 364e370.

Woodbury, D., Reynolds, K., Black, I.B., 2002. Adult bone marrow stromal stem cellsexpress germline, ectodermal, endodermal, and mesodermal genes prior toneurogenesis. J. Neurosci. Res. 69, 908e917.

Xu, L., Enyeart, J.J., 1999. Adenosine inhibits a non-inactivating Kþ current in bovineadrenal cortical cells by activation of multiple P1 receptors. J. Physiol. 521 (Pt 1),81e97.

Yao, S.Q., Li, Z.Z., Huang, Q.Y., Li, F., Wang, Z.W., Augusto, E., He, J.C., Wang, X.T.,Chen, J.F., Zheng, R.Y., 2012. Genetic inactivation of the adenosine A(2A) re-ceptor exacerbates brain damage in mice with experimental autoimmuneencephalomyelitis. J. Neurochem. 123, 100e112.

Zocchi, C., Ongini, E., Conti, A., Monopoli, A., Negretti, A., Baraldi, P.G., Dionisotti, S.,1996. The non-xanthine heterocyclic compound SCH 58261 is a new potentand selective A2A adenosine receptor antagonist. J. Pharmacol. Exp. Ther. 276,398e404.