R

UdG

ZTa

b

c

d

H

ARRAA

KPASMEE

Em

(

T

f

0h

Brain Research Bulletin 97 (2013) 16– 23

Contents lists available at SciVerse ScienceDirect

Brain Research Bulletin

jo ur nal home p age: www.elsev ier .com/ locate /bra inresbul l

esearch report

ridine modulates neuronal activity and inhibits spike-waveischarges of absence epileptic Long Evans and Wistar Albinolaxo/Rijswijk rats

solt Kovácsa,∗, Andrea Sléziab,1, Zsolt Kristóf Bali c,2, Péter Kovácsc,3, Arpád Dobolyid,4,amás Szikrab,5, István Hernádic,6, Gábor Juhászb,7

Department of Zoology, University of West Hungary, Savaria Campus, Károlyi Gáspár tér 4, Szombathely 9700, HungaryLaboratory of Proteomics, Eötvös Loránd University, Pázmány Péter sétány 1C, Budapest 1117, HungaryDepartment of Experimental Zoology and Neurobiology, Institute of Biology, University of Pécs, Ifjúság u. 6, Pécs 7624, HungaryNeuromorphological and Neuroendocrine Research Laboratory, Department of Anatomy, Histology and Embryology, Semmelweis University and theungarian Academy of Sciences, Tuzoltó u. 58, Budapest 1094, Hungary

a r t i c l e i n f o

rticle history:eceived 2 August 2012eceived in revised form 20 April 2013ccepted 6 May 2013vailable online xxx

eywords:yrimidine nucleosidebsence epileptic ratspike-wave dischargesultibarrel microiontophoresis

xtracellular neuronal activity

a b s t r a c t

Pharmacological and functional data suggest the existence of uridine (Urd) receptors in the cen-tral nervous system (CNS). In the present study, simultaneous extracellular single unit recording andmicroiontophoretic injection of the pyrimidine nucleoside Urd was used to provide evidence for thepresence of Urd-sensitive neurons in the thalamus and the cerebral cortex of Long Evans rats. Twenty-two neurons in the thalamus (24% of recorded neurons) and 17 neurons in the cortex (55%) responded tothe direct iontophoresis of Urd. The majority of Urd-sensitive neurons in the thalamus and cortex (82%and 59%, respectively) increased their firing rate in response to Urd. In contrary, adenosine (Ado) anduridine 5′-triphosphate (UTP) decreased the firing rate of all responding neurons in the thalamus, andthe majority of responding neurons in the cortex (83% and 87%, respectively). Functional relevance ofUrd-sensitive neurons was investigated in spontaneously epileptic freely moving Long Evans and WistarAlbino Glaxo/Rijswijk (WAG/Rij) rats. Intraperitoneal (i.p.) injection of 500 mg/kg Urd decreased epileptic

EG activity (210–270 min after injection) in both rat strains. Intraperitoneal administration of 1000 mg/kgUrd decreased the number of spike-wave discharges (SWDs) between 150–270 min and 90–270 minin Long Evans and WAG/Rij rats, respectively. The effect of Urd was long-lasting in both rat strains asthe higher dose significantly decreased the number of SWDs even 24 h after Urd injection. The present

results suggest that Urd-sensitive neurons in the thalamus and the cerebral cortex may play a role in the antiepileptic action of Urd pos

Abbreviations: 3-AP, 3-aminopyridine; Ado, adenosine; ANOVA, analysis of variance;

EG, electroencephalogram; FFT, Fast Fourier Transform; GABA, gamma-aminobutyric acean; SWD, spike-wave discharge; UDP, uridine 5′-diphosphate; Urd, uridine; UTP, uridi∗ Corresponding author. Tel.: +36 94 504 409; fax: +36 94 504 404.

E-mail addresses: zskovacs@ttk.nyme.hu (Z. Kovács), slezia@koki.hu (A. Slézia), ballli@gA. Dobolyi), tamas.szikra@fmi.ch (T. Szikra), hernadi@gamma.ttk.pte.hu (I. Hernádi), gjuh

1 Present address: Laboratory of Thalamus Research, Institute of Experimental Medicinel.: +36 1 210 9400; fax: +36 1 210 9412.2 Tel.: +36 72 503607; fax: +36 72 501517.3 Present address: Department of Neuropharmacology, Gedeon Richter Plc., Budapest,

4 Tel.: +36 1 215 6920/53634; fax: +36 1 218 1612.5 Present address: Neural Circuit Laboratories, Friedrich Miescher Institute for Biomed

ax: +41 61 6973976.6 Tel.: +36 72 503607; fax: +36 72 501517.7 Tel.: +36 1 372 2500; fax: +36 1 381 2204.

361-9230/$ – see front matter © 2013 Published by Elsevier Inc.ttp://dx.doi.org/10.1016/j.brainresbull.2013.05.009

sibly via modulation of thalamocortical neuronal circuits.© 2013 Published by Elsevier Inc.

ATP, adenosine 5′-triphosphate; CNS, central nervous system; d.w., distilled water;id; i.p., intraperitoneal; NMDA, N-methyl-d-aspartate; S.E.M., standard error of thene 5′-triphosphate; WAG/Rij, Wistar Albino Glaxo/Rijswijk.

amma.ttk.pte.hu (Z.K. Bali), Pet.Kovacs@richter.hu (P. Kovács), dobolyi@ana.sote.huasz@dec001.geobio.elte.hu (G. Juhász).

e, Hungarian Academy of Sciences, Budapest, Szigony u. 43, 1083, Hungary.

Gyömroi út 19-21, 1103, Hungary. Tel.: +36 1 505 7331; fax: +36 1 8898400.

ical Research, Maulbeerstrasse 66, 4058 Basel, Switzerland. Tel.: +41 61 3873192;

search

1

lsem2

aFsroB12is2(lo32

itaoAAKIatreU

rakadi2suKTUtteew1a

2

2

om

Z. Kovács et al. / Brain Re

. Introduction

Nucleosides participate in the physiological and pathophysio-ogical mechanisms of the CNS such as memory, sleep, depression,chizophrenia, epilepsy, Alzheimer’s disease and Parkinson’s dis-ase. Therefore, increasing attention has been paid to nucleosideechanisms in the CNS (Burnstock, 2007; Kovacs and Dobolyi,

013).Uridine is an endogenous pyrimidine nucleoside, which serves

s a precursor of UTP synthesis required for RNA synthesis.ollowing the discovery of the neuromodulatory function of adeno-ine 5′-triphosphate (ATP), UTP, and the purine nucleoside Adoeviewed recently (Burnstock, 2007), the specific neuroactive rolef Urd has also been hypothesized (Connolly and Duley, 1999).ased on the sleep-promoting effect of Urd (Borbely and Tobler,989), the existence of a Urd receptor was proposed (Kimura et al.,001). Indeed, Urd was shown to activate fast transmembrane Ca2+

on fluxes in rat brain homogenates and a specific Urd bindingite has also been characterized (Kardos et al., 1999; Kovács et al.,003). We demonstrated an uneven distribution of Urd in the CNSKovacs et al., 2010). Furthermore, elevated levels of extracellu-ar Urd were found in response to depolarizing agents in the brainf anesthetized rats (Dobolyi et al., 1999, 2000) and also during-aminopyridine (3-AP)-induced epileptic seizures (Slézia et al.,004).

In the present study, we aimed to identify Urd-sensitive neuronsn the cerebral cortex and the thalamus. The effect of locally ion-ophoretized Urd was examined on maintained extracellular firingctivity of neurons intrinsic to the cerebral cortex and the thalamusf Long Evans rats. The effects of microiontophoretically applieddo and UTP on neuronal spiking activity were also examined:do is an established neuromodulator nucleoside (Burnstock, 2007;ovacs and Dobolyi, 2013) and uridine could have direct effects, too.

n addition, Urd may be metabolized to UTP; thus Urd may exert itsction also via the G-protein coupled P2Y2, P2Y4 and P2Y6 recep-ors of UTP (Köles et al., 2005; Richardson et al., 2003). In turn,eleased UTP is considered to be a potential source of Urd in thextracellular space (Zimmermann, 1996). Based on these data Ado,rd and UTP were investigated in this study.

Ketamine is a noncompetitive N-methyl-d-aspartate (NMDA)eceptor antagonist, which changes the excitatory/inhibitory bal-nce in brain areas involved in genesis of absence epilepsy. Underetamine treatment absence seizure generators may be inhibitednd, as a consequence, SWDs disappear. Indeed, ketamine mayecrease/abolish absence epileptic activity dose-dependently sim-

larly to other NMDA receptor antagonists (Midzyanovskaya et al.,004; Peeters et al., 1990). Thus, in the first experiment, wheningle-neuron electrophysiological recordings were performednder ketamine anaesthesia (similarly as described previously byovács and Hernádi (2006), spontaneous SWDs were suppressed.herefore, in a second experiment, to test the functionality ofrd-sensitive neurons, the effect of Urd was examined on the spon-

aneous epileptic activity of freely moving Long Evans rats. Besides,o investigate whether i.p. administered Urd evokes strain specificffects on SWDs in Long Evans rats, we tested the antiepilepticffects of Urd also in WAG/Rij rats. Awake, freely moving animalsere injected with two different doses of Urd (500 mg/kg, and

000 mg/kg i.p.) and the effects on spontaneous absence epilepticctivity were analyzed on different time scales.

. Material and methods

.1. Animals

Eight months old adult WAG/Rij male rats (housed at the Department of Zool-gy, University of West Hungary, Savaria Campus, Szombathely, Hungary) and 6–8onths old male hooded Long Evans rats (Charles River Laboratories, Gödöllo,

Bulletin 97 (2013) 16– 23 17

Hungary, housed at the Institute of Biology, University of Pécs, Hungary) were usedin the experiments. Animals were initially kept in groups of 3–4 under standardlaboratory conditions (12:12 h light–dark cycle, light was on from 08.00 AM to08.00 PM), with free access to water and food pellets. Rats were maintained inair-conditioned rooms at 22 ± 2 ◦C and were housed individually after surgery andduring the experiments.

Animal treatment and surgery procedures were carried out according to the localethical rules in accordance with the Hungarian Act of Animal Care and Experimenta-tion (1998. XXVIII. Section 243/1998) in conformity with the regulations for animalexperimentation in the European Communities Council Directive of 24 November1986 (86/609/EEC). All efforts were made to minimize pain and suffering and toreduce the number of animals used.

2.2. Microiontophoresis and extracellular unit recording

2.2.1. Surgical treatmentsElectrophysiological recordings were performed under ketamine anaesthesia

following single i.p. injection of 100 mg/kg ketamine (CP Ketamine, RG, Hungary).Long Evans rats (n = 25) were placed in a stereotaxic apparatus and a hole wasdrilled in the skull. Stereotaxic coordinates for the track of multibarrel micropipettesselected on the basis of the rat brain stereotaxic atlas by Paxinos and Watson (2005)were AP: −3.6 mm; L: 2.5 mm; V (from the dura): 1–7 mm.

2.2.2. Single-unit recording and microiontophoresisSeven-barreled micropipettes were used for electrophysiological recording and

microiontophoresis, with tips of 8–10 micrometre in total diameter (Carbostar-7,Kation Scientific Ltd., MN, USA). The impedance of the central, recording chan-nel was 0.4–0.8 M� (at 50 Hz), whereas the impedance for each drug channel was20–200 M�. One drug channel (filled with 0.5 M NaCl) was used for the applicationof a continuous balancing current, while each of the remaining pipettes were filledwith one of the following five bioactive substances: kainate (Sigma, 60 mM, dis-solved in distilled water (d.w.)); gamma-aminobutyric acid (GABA; Sigma, 500 mM,dissolved in d.w.); Ado (Sigma, 250 mM, dissolved in 50% dimethyl sulfoxide 50%d.w.); UTP (Sigma, 125 mM, dissolved in d.w.); and Urd (Sigma, 100 mM, dissolved ind.w.), respectively. Ejection currents were controlled by individual constant currentcircuits (Neurophore BH-2, Medical Systems Corp., NY, USA). Extracellular actionpotentials were passed to a high performance biological amplifier (Supertech Ltd.,Pécs, Hungary), then to an analogue-digital conversion interface (Power 1401, CED,Cambridge, UK) and were stored and archived on PC.

2.2.3. Statistical analysisWaveform data were analyzed with the Spike2 software (CED, Cambridge, UK).

In case of ambiguous neuronal identification, waveform dependent spike-sortingroutines were run off-line. Firing rates of individual neurons were tested before(as control) and during iontophoretic drug administration (treatment). Each neuronwas tested 3–5 times and firing rate data was averaged for each neuron. Data wereexpressed as mean percentage of control firing rate ± standard error of the mean(S.E.M.). Statistics between control and treatment firing activity were performedusing one-way analysis of variance (ANOVA) and Student’s paired and unpairedt-tests, where applicable; p < 0.05 was set as threshold for significance.

2.3. Recording and analysis of absence epileptic activity

2.3.1. Electrode implantationLong Evans (n = 10) and WAG/Rij (n = 10) rats were anaesthetized by

halothane–air mixture (0.8–1%). Stainless steel screw (outer diameter of the thread:0.8 mm) electrodes were placed into the bone above the primary motor cortex (AP:0.8 mm; L: 1.8 mm) and the somatosensory cortex (AP: 0.2 mm; L: 6.2 mm) based onthe stereotaxic atlas of the rat brain by Paxinos and Watson (2005). A stainless steelreference electrode (a plate of 3 mm × 4 mm) was implanted under the skin and overthe masseter muscle while a screw electrode was placed above the cerebellum asground electrode. All electrodes were soldered to a ten-pin socket and were fixedto the skull bone with acrylic dental cement. Rats were allowed to recover fromsurgery for 2 weeks.

2.3.2. EEG recordingA differential biological amplifier (Bioamp4, Supertech Ltd., Pécs, Hungary) was

connected to a CED 1401 mII data capture and analysis device and the Spike2software (CED, Cambridge, UK) was used for electroencephalogram (EEG) datarecording. To detect SWDs (generated in reverberating thalamocortical neuronalcircuitry and manifested in EEG) primary motor cortex-plate and somatosensorycortex-plate leads were recorded. The bandwidth of the EEG signal filtering was0.53–150 Hz. The analogue signal was digitized at 1 kHz sampling rate and raw EEGdata were stored on a PC for further analysis.

2.3.3. Experimental design and data analysisWe measured the number of SWD occurrences and duration between 30 and

270 min of post-injection time. The first 30 min post injection time was omitted fromanalysis as stress induced by i.p. injection is known to change the frequency of SWDsfor up to 30 min (Kovacs et al., 2006). SWDs were extracted from the raw data files

18 Z. Kovács et al. / Brain Research Bulletin 97 (2013) 16– 23

ed fro

(bpap(

3n1dceoSidCdce

3

3

mis1titrsoo

Fig. 1. Typical SWDs (A1, B1) and their power spectrum (A2 and B2) record

Fig. 1A1 and B1) and EEG time fragments containing SWDs were further analyzedy fast Fourier transform (FFT) as described earlier in Kovacs et al. (2006). The mainroperties of a typical SWD of Long Evans and WAG/Rij rats (Long Evans/WAG/Rij)re as follows: a train of spikes and slow waves starting and ending with spikes,ower spectra 6–9.5/7–11 Hz (Fig. 1A2 and B2) and amplitude 0.1–2.0/0.2–1.0 mVCoenen and Van Luijtelaar, 2003; Polack and Charpier, 2006; Shaw, 2004).

To establish a control SWD level, rats were daily injected with 1 ml saline i.p. for days. Following the three-day control period, half of the rats (n = 5 Long Evans and

= 5 WAG/Rij rats) were i.p. injected by a 500 mg/kg Urd solution (lower dose; in ml saline) and the other half of the rats (n = 2 × 5) received 1000 mg/kg Urd (higherose; in 1 ml saline). At the fifth day, as saline control experiment (post-treatmentontrol day), 1 ml saline was injected i.p. again to examine the possible long lastingffects of Urd on SWDs (compared with control SWD levels). As we, in line withther studies, described previously (Kovacs et al., 2006; Polack and Charpier, 2006;haw, 2004) both the number and the duration of SWDs recorded in control exper-ments varied individually (SWD frequency 2–57 h–1 and 7–49 h−1 whereas SWDuration 1.8–44.5 s and 2.0–36.9 s in Long Evans and WAG/Rij rats, respectively).onsequently, derived EEG results data were expressed in SWD numbers and SWDuration in the percentage of average number and duration of SWDs of three-dayontrol period. Statistical analyses of the data were performed by ANOVA. Data arexpressed as means ± S.E.M.

. Results

.1. Neuronal firing activity in the thalamus and the cortex

Firing rate changes of altogether 40 neocortical and 101 thala-ic neurons were analyzed in this study. All reported neurons were

ontophoretized with at least one test compound. The maintainedpontaneous firing activity of the neurons was 9.7 ± 1.5 Hz (range.0–40.6 Hz) in the cortex, and 16.6 ± 1.7 Hz (range 0.3–66.4 Hz) inhe thalamus. We identified Urd-, Ado- and UTP-sensitive neuronsn the thalamus and the cortex. A neuron was considered responsiveo the iontophoretic treatment when its firing rate changed ±20%

espective to its baseline level and the difference was statisticallyignificant (Student’s t-test, p < 0.05). The schematic presentationf recording sites where neurons responded to the iontophoresisf Urd, Ado, or UTP are depicted in Fig. 2(B) as superimposed on the

m freely moving Long Evans (A) and WAG/Rij (B) rats on the control days.

coronal plane of the stereotaxic brain atlas (Paxinos and Watson,2005).

3.1.1. The effect of Urd on neuronal firing activity in the thalamusand the cortex

Twenty-two of 90 examined neurons (24%) in the thalamusand 17 of 31 neurons (55%) in the cortex responded to the directiontophoresis of Urd. In the thalamus, Urd administration wasinhibitory in 4 out of 90 examined neurons (4%) with mean fir-ing activity decrease to 54.4 ± 1.5% of the baseline level, while in18 neurons (20%) the effect of Urd was excitatory with mean firingactivity increase to 202.8 ± 15.8% (Fig. 2A1, A3 and C). In the cortex,the effect of Urd was inhibitory in 7 out of 31 neurons (23%) withmean firing activity decrease to 73.8 ± 1.4% of the baseline level,while, the effect of Urd was excitatory in 10 neurons (32%) withmean firing activity increase to 162.1 ± 16.4%.

3.1.2. The effect of UTP on the neuronal firing activity in thethalamus and the cortex

In the thalamus, UTP administration was inhibitory in 21 outof 30 examined neurons (70%) with mean firing activity decreaseto 56.9 ± 4.2% of the baseline level (Fig. 2A2), while no excitatoryeffect (i.e., increase of mean firing rate) of UTP was observed onthalamic neurons. In the cortex, the effect of UTP was inhibitoryin 13 out of 27 neurons (48%) with mean firing activity decreaseto 60.5 ± 4.5% of the baseline level, while in 2 neurons (7%) theeffect of UTP was excitatory with mean firing activity increase to150.0 ± 5.2% (Fig. 2C).

3.1.3. The effect of Ado on the neuronal firing activity in thethalamus and the cortex

Iontophoretic administration of Ado caused rapid inhibitionin 8 out of 26 thalamic neurons (31%) with mean firing activitydecrease to 56.8 ± 3.9% of the baseline level (Fig. 2A4). Adenosinedid not increase the mean firing activity of thalamic neurons. In the

Z. Kovács et al. / Brain Research Bulletin 97 (2013) 16– 23 19

Fig. 2. Representative sample of the effect of Urd (A1: excitatory effects on maintained firing activity, A3: inhibitory effects on maintained firing activity), UTP (A2) and Ado(A4) on the firing rate of single thalamic neurons with average amplitude of action potentials and overall effects normalized on the total pool of recorded neurons expressedin mean percentage of control firing rate ± S.E.M. (Part A). Schematic presentation of recording sites superimposed on the representative coronal plane (AP: −3.6 mm) oft he ani decrea( .

c2oot

3a

naqiStrLd

he standard stereotaxic rat brain atlas (Paxinos and Watson, 2005). Collapsing in tndicating the number of cortical and thalamic neurons that showed increased (↑),

Part C). Abbreviations: Ado: adenosine; Urd: uridine; UTP: uridine 5′-triphosphate

ortex, the effect of Ado administration was inhibitory in 10 out of3 neurons (43%) with mean firing activity decrease to 60.2 ± 3.8%f the baseline level, while excitatory effect of Ado was observednly in 2 examined neurons (9%) with mean firing activity increaseo 143.1 ± 17.4% (Fig. 2C).

.2. Effect of Urd on SWD frequency and duration in Long Evansnd WAG/Rij rats

The effects of Urd on spontaneous epileptic activity were sig-ificant in both Long Evans and WAG/Rij rats (Fig. 3). Uridinedministered in a dose of 500 mg/kg significantly decreased the fre-uency of SWD occurrence between 210 and 270 min after injection

n both strains (Fig. 3A1 and B1). On the post-treatment control day,WD frequency returned to the baseline level (Fig. 3A2 and B2). In

urn, the higher dose of Urd significantly decreased SWD occur-ences between 150–270 min and 90–270 min after injection intoong Evans and WAG/Rij rats, respectively (Fig. 3A1 and B1). Theecrease in the number of SWDs was maintained during the whole

terior–posterior dimension resulted in symbol overlap (Part B). Summary diagramsed (↓) or unchanged (0) firing activity after the administration of Urd, UTP or Ado

recording period including the post-treatment control day in bothstrains (Fig. 3A2 and B2).

The average duration of SWDs was not changed by either ofthe doses of Urd injection or on the post treatment control day.However, the total duration of SWDs was significantly decreasedin both Long Evans and WAG/Rij rats. This decrease of ictal activ-ity was observed on the day of i.p. injection of 1000 mg/kg Urd(between 150 and 270 min) and after its post-treatment controlday (between 30 and 270 min) as well (Fig. 3A3, B3, A4 and B4).The lower dose of Urd significantly decreased the total durationof SWD only between 210 and 270 min on the day of i.p. injection(Long Evans rats: 45.1 ± 8.9%, p < 0.05; WAG/Rij rats: 54.6 ± 9.7%,p < 0.05) whereas we did not find any differences in total durationof SWDs on the post-treatment control day (data not shown). Also,we did not observe any apparent adverse behavioural effects in the

treated animals.

Several differences were revealed between the two rat strainsin the effect of i.p. injected Urd on SWDs (Fig. 3). For example,after higher dose of Urd, significant decreasing of SWD number in

20 Z. Kovács et al. / Brain Research Bulletin 97 (2013) 16– 23

Fig. 3. The effect of Urd on SWDs in Long Evans (A) and WAG/Rij (B) rats. The effect of two doses of Urd on SWD frequency in Long Evans rats (A1) and WAG/Rij rats (B1)(lower dose: 500 mg/kg i.p., black columns; higher dose: 1000 mg/kg i.p., grey columns). The changes in SWD frequency on the post-treatment control day (post-tr. cont. day)are also shown (A2 and B2; post-tr. cont. day after lower dose of Urd: stripped columns; post-tr. cont. day after higher dose of Urd: open columns). The effect of 1000 mg/kgi.p. Urd is shown on average SWD duration (A3 and B3; black columns) and on total time (A3 and B3; open columns) on the day of i.p. injection as well as on average duration(A4 and B4; grey columns) and on total time (A4 and B4; stripped columns) of SWDs on post-tr. cont. day in Long Evans rats and WAG/Rij rats. Abbreviations: % (y-axis): EEGresults data expressed in the percentage of average number or duration of SWDs of three-day control period; i.p.: intraperitoneal; post-tr. cont. day: post-treatment controlday; Urd: uridine; *p < 0.05.

search

Wts(

4

4a

ocTsbiwwttiStrn

Utcbostrmbw(moaeeotetite

codhaeaattsit

Z. Kovács et al. / Brain Re

AG/Rij rats started 60 min earlier (from 90 to 270 min) comparedo Long Evans rat (from 150 to 270 min). However, Long Evans ratshowed slight decrease in SWD number between 90 and 150 minFig. 3A1 and B1).

. Discussion

.1. Effect of Urd, UTP, and Ado on neuronal firing rate and onbsence epileptic activity

Local application of Urd, UTP, and Ado affected the firing activityf a considerable percentage of the examined neurons located in theerebral cortex and the thalamus of anaesthetized Long Evans rats.here were, however, marked differences in the actions of the threeubstances. The action of Ado was mainly firing rate decreasing inoth the cerebral cortex activity (in 83% of responding neurons) and

n the thalamus (in 100% of responding neurons). This action of Adoas likely to be mediated by the inhibitory A1 type of Ado receptors,hich represent the dominant receptor type in the neocortex and

he thalamus (Burnstock, 2007; Kovacs and Dobolyi, 2013). In fact,he present results concerning the inhibitory action of iontophoret-cally applied Ado are in line with previous findings (Perkins andtone, 1980), whereas the present results on UTP and Urd providehe first evidence on the direct action of these substances on neu-onal activity especially in the thalamus where 82% of Urd-sensitiveeurons increased their firing rate to the iontophoresis of Urd.

Intraperitoneal injection of Urd has been shown to increaserd plasma levels, which in turn leads to elevated Urd levels in

he brain (Cansev, 2006), as Urd has (both equilibrative and con-entrative) transporters, through which it can pass through thelood-brain barrier (Redzic et al., 2005). Therefore, a central actionf the i.p. administered Urd is likely to have caused the suppres-ion of epileptic activity in the present study. Nevertheless, theime course of brain Urd concentration changes is not known inesponse to i.p. injections of 500–1000 mg/kg Urd. We previouslyeasured a 4.7-fold increase in the extracellular Urd level of the

rain tissue following an i.p. injection of 500 mg/kg dose Urd, whichas not effective in reducing the 3-AP induced epileptic seizures

Slézia et al., 2004). In contrast, this dose was effective in the currentodel of absence epilepsy suggesting a specific antiepileptic effect

f Urd. This antiepileptic effect was also dose-dependent as thedministration of 1000 mg/kg Urd resulted in a more pronouncedffect compared to 500 mg/kg Urd. However, it is not likely that anlevated level of Urd can be maintained for several hours becausef its rapid turnover in the brain tissue. Therefore, we suggesthat temporarily elevated Urd may have long-lasting antiepilepticffects, which could be mediated by the actions of Urd on synap-ic plasticity. Uridine can reportedly enhance neurite outgrowthn rat pheochromocytoma cells (Pooler et al., 2005) and increasehe amount of presynaptic proteins and dendritic spines (Sakamotot al., 2007; Wurtman et al., 2009) in different brain regions.

Absence epilepsy is a generalized form of epilepsy and may beharacterized by a complete loss of consciousness, with abruptnset and termination which are in relation to appearance andisappearance of synchronous bilateral SWDs recorded in theuman EEG (Snead, 1995). Genetically absence epileptic rats suchs Long Evans rats and WAG/Rij rats are both well established mod-ls of absence epilepsy (Coenen and Van Luijtelaar, 2003; Polacknd Charpier, 2006; Shaw, 2004), as they spontaneously generatebsence epileptic seizures; thus these animals may better reflecthe pathophysiological background of human absence epilepsy (i.e.

hey are better models of human absence epilepsy) than epilepticeizures in experimentally induced models. Spike-wave dischargesn these animals may originate from pathological alterations in thehalamocortical neuronal circuitry generating a hyper-synchronic

Bulletin 97 (2013) 16– 23 21

activity of the thalamocortical and corticothalamic connectionstriggered by the perioral/lateral region of the somatosensory cor-tex (Coenen and Van Luijtelaar, 2003; Snead, 1995). Spontaneouslyepileptic strains may have genetic alterations that lead to alteredcomposition of ion channels and receptors in thalamic as well ascortical neurons. Based on the presence of Urd-sensitive neuronsin the neocortex and the thalamus, these brain regions are the can-didate targets where Urd can exert its effect on SWDs. Althoughthe possibility that rats exhibiting absence discharges have alteredsensitivity to UTP and Urd cannot be excluded we have no reason tosuspect that either. Therefore, we believe that the present results ofthe extracellular single unit experiment obtained in rats is relevantto the finding that Urd affects SWDs in absence epileptic animals.The excitatory nature of Urd on maintained firing activity both inthe cortex and the thalamus is in good agreement with previousfindings that excitatory tone increasing agents, such as GABAer-gic and Ado receptor antagonists, successfully suppress absenceepileptic activity (Ates et al., 2004; Peeters et al., 1989) as do drugsused to treat human absence epilepsy (Chen et al., 2011; Van Rijnet al., 2004). In contrast, agents that increase the inhibitory tone,such as GABAergic agonists, GABA reuptake blockers or Ado, ele-vate spontaneous epileptic activity in Long Evans rats and WAG/Rijrats (Coenen et al., 1995; Ilbay et al., 2001; Peeters et al., 1989).A potential explanation for this phenomenon could reside in thenature of thalamic relay neurons whose SWD evoking burst fir-ing mode is activated by membrane hyperpolarization (Sherman,2001). It has also been suggested that increased excitability of neu-rons in the perioral/lateral region of the somatosensory cortex, aproposed trigger zone of SWDs, could augment absence epilepticactivity (Chen et al., 2011; Coenen and Van Luijtelaar, 2003). There-fore, an alternative explanation of the inhibitory effect of Urd onSWD frequencies is that the cortical neurons that may decreasetheir activity by Urd in this trigger zone play a key role in the initi-ation of SWDs (Snead, 1995). Low-threshold calcium potentials arein relation with the oscillatory activity of thalamocortical neuronsand, as a consequence, with the pathogenesis of absence epilepsyand the appearance of SWDs (Snead, 1995). Initial excitation isneeded to evoke the low-threshold calcium current in thalamicneurons whereas prolonged activation of thalamic neurons maydecrease the likelihood of the appearance of low-threshold calciumpotentials and absence seizures (Snead, 1995); thus both inhibi-tion (by prevention of initial excitation) and excessive activation ofthalamic (thalamocortical) neurons (by limitation of appearance oflow-threshold calcium potentials) may cause decrease in absenceepileptic activity. In addition, both stimulation of cortical inhibitoryinterneurons (which are in connection with pyramidal cells) anddirect inhibition of cortical excitatory pyramidal (corticothalamic)cells may attenuate corticothalamic rhythmicity and decrease thenumber of SWDs (Snead, 1995). Consequently, we can hypoth-esize that the inhibition of cortical pyramidal and/or thalamicneuronal activity as well as stimulation of the inhibitory interneu-ronal activity in the cortex and/or excessive stimulation of thalamicneuronal activity by Urd may decrease absence epileptic activity.These results also suggest that inhibitory and excitatory effects ofUrd on both cortical and thalamic neurons may be involved in itsantiepileptic effects at least in animal models of absence epilepsy.However, the present in vivo technique does not enable directaccess to information on the neuronal types that were excited orinhibited by Urd. Therefore, a complex and cell-type specific effectof Urd cannot be excluded. In addition, we also cannot rule outthe possibility that brain areas, other than the cerebral cortex andthe thalamus investigated in the present study might also play a

role in epileptogenesis or modulation of epileptic seizures (Snead,1995), and may also contain Urd-sensitive neurons. To reveal neu-ron type(s) and brain areas which are involved in Urd evokeddecrease in absence epileptic activity, direct investigation of the

2 search

rhoi2

tKgteowCeb2vn(iaimUabf1rilm

4

daefisEoviobia

aalIs(o(o1oeaiU

2 Z. Kovács et al. / Brain Re

eceptor effect of Urd would be needed. However, no Urd receptorsave been identified so far and their existence has been suggestednly on the basis on certain sporadic functional and pharmacolog-cal data (Dobolyi et al., 1999; Kardos et al., 1999; Kimura et al.,001; Kovács et al., 2003).

To date, several authors postulated that Urd may bind to a puta-ive Urd receptor and/or the GABAA receptor (Dobolyi et al., 1999;ardos et al., 1999; Kimura et al., 2001; Kovács et al., 2003) sug-esting that (i) the GABAA receptor may be a potential candidateo mediate the direct inhibitory action of Urd and (ii) the exist-nce of an excitatory Urd receptor. The direct increasing effectf Urd on the neuronal activity in the present study is consistentith previous findings reporting that Urd application evokes fasta2+ entry in membrane preparations (Kardos et al., 1999). Nev-rtheless, inhibitory effect of Urd may be mediated via GABA andenzodiazepine binding sites of the GABAA receptor (Kimura et al.,001). In addition, increased Urd concentrations could lead to ele-ated UTP levels (Richardson et al., 2003). Since UTP can modify theeuronal activity through its neuronal plasma membrane receptorsBrunschweiger and Muller, 2006), an indirect action of Urd via UTPs also conceivable. However, as UTP had no major observed excit-tory action on unit activity (less than 6% of UTP-sensitive neuronsncreased their activity measured both in the cortex and the thala-

us), the present results argue against the mediation of excitatoryrd actions via UTP and its receptors. In contrast, the inhibitoryction of Urd on mean firing activity of several neurons could welle mediated by UTP or uridine 5′-diphosphate (UDP), which isormed from UTP by ecto-nucleotidase enzymes (Zimmermann,996). Furthermore, a functional relevance of Urd-sensitive neu-ons is plausible as elevated extracellular Urd levels can be observedn response to depolarizing agents (Dobolyi et al., 2000). Neverthe-ess, our knowledge is not sufficient at present to explain the exact

echanisms by which Urd exerts its effects on neuronal activity.

.2. Uridine as a potential antiepileptic agent

The anticonvulsant effect of Urd was demonstrated a fewecades ago and provided controversial results in experimentalnimal models of epilepsy (Dwivedi and Harbison, 1975; Robertst al., 1974; Zhao et al., 2006). However, the present study is therst to investigate the effect of Urd on absence epilepsy: we demon-trated an inhibitory effect of Urd on SWD frequency in both Longvans and WAG/Rij rats. The appearance and duration of the effectf i.p. administered Urd was dose-dependent. In turn, we foundery little difference in the effects of Urd between the two rat strainsnvestigated. These results suggest that the decreasing effect of Urdn absence epileptic activity is not strain specific, and Urd maye beneficial in alleviating the severity of absence epileptic activ-

ty. Consequently, Urd has the strong potential to be tested as anntiepileptic agent in absence epilepsy.

Uridine, as a natural endogenous molecule, is attractive for ther-peutic use because of its low toxicity (Van Groeningen et al., 1991)nd because in orally administered formulation it can increase Urdevels both in the plasma and in the brain (Wurtman et al., 2000).n fact, different doses of Urd have already been tested in humantudies in order to alleviate the side-effects of anticancer drugsVan Groeningen et al., 1993) and in a rare developmental dis-rder with elevated 5′-nucleotidase activity and EEG abnormalityPage et al., 1997). High doses (e.g. 1000 mg/kg or similar) of Urd byral administration were well tolerated in humans (e.g. Page et al.,997; Sutinen et al., 2007) but we have no data on the tolerabilityf parenterally injected 1000 mg/kg dose Urd in humans. To find an

ffective and well tolerable dose of Urd as a putative antiepilepticgent and to reveal its exact effect on epileptic activity further stud-es are needed including the application of (i) different methods ofrd administration, (ii) various Urd analogues (Kimura et al., 2001)

Bulletin 97 (2013) 16– 23

or (iii) food supplements containing nucleosides to raise plasmaUrd levels (e.g. NucleomaxX; Weinberg et al., 2011).

5. Conclusion

The long-lasting effect of a single dose of Urd on the suppres-sion of SWDs in the present study, together with the observationof its fast, direct action at the neuronal level, both suggest that Urdis a neuroactive agent, which may be involved in the regulationof neuronal activity. The identification of the structure of putativeUrd receptors in different brain areas that are involved in absenceepilepsy genesis and the synthesis of specific Urd receptor antag-onists will be able to open up new future avenues for researchon receptor level effects of Urd. We believe that the present datastrengthen the hypotheses that Urd and/or its analogues may bepotential antiepileptic agents in absence epilepsy and the researchmay contribute to generate further studies on the anti-epilepticeffects of Urd and Urd analogues.

Conflict of interest

None of the authors has any conflict of interest to disclose.

Acknowledgements

The European Union and the European Social Fund have pro-vided financial support to the project under the grant agreementno. TÁMOP 4.2.1./B-09/1/KMR-2010-0003 to G. Juhász. This workwas supported by the National Office for Research and Technol-ogy (NKTH): TÁMOP-4.2.2/08/1 to G. Juhász, the NKTH TECH 09 A1Grant to Á. Dobolyi, the National Development Agency of Hungary(under Grant No. TIOP-1.3.1.-07/2-2F-2009-2008) for Z. Kovács, andthe Szentágothai János Research Centre of the University of Pécs forZ. K. Bali and I. Hernádi. Á. Dobolyi is a recipient of the Bolyai JánosAward of the Hungarian Academy of Sciences. We wish to thankTamás Török (NYME SEK) for technical assistance.

References

Ates, N., Sahin, D., Ilbay, G., 2004. Theophylline, a methylxanthine derivative, sup-presses absence epileptic seizures in WAG/Rij rats. Epilepsy and Behavior 5,645–648.

Borbely, A.A., Tobler, I., 1989. Endogenous sleep-promoting substances and sleepregulation. Physiological Reviews 69, 605–670.

Brunschweiger, A., Muller, C.E., 2006. P2 receptors activated by uracil nucleotides-anupdate. Current Medicinal Chemistry 13, 289–312.

Burnstock, G., 2007. Physiology and pathophysiology of purinergic neurotransmis-sion. Physiological Reviews 87, 659–797.

Cansev, M., 2006. Uridine and cytidine in the brain: their transport and utilization.Brain Research Reviews 52, 389–397.

Chen, S.D., Yeh, K.H., Huang, Y.H., Shaw, F.Z., 2011. Effect of intracranial adminis-tration of ethosuximide in rats with spontaneous or pentylenetetrazol-inducedspike-wave discharges. Epilepsia 52, 1311–1318.

Coenen, A.M., Blezer, E.H., Van Luijtelaar, E.L., 1995. Effects of the GABA-uptake inhibitor tiagabine on electroencephalogram, spike-wave discharges andbehaviour of rats. Epilepsy Research 21, 89–94.

Coenen, A.M., Van Luijtelaar, E.L., 2003. Genetic animal models for absence epilepsy:a review of the WAG/Rij strain of rats. Behavior Genetics 33, 635–655.

Connolly, G.P., Duley, J.A., 1999. Uridine and its nucleotides: biological actions,therapeutic potentials. Trends in Pharmacological Sciences 20, 218–225.

Dobolyi, A., Reichart, A., Szikra, T., Nyitrai, G., Kekesi, K.A., Juhasz, G., 2000. Sustaineddepolarisation induces changes in the extracellular concentrations of purine andpyrimidine nucleosides in the rat thalamus. Neurochemistry International 37,71–79.

Dobolyi, A., Szikra, T., Kékesi, A.K., Kovács, Z., Juhász, G., 1999. Uridine is released bydepolarization and inhibits unit activity in the rat hippocampus. Neuroreport10, 3049–3053.

Dwivedi, C., Harbison, R.D., 1975. Anticonvulsant activities of delta-8 and delta-9

tetrahydrocannabinol and uridine. Toxicology and Applied Pharmacology 31,452–458.

Ilbay, G., Sahin, D., Karson, A., Ates, N., 2001. Effects of adenosine administrationon spike-wave discharge frequency in genetically epileptic rats. Clinical andExperimental Pharmacology and Physiology 28, 643–646.

search

K

K

K

K

K

K

K

K

M

P

P

P

P

P

P

P

R

Z. Kovács et al. / Brain Re

ardos, J., Kovács, I., Szarics, E., Kovács, R., Skuban, N., Nyitrai, G., Dobolyi, A., Juhász,G., 1999. Uridine activates fast transmembrane Ca2+ ion fluxes in rat brainhomogenates. Neuroreport 10, 1577–1582.

imura, T., Ho, I.K., Yamamoto, I., 2001. Uridine receptor: discovery and its involve-ment in sleep mechanism. Sleep 24, 251–260.

ovács, I., Lasztóczi, B., Szarics, E., Héja, L., Sági, G., Kardos, J., 2003. Characteri-sation of an uridine-specific binding site in rat cerebrocortical homogenates.Neurochemistry International 43, 101–112.

ovács, P., Hernádi, I., 2006. Yohimbine acts as a putative in vivo alpha2A/D-antagonist in the rat prefrontal cortex. Neuroscience Letters 402, 253–258.

ovacs, Z., Dobolyi, A., 2013. Anatomical distribution of nucleoside system in thehuman brain and implications for therapy. In: Masino, S.A., Boison, D. (Eds.),Adenosine: A Key Link Between Metabolism and Brain Activity. Springer Science,Business Media, New York, pp. 621–656.

ovacs, Z., Dobolyi, A., Juhász, G., Kékesi, K.A., 2010. Nucleoside map of the humancentral nervous system. Neurochemical Research 35, 452–464.

ovacs, Z., Kékesi, K.A., Szilágyi, N., Abrahám, I., Székács, D., Király, N., Papp, E.,Császár, I., Szego, E., Barabás, K., Péterfy, H., Erdei, A., Bártfai, T., Juhász, G., 2006.Facilitation of spike-wave discharge activity by lipopolysaccharides in Wistar

Albino Glaxo/Rijswijk rats. Neuroscience 140, 731–742.öles, L., Furst, S., Illes, P., 2005. P2X and P2Y receptors as possible targets of thera-

peutic manipulations in CNS illnesses. Drug News and Perspectives 18, 85–101.idzyanovskaya, I.S., Salonin, D.V., Bosnyakova, D.Y., Kuznetsova, G.D., van Luijte-

laar, E.L., 2004. The multiple effects of ketamine on electroencephalographicactivity and behavior in WAG/Rij rats. Pharmacology, Biochemistry, and Behav-ior 79, 83–91.

age, T., Yu, A., Fontanesi, J., Nyhan, W.L., 1997. Developmental disorder associ-ated with increased cellular nucleotidase activity. Proceedings of the NationalAcademy of Sciences of the United States of America 94, 11601–11606.

axinos, G., Watson, C., 2005. The rat brain in stereotaxic coordinates. AcademicPress, San Diego.

eeters, B.W., Van Rijn, C.M., Vossen, J.M., Coenen, A.M., 1989. Effects of GABA-ergic agents on spontaneous non-convulsive epilepsy, EEG and behaviour, inthe WAG/RIJ inbred strain of rats. Life Sciences 45, 1171–1176.

eeters, B.W., van Rijn, C.M., Vossen, J.M., Coenen, A.M., 1990. Involvement of NMDAreceptors in non-convulsive epilepsy in WAG/Rij rats. Life Sciences 47, 523–529.

erkins, M.N., Stone, T.W., 1980. Aminophylline and theophylline derivatives asantagonists of neuronal depression by adenosine: a microiontophoretic study.Archives Internationales de Pharmacodynamie et de Thérapie 246, 205–214.

olack, P.O., Charpier, S., 2006. Intracellular activity of cortical and thalamic neu-rones during high-voltage rhythmic spike discharge in Long-Evans rats in vivo.Journal of Physiology 571, 461–476.

ooler, A.M., Guez, D.H., Benedictus, R., Wurtman, R.J., 2005. Uridine enhances neu-rite outgrowth in nerve growth factor-differentiated pheochromocytoma cells.Neuroscience 134, 207–214.

edzic, Z.B., Biringer, J., Barnes, K., Baldwin, S.A., Al-Sarraf, H., Nicola, P.A., Young, J.D.,Cass, C.E., Barrand, M.A., Hladky, S.B., 2005. Polarized distribution of nucleoside

Bulletin 97 (2013) 16– 23 23

transporters in rat brain endothelial and choroid plexus epithelial cells. Journalof Neurochemistry 94, 1420–1426.

Richardson, U.I., Watkins, C.J., Pierre, C., Ulus, I.H., Wurtman, R.J., 2003. Stimulationof CDP-choline synthesis by uridine or cytidine in PC12 rat pheochromocytomacells. Brain Research 971, 161–167.

Roberts, C.A., Kreisman, N.R., Waltman, M., 1974. Uridine anticonvulsant effects:selective control of nucleoside incorporation in experimental epilepsy. Epilepsia15, 479–500.

Sakamoto, T., Cansev, M., Wurtman, R.J., 2007. Oral supplementation with docosa-hexaenoic acid and uridine-5’-monophosphate increases dendritic spine densityin adult gerbil hippocampus. Brain Research 1182, 50–59.

Shaw, F.Z., 2004. Is spontaneous high-voltage rhythmic spike discharge in LongEvans rats an absence-like seizure activity? Journal of Neurophysiology 91,63–77.

Sherman, S.M., 2001. Tonic and burst firing: dual modes of thalamocortical relay.Trends in Neurosciences 24, 122–126.

Slézia, A., Kekesi, A.K., Szikra, T., Papp, A.M., Nagy, K., Szente, M., Magloczky, Z.,Freund, T.F., Juhasz, G., 2004. Uridine release during aminopyridine-inducedepilepsy. Neurobiology of Disease 16, 490–499.

Snead, O.C., 1995. Basic mechanisms of generalized absence seizures. Annals ofNeurology 37, 146–157.

Sutinen, J., Walker, U.A., Sevastianova, K., Klinker, H., Häkkinen, A.M., Ristola,M., Yki-Järvinen, H., 2007. Uridine supplementation for the treatment ofantiretroviral therapy-associated lipoatrophy: a randomized, double-blind,placebo-controlled trial. Antiviral Therapy 12, 97–105.

Van Groeningen, C.J., Peters, G.J., Nadal, J.C., Laurensse, E., Pinedo, H.M., 1991. Clinicaland pharmacologic study of orally administered uridine. Journal of the NationalCancer Institute 83, 437–441.

Van Groeningen, C.J., Peters, G.J., Pinedo, H.M., 1993. Reversal of 5-fluorouracil-induced toxicity by oral administration of uridine. Annals of Oncology 4,317–320.

Van Rijn, C.M., Sun, M.S., Deckers, C.L., Edelbroek, P.M., Keyser, A., Renier, W.,Meinardi, H., 2004. Effects of the combination of valproate and ethosuximideon spike wave discharges in WAG/Rij rats. Epilepsy Research 59, 181–189.

Weinberg, M.E., Roman, M.C., Jacob, P., Wen, M., Cheung, P., Walker, U.A., Mulligan,K., Schambelan, M., 2011. Enhanced uridine bioavailability following adminis-tration of a triacetyluridine-rich nutritional supplement. PLoS ONE 6, e14709.

Wurtman, R.J., Cansev, M., Ulus, I.H., 2009. Synapse formation is enhanced by oraladministration of uridine and DHA, the circulating precursors of brain phos-phatides. Journal of Nutrition, Health and Aging 13, 189–197.

Wurtman, R.J., Regan, M., Ulus, I., Yu, L., 2000. Effect of oral CDP-choline onplasma choline and uridine levels in humans. Biochemical Pharmacology 60,

989–992.

Zhao, Q., Marolewski, A., Rusche, J.R., Holmes, G.L., 2006. Effects of uridine in modelsof epileptogenesis and seizures. Epilepsy Research 70, 73–82.

Zimmermann, H., 1996. Biochemistry, localization and functional roles of ecto-nucleotidases in the nervous system. Progress in Neurobiology 49, 589–618.