ARTICLE IN PRESS

PURKINJE CELL RHYTHMICITY AND SYNCHRONICITY DURING MODULATION OF FAST CEREBELLAR OSCILLATION

L. SERVAIS

* AND G. CHERON

**

aLaboratory of Electrophysiology, Université Mons-Hainaut, Place du Parc 20, 7000 Mons, Belgium

bLaboratory of Movement Neurophysiology and Biomechanics, Uni- versité Libre de Bruxelles, Brussels, Belgium

cLaboratory of Neurophysiology, Université Libre de Bruxelles, Brus- sels, Belgium

Abstract—Fast ( ⬃ 160 Hz) cerebellar oscillation has been re- cently described in different models of ataxic mice, such as mice lacking calcium-binding proteins and in a mouse model of Angelman syndrome. Among them, calretinin– calbindin double knockout mice constitute the best model for evaluat- ing fast oscillations in vivo. The cerebellum of these mice may present long-lasting episodes of very strong and stable local field potential oscillation alternating with the normal non-oscillating state. Spontaneous firing of the Purkinje cells in wild type and double knockout mice largely differs. Indeed, the Purkinje cell firing of the oscillating mutant is character- ized by an increased rate and rhythmicity and by the emer- gence of synchronicity along the parallel fiber axis. To better understand the driving role played by these different param- eters on fast cerebellar oscillation, we simultaneously re- corded Purkinje cells and local field potential during the induction of general anesthesia by ketamine or pentobarbi- tone. Both drugs significantly increased Purkinje cell rhyth- micity in the absence of oscillation, but they did not lead to Purkinje cell synchronization or to the emergence of fast oscillation. During fast oscillation episodes, ketamine abol- ished Purkinje cell synchronicity and inhibited fast oscilla- tion. In contrast, pentobarbitone facilitated fast oscillation, induced and increased Purkinje cell synchronicity.

We propose that fast cerebellar oscillation is due to the synchronous rhythmic firing of Purkinje cell populations and is facilitated by positive feedback whereby the oscillating field further phase-locks recruited Purkinje cells onto the same rhythmic firing pattern. © 2005 Published by Elsevier Ltd on behalf of IBRO.

Key words: cerebellum, oscillation, rhythmicity, synchronic- ity, ketamine, pentobarbitone

The understanding of interactions between neuronal net- work oscillation and single neuronal firing is central to the study of oscillation genesis and function. Local field poten- tial oscillations (LFPO) have been described in many re-

gions of the mammalian brain, such as the thalamus (Ste- riade et al., 1996), cerebellum (Pellerin and Lamarre, 1997), hippocampus (Whittington et al., 1995; Faulkner et al., 1999), and other parts of the cerebral cortex (Steriade et al., 1993; Collins et al., 2001; Grenier et al., 2003). They may support physiological processes (Hartmann and Bower, 1998; Baker et al., 1999), result from pharmaco- logical impregnation (Fisahn et al., 1998) or reflect patho- logical conditions, such as seizures (Khazipov and Holmes, 2003; Grenier et al., 2003) or ataxia (Cheron et al., 2004, 2005).

In the cerebellar cortex, fast LFPO (130 –260 Hz) was first described in mice lacking calretinin and/or calbindin (Cheron et al., 2004), two proteins involved in calcium buffering in granule and Purkinje cells, respectively. A similar fast cerebellar LFPO is also present in a mouse model of Angelman syndrome (Cheron et al., 2005), and in mice lacking parvalbumin (Servais et al., 2005). All these mice with fast cerebellar LFPO (calbindin, calretinin, cal- bindin/calretinin and Ube3a deficient mice) are ataxic (Airaksinen et al., 1997; Schiffmann et al., 1999; Miura et al., 2002). In contrast, wild type (WT) mice do not present cerebellar LFPO suggesting that this oscillation is a cause or a marker of cerebellar dysfunction.

The cerebellum of mice with LFPO may switch be- tween long-lasting periods of fast oscillation and of the normal non-oscillating state. Although the factors leading to the spontaneous emergence of such fast oscillatory episodes are not known, simple and complex spikes of Purkinje cell populations have been demonstrated to be tightly phase-locked to LFPO. In mice with fast oscillation, the simple spikes of Purkinje cells recorded close to the oscillating field present increased rhythmicity, firing rate and synchronicity (Cheron et al., 2004). Different hypoth- eses attempt to explain the emergence of LFPO and Pur- kinje cell synchronicity in mice lacking calcium-binding proteins. First, increased excitability of granule (Gall et al., 2003) and/or Purkinje cells could lead to increased Pur- kinje cell firing rate and to increased rhythmicity and syn- chronicity of these cells, possibly giving rise to fast oscil- lation. Indeed, Stratton et al. (1988) have reported a rela- tionship between Purkinje cell rhythmicity and firing rate.

Fast LFPO would thus be considered as a side effect of increased simple spike firing rate, which would be the primary cause of ataxia and other cerebellar electrophys- iological abnormalities. Another view is that fast oscillation itself would be the primary cause of Purkinje cell synchro- nization in a high rhythmic frequency range. This fast oscillation would then “trap” Purkinje cells, rendering their firing less adaptable, possibly leading to ataxia. These two

*Corresponding author. Tel:

32-65-37-35-66; fax:

32-65-37-35-66.

**Correspondence to: G. Cheron, Laboratory of Electrophysiology, Université Mons-Hainaut, Mons, Belgium. Tel:

32-65-37-35-66; fax:

⫹

32-65-37-35-66.

E-mail addresses: servais.laurent@ulb.ac.be (L. Servais), gcheron@ulb.ac.be (G. Cheron)

Abbreviations:

APV, D,L-2-amino-5-phosphonovaleric acid; Cb

Cr

, mice lacking calbindin and calretinin; LFPO, local field potential oscillation; NMDA, N-methyl-

-aspartate; WT, wild type mice.

Neuroscience xx (2005) xxx

0306-4522/05$30.00⫹0.00 © 2005 Published by Elsevier Ltd on behalf of IBRO.

doi:10.1016/j.neuroscience.2005.06.001

1

mechanisms may coexist. In neocortical fast LFPO, Gre- nier et al. (2003) proposed a vicious feedback loop in which very fast oscillations in field potentials reflect the synchronous action of neocortical neurons and help to generate and synchronize action potentials in adjacent neurons through electrical interactions (Jefferys, 1995).

In order to test these hypotheses, simultaneous re- cordings of fast oscillation and Purkinje cells are required during the emergence, inhibition or facilitation of LFPO.

Cerebellar microinjections (Cheron et al., 2004, 2005; Ser- vais et al., 2005) have demonstrated the involvement of GABA

and N-methyl-

-aspartate (NMDA) receptors in maintaining cerebellar LFPO. However, long-lasting stable multiple recording of single-unit Purkinje cells in alert mice during local microinjections is extremely difficult to per- form. Moreover, the drug concentration at the different recording sites is currently impossible to standardize dur- ing microinjection. Therefore, we studied fast LFPO and Purkinje cell firing before and during general anesthesia induced by pentobarbitone (a GABA

receptor positive modulator) or ketamine (a NMDA receptor antagonist).

Given their modes of action, these substances were ex- pected to respectively facilitate and inhibit fast LFPO. Both drugs are known to diffuse rapidly and homogenously in mouse brain (Saubermann et al., 1974; Blednov and Simp- son, 1999). The aim of this study was to characterize Purkinje cell firing rate, rhythmicity and synchronicity dur- ing the inhibition or the facilitation of LFPO.

EXPERIMENTAL PROCEDURES Mice

Sex- and age- (3– 8 months) matched WT and calretinin/calbindin double knockout mice (Cb

Cr

) (Cheron et al., 2004) gener- ated on a mixed 129 ⫻ C57Bl/6 genetic background were used in all experiments.

Surgical preparation for recording

Mice were anesthetized with xylido-dihydrothiazin (Rompun®, Bayer, Germany, 10 mg/kg) and ketamine (Ketalar®, CEVA, Bel- gium, 100 mg/kg). Animals were administered an additional dose of xylido-dihydrothiazin (3 mg/kg) and ketamine (30 mg/kg) if they presented agitation or markedly increased respiration or heart rate during the procedure. In addition, local anesthesia with 0.5 ml lidocaine 20 mg/ml ⫹ Adrenaline 1:80,000 (Xylocaine®, Astra Zen- eca, UK) was administered s.c. during soft tissue removal. Two small bolts were cemented to the skull to immobilize the head during the recording sessions and a silver reference electrode was placed on the surface of the parietal cortex. An acrylic recording chamber was constructed around a posterior craniostomy, cov- ered by a thin layer of bone wax (Ethicon®, Johnson and Johnson, USA) before the recording sessions. An i.p. catheter was placed and fixed to the skin. Twenty-four hours after anesthesia, alert mice were immobilized for the recording session. The dura mater was removed locally above the vermis. Recordings were per- formed in lobules IV–VIII, and the location of the electrodes (depth and lobule) was noted. We did not select a specific lobule for the different experiments, but during multiple recordings, electrodes were placed in the same lobule.

Single-unit recordings

Single-unit recordings were performed with glass micropipettes filled with NaCl 0.2 M (1.5–5 M ⍀ of impedance). A neural signal was considered as originating from a Purkinje cell if it presented two types of spiking activities: simple spikes characterized by single depolarization (300 – 800 ␮ s) occurring between 20 and 200 Hz and complex spikes characterized by an initial fast depo- larization (300 – 600 ␮ s) followed by smaller and relatively con- stant wavelets. It was considered that simple and complex spikes originated from the same Purkinje cell when a transient pause ( ⬃ 20 ms) in simple spike firing followed each complex spike.

Multi-unit recordings

Multiple recordings along the same parallel fiber axis were per- formed by means of seven linearly arranged, quartz-insulated, platinum-tungsten fiber-microelectrodes (outer and shaft diameter of 80 ␮ m and 25 ␮ m, respectively) with 250- ␮ m inter-electrode spacing. Each microelectrode was mounted into a stretched elas- tic rubber tube enabling proper positioning via DC-micromotors (resolution of 0.27 ␮ m) (Eckhorn and Thomas, 1993).

Drug injection

After the recording of oscillation and/or Purkinje cell activity lasting a minimum of 120 s, ketamine 100 mg/kg, dilution 1/5 with NaCl 0.9% and pentobarbitone (Nembutal®, CEVA) 30 mg/kg, dilution 1/20 with NaCl 0.9% were slowly ( ⬃ 30 s) injected through the i.p.

catheter. No mice underwent more than one injection on the same day and no more than a total of two injections. Recording and analysis were performed until 15 min after the injection unless otherwise mentioned. At the end of the recording session, mice were killed by a lethal dose of pentobarbitone through the i.p.

catheter.

Microinjections

Pentobarbitone (24 mM), ketamine (42 mM) or D,L-2-amino-5- phosphonovaleric acid (APV) (Sigma, France) (a NMDA-specific antagonist) (42 mM) was injected through a micropipette, drawn from calibrated 1.16 mm internal diameter glass tubing (internal diameter: 30 ␮ m), using an air pressure system (Picospritzer II), pulse of 5 ms, five pulses. Each pulse delivered a volume of 0.125 ␮ L. No mice underwent more than one microinjection the same day and more than a total of three microinjections.

Data analysis

All recordings were performed with a bandwidth of 0.01 kHz to 10 kHz. They were stored on 4 mm digital audio tapes (Sony PCM- R500) and transferred to a Pentium III personal computer with analog-to-digital converter boards (Power 1401, CED). The re- corded data were digitized continuously at 10 kHz. Off-line anal- ysis and illustrations were performed using the Spike 2 CED software. The discrimination between Purkinje cell simple and complex spikes was performed by the same software (waveform recognition) and controlled visually before analysis.

The rhythmic frequency was defined as the reciprocal of the latency of the first peak in the autocorrelogram of simple spike firing (width ⫽ 1 s, bin size ⫽ 0.2 ms). Consequently, rhythmic fre- quency could not be determined on flat autocorrelograms. The strength of the rhythmicity was quantified with a rhythm index measured on the simple spike autocorrelogram (120 s-lasting recording, width ⫽ 1 s, bin size ⫽ 1 ms) (Sugihara et al., 1995; Lang et al., 1999). Briefly, the height and depth of all peaks and valleys that were significantly different from the baseline and occurred at specific latencies with regard to the initial peak were summed. The sum was divided by the total number of spikes in the recording. In the autocorrelograms with no significant peaks and valleys, a L. Servais and G. Cheron / Neuroscience xx (2005) xxx

2

ARTICLE IN PRESS

value of zero was given to the rhythm index, and the activity was considered as non-rhythmic. In these cases, or when the rhythm index was less than an empirically determined value of 0.01, the rhythmic frequency was not determined. Thus, we quantified the rhythmicity according to its frequency (by the rhythmic frequency) and to its strength by the rhythm index (the higher the rhythmicity, the higher the rhythm index).

The strength of synchronicity between Purkinje cells was quantified by a synchronicity index, measured on the cross-cor- relogram (2 min recording of Purkinje cells pairs, width ⫽ 1 s, bin size ⫽ 1 ms). This was calculated by dividing the difference be- tween the central peak of the cross-correlogram and the baseline level by the total number of spikes during the same period of time (the higher the synchronicity, the higher the synchronicity index).

The central peak was defined as the highest peak in a 20 ms interval centered on the 0 ms time-bin.

Local field potentials were recorded using a distinct electrode in absence of detectable spiking activity. The local field potentials were analyzed by the wave-triggered averaging technique (Ste- riade et al., 1998). Three hundred successive negative peaks of the LFPO and equal windows around that point (50 ms before and 50 ms after) were extracted and averaged. These averaged os- cillation sequences were quantified by the Fast-Fourier-Transform algorithm. An oscillation index was computed by dividing the maximum amplitude of the power spectrum peak by the total area of the power spectrum (Cheron et al., 2004). Therefore, the oscil- lation was quantified according to its frequency, corresponding to the highest peak of the Fast-Fourier-Transform, and to its strength, by the LFPO index.

We used Student’s t-test for unpaired samples to compare WT and mutant mice values and Student’s t-test for paired sam- ples to compare values before and after drug injection. ANOVA test for repeated measures followed by post hoc comparison was also used to compare values at different times following injection.

Results are expressed and illustrated as mean ⫾ S.D. and are considered significant if P ⬍ 0.05. All statistical analyses were per- formed using Statistica 6.0.

The mice were treated according to international guiding prin- ciples for research involving animals and human beings and guidelines established by the ethical committee of the UMH for the care and use of laboratory animals. Efforts were made to minimize the number of animals used and their suffering.

We first verified the predicted effect of pentobarbitone and ketamine on LFPO. Then, the effect of both drugs on single Purkinje cell firing in the absence of LFPO was characterized.

Finally, we examined the effect of ketamine and pentobarbitone on Purkinje cell synchronicity and on LFPO-Purkinje cell interac- tions.

RESULTS

Effect of ketamine and pentobarbitone on LFPO Ketamine or pentobarbitone was injected 10 min after stable LFPO recording (n ⫽ 5 for each drug) in six Cb

Cr

mice.

Fig. 1 illustrates a LFPO recording and its Fourier spectrum before and 10 min after ketamine (a) or pentobarbitone (b) injection. Ten minutes following the ketamine injection, the peak of the Fourier spectrum is decreased, indicating in- hibition, and shifted toward the left, indicating slowing down of the LFPO (Fig. 1a, c, d). In contrast, 10 minutes following pentobarbitone injection, LFPO is maintained, and even facilitated, its Fourier spectrum being sharper and shifted toward the left (Fig. 1b– d). No significant change in LFPO index or frequency occurred after saline injection (n ⫽ 5) (Fig. 1c, d). Recordings were performed

until spontaneous movement recovery (125 ⫾ 89 min and 104 ⫾ 12 min after ketamine and pentobarbitone injec- tion, respectively). At this time, LFPO index and frequency were not significantly different from the initial values (Fig.

1c, d).

In order to test if the effects were due to a local action on the cerebellar neurons, pentobarbitone (n ⫽ 6) or ket- amine (n ⫽ 6) was locally injected into the cerebellum dur- ing fast LFPO episodes recorded in five Cb

Cr

mice.

One minute following pentobarbitone microinjection, LFPO index increased to 129 ⫾ 23% of its initial value and its frequency decreased to 91 ⫾ 3%. In contrast, one minute after ketamine injection, LFPO index decreased to 62 ⫾ 21% and its frequency to 89 ⫾ 4%. In order to deter- mine if the ketamine action was related to its NMDA an- tagonism or to non-specific effects, APV, a specific NMDA blocker, was locally injected during fast LFPO episodes (n ⫽ 5) in three mice. One minute after APV injection, LFPO index decreased to 66 ⫾ 18% and its frequency to 87 ⫾ 5%.

Effect of ketamine and pentobarbitone on Purkinje cell firing in the absence of LFPO

Ketamine was injected 3 min after stable Purkinje cell recording in WT (n ⫽ 7 injections) and Cb

Cr

mice (n ⫽ 7 injections) in the absence of LFPO. LFPO episodes never emerged following ketamine. Fig. 2a, b illustrates the spontaneous firing of a Purkinje cell recorded in a Cb

Cr

mouse and the autocorrelogram of the simple spikes before and after the injection of ketamine. Before the injection, the autocorrelogram (Fig. 2a) presents only two small side peaks, indicating a low level of rhythmicity.

After the injection, the number of side peaks greatly in- creases, despite the slight reduction in simple spike firing rate (Fig. 2b). In WT and in Cb

Cr

, rhythm index values significantly increased 3 min after ketamine injec- tion, but reached higher values in Cb

Cr

than in WT mice between the 8th and the 15th minute after the injection (Fig. 2c). Simple spike firing rate significantly decreased 3 min after ketamine injection in WT and Cb

Cr

mice, and remained lower than pre-injection value until the 8th minute, with no significant differences between WT and Cb

Cr

(Fig. 2e). Because of the low rhythmicity before injection, it was possible to reliably determine rhythmic frequency in only 10 cells (five in WT and five in Cb

Cr

mice). Rhythmic frequency de- creased between the 4th and the 15th minute after ket- amine injection (126 ⫾ 60 – 85 ⫾ 40 Hz and 78 ⫾ 35–52 ⫾ 20 in WT and mutants, respectively, values 10 min following the injection), with no significant differences between WT and Cb

Cr

mice.

In the absence of LFPO, pentobarbitone injection in-

duced the same modifications in Purkinje cell firing,

namely a decrease in firing rate 3 min after the injection

(Fig. 2f) and an increase in rhythmicity (Fig. 2d). In mutant

and WT, simple spike firing rate rapidly resumed to pre-

injection values, then decreased again after the 10th and

the 13th minute, respectively. Rhythmicity reached higher

values in Cb

Cr

(n ⫽ 6 injections) than in WT (n ⫽ 8

injections) mice between the 7th and the 15th minute after

pentobarbitone injection (Fig. 2d). Rhythmic frequency be- fore pentobarbitone injection could be reliably determined in 10 cells (six in WT and four in Cb

Cr

) and it significantly decreased after the third minute following the injection (71⫾16 –35⫾14 Hz and 38⫾10 –20⫾15 Hz in WT and mutants, respectively).

In order to determine if these effects were due to a local action on the cerebellar neurons, we performed cer- ebellar microinjections of pentobarbitone (n ⫽6) or ket- amine (n ⫽ 6) during the Purkinje cell recording in absence of LFPO in six Cb

Cr

mice. One minute following pentobarbitone microinjection, rhythm index values sig- nificantly increased from 0.05⫾0.04 – 0.17⫾0.13, and simple spike firing rate decreased from 74 ⫾ 24 Hz to 60⫾23 Hz. Following ketamine microinjection, rhythm index values significantly increased from 0.04 ⫾ 0.05–

0.12⫾0.06 and simple spike firing decreased from

54⫾19 Hz to 40⫾18 Hz. Similar effects were observed following APV microinjections (n ⫽8). Indeed, rhythm in- dex values significantly increased from 0.06⫾0.05– 0.11- 0.07 and simple spike firing rate decreased from 77⫾42–

63 ⫾ 27 Hz.

Pentobarbitone slows down LFPO by decreasing Purkinje cells rhythmic frequency

To gain a better understanding in the interaction between LFPO and Purkinje cell rhythmicity, phase-locked (n ⫽ 3, space between electrodes ⫽ 250 ␮ m (n ⫽ 2) or 500 ␮ m (n ⫽ 1)) and non-phase-locked (n ⫽ 6, space between elec- trode ⫽ 250 ␮ m (n ⫽ 1), 500 ␮ m (n ⫽ 3) or 750 ␮ m (n ⫽ 2)) Purkinje cells were recorded during LFPO episodes before and after pentobarbitone injection. Fig. 3a– h illustrates one of these experiments. Averaged traces demonstrate

Ketamine inhibits and slows down fast cerebellar oscillation. In contrast, pentobarbitone facilitates and slows down fast cerebellar oscillation.

(a, b) Fast cerebellar oscillation (raw recording) and corresponding Fourier spectrum recorded in a Cb

Cr

mouse before (upper trace) and after (lower trace) a ketamine (a) and a pentobarbitone (b) injection. (c, d) Mean values of normalized LFPO index (c) and frequency (d) after pentobarbitone (

), ketamine (

) or saline (

) injections (n

5 for each substance).

L. Servais and G. Cheron / Neuroscience xx (2005) xxx 4

ARTICLE IN PRESS

that both simple and complex spikes were tightly phase- locked to the LFPO recorded simultaneously (Fig. 3a, b) both before and after the pentobarbitone injection. As ex- pected from previous single-unit recordings (Fig. 2), simple spike firing rate decreased, but rhythmicity increased (Fig.

3c, d) five minutes after the injection. During the same

period, LFPO frequency decreased, but LFPO index in-

creased (Fig. 3e, f), as reflected in the sharpening of the

Fourier spectrum. The rhythmic frequency of the cell re-

mained closely correlated with LFPO frequency (r ⫽ 0.99)

(Fig. 3h). This correlation was significantly stronger than

the correlation between LFPO frequency and simple spike

Ketamine and pentobarbitone decrease Purkinje cell simple spike firing rate but increase simple spike rhythmicity. (a, b) Purkinje cell recorded

in a Cb

Cr

mouse before (a) and after (b) ketamine injection. Complex and simple spikes are easily distinguished. The corresponding

autocorrelograms demonstrate the increase in rhythmicity and the decrease in simple spike firing rate. (c, d) Evolution of rhythm index after the

ketamine (c) and pentobarbitone (d) injections in WT (

) and in Cb

Cr

(o) mice. The largest error bars belong to WT mice. The change in error

bar amplitude in WT rhythm index values at 14 min after injection is due to a highly rhythmic cell that entered into a silent mode at this time (e, f)

Evolution of simple spike firing rate after ketamine (e) and pentobarbitone (f) injections, same Purkinje cells.

firing rate (r ⫽ 0.91, one-sided test for r comparison) (Fig.

3g). In all experiments, the rhythmic frequency of phase- locked Purkinje cells remained tightly correlated with the frequency of oscillation (r values ranging from 0.93– 0.99), demonstrating a close relationship between LFPO and Purkinje cell rhythmicity. In all experiments, whether or not the cells were phase-locked to the LFPO, the rhythmic

frequency of the Purkinje cells significantly decreased 3 min after injection (Fig. 3i). This decrease in Purkinje cell rhythmic frequency in Cb

Cr

mice was similar to that observed in the absence of LFPO (Fig. 3i). In all experiments, complex spike activity was phase-locked to the oscillating field concomitantly with the simple spike activity.

Fig. 3.

Pentobarbitone slows down LFPO through slowing down Purkinje cell rhythmic frequency. (a) Simultaneous recording of a Purkinje cell and a LFPO (raw recording), distance between electrodes

250

m. Complex and simple spikes are easily distinguished. Note the LFPO on the Purkinje cell recording (left). Averaged traces of the Purkinje cell and the LFPO recording, the trigger being the Purkinje cell simple (center) and complex (right) spike. (b) Idem, 10 minutes after a pentobarbitone injection. (c, d) Autocorrelograms of the Purkinje cell simple spike illustrated in A before (c) and 10 minutes after (d) the injection. (e, f) Fourier spectrum of the LFPO illustrated in A before (e) and 10 minutes after (f) the injection. (g, h) Plots of LFPO frequency and Purkinje cell simple spike firing rate (g) and rhythmic frequency (h) measured before and every minute after a pentobarbitone injection (experiment illustrated in a– b). (i) Plots of Purkinje cell rhythmic frequency values through time after pentobarbitone injections in the absence of LFPO (o), during LFPO for phase-locked ( ‘ ) and non-phase-locked (

) Purkinje cells.

L. Servais and G. Cheron / Neuroscience xx (2005) xxx 6

ARTICLE IN PRESS

Ketamine inhibition of LFPO is associated with Purkinje cells synchronicity abolition

Eight pairs of Purkinje cells (distance between elec- trodes ⫽ 250 ␮ m) were recorded along the parallel fiber axis in six Cb

Cr

mice during (n ⫽ 4) and in the ab- sence (n ⫽ 4) of LFPO. As expected, the cells were syn- chronous during LFPO episodes (synchronicity in- dex ⫽ 0.078 ⫾ 0.016), whereas little if any synchronicity could be detected in the absence of LFPO (synchronicity index ⫽ 0.005 ⫾ 0.003). Ketamine was injected after 2 min of stable recording. Ten minutes after the injection, synchro- nicity index values decreased to 0.014 ⫾ 0.017 (Fig. 4a) for Purkinje cell pairs recorded during LFPO whereas the synchronicity index of Purkinje cell pairs recorded in the absence of LFPO remained close to zero (Fig. 4b). Fig. 5 illustrates one of these experiments. Before the injection, two Purkinje cells were LFPO phase-locked and synchro- nous, as can be seen in the averaged traces (Fig. 5a, b) and the cross-correlogram (Fig. 5c), respectively. Ten min-

utes after the injection, LFPO was strongly inhibited (Fig.

5d, e). At this time, synchronicity had disappeared (Fig. 5f).

Twenty minutes after the injection, LFPO had also disap- peared on both traces (Fig. 5 g, h) and synchronicity remained equal to zero (Fig. 5i). Phase-locking of both simple (Fig. 5 g, h) and complex spikes (not shown) had also completely disappeared.

Pentobarbitone increases Purkinje cell synchronicity during LFPO

Eight pairs of Purkinje cells were recorded during LFPO episodes (synchronicity index ⫽ 0.013 ⫾ 0.014). The distance between Purkinje cells was 250 ␮ m for the three most syn- chronous pairs, 500 (n ⫽ 1), 750 (n ⫽ 3) and 1000 (n ⫽ 1) ␮ m for the others. Five minutes after pentobarbitone injection, synchrony index values significantly increased for all pairs (synchronicity index ⫽ 0.030 ⫾ 0.017) (Fig. 4c). Fig. 6 illus- trates one of these experiences. Before pentobarbitone injection, Purkinje cells #2 and #3 were phase-locked to

Fig. 4.

Differential effect of ketamine and pentobarbitone on Purkinje cell synchronicity. (a) Plots of Purkinje cell pair synchronicity before (left) and

after (right) ketamine injection during LFPO (n

4). (b) Idem, in the absence of LFPO. (c) Plots of Purkinje cell pair synchronicity before (left) and after

(right) pentobarbitone injection during LFPO (n

8). (d) Idem, in the absence of LFPO (n

10).

the LFPO, as demonstrated by the averaged traces (Fig.

6b, c). Peaks in the auto (Fig. 6f, i) and cross-correlogram (Fig. 6h) indicate rhythmicity and synchronicity, respec- tively. Purkinje cell #1 was also slightly rhythmic (Fig. 6d), but it was not synchronous with the other cells (Fig. 6e, g) and was not phase-locked to the LFPO (Fig. 6a). Ten minutes after the injection, the rhythmicity of all cells had increased (Fig. 6m, o, r), and synchronicity had appeared between cell #1 and the other two cells (Fig. 6n, p, see arrow). Moreover, cell #1 had phase-locked to the LFPO, as demonstrated by the averaged trace (Fig. 6j, see ar- row). Cells #2 and #3 remained phase-locked to LFPO (Fig. 6k, l) and synchronous (Fig. 6q).

As previously demonstrated, pentobarbitone increases Purkinje cell rhythmicity (Fig. 2), LFPO index (Fig. 1) and Purkinje cell synchronicity during LFPO (Figs. 4 and 6). How- ever, pentobarbitone injection never elicited the emergence of LFPO in previously non-oscillating WT or Cb

Cr

mice. Therefore, we studied the influence of pentobarbitone injection on non-synchronous Purkinje cells located along the same parallel fiber axis in seven Cb

Cr

mice in the absence of LFPO (distance between electrodes ⫽ 250 ␮ m). A total of 10 pairs of non-synchronous Purkinje cells (syn- chronicity index ⫽ 0.0028 ⫾ 0.0024) were studied before and after pentobarbitone injection. No synchronicity ap- peared after the injection (synchronicity index ⫽ 0.0027

⫾ 0.0021) (Fig. 4d). This demonstrates that the enhance- ment of Purkinje cell synchronicity by pentobarbitone re- quires the presence of LFPO. Fig. 7 (same scale and display as Fig. 6) illustrates one of these experiments.

Before pentobarbitone injection, three Purkinje cells lo- cated along the parallel fiber axis were recorded in a Cb

Cr

mouse in the absence of LFPO, as illustrated by averaged traces triggered by the simple spikes of each cell (Fig. 7a– c). The cells presented little if any rhythmicity,

all three autocorrelograms being nearly flat (Fig. 7d, f, i). In addition, they were not synchronous, as demonstrated by flat cross-correlograms (Fig. 7e, g, h). Ten minutes after the injection, the rhythmicity of each cell increased despite a decrease in the simple spike firing rate (Fig. 7m, o, r), but no synchronicity between cells (Fig. 7n, p, q) no LFPO (Fig. 7j–l) appeared.

DISCUSSION

These results demonstrate the close and reciprocal inter- action between fast cerebellar oscillation and Purkinje cell firing. LFPO not only is supported by the synchronous firing of rhythmic Purkinje cells, but also participate in their synchronization.

Pharmacological action of pentobarbitone and ketamine

Pentobarbitone facilitates GABA

receptors and increases membrane permeability to Cl

, which hyperpolarizes cell membranes (Olsen and Snowman, 1982). In the cerebellum, GABA

receptors are widely distributed. In Purkinje cells, they mediate the inhibitory input of molecular interneurons.

Ketamine blocks NMDA receptors, but has also a number of other pharmacological properties, including effects on acetyl- cholinesterase (Cohen et al., 1974), monoamine transporters (Smith et al., 1981) and opiate receptors (Smith et al., 1980).

APV is a much more specific NMDA antagonist. In the cere- bellar cortex, NMDA receptors are located on interneurons, granule and brush cells, and they mediate the mossy fibers excitation. They are also present at the presynaptic level, at the parallel fiber-Purkinje cell and at the interneuron-Purkinje cell synapse (Casado et al., 2002; Duguid and Smart, 2004).

Despite the immunolabeling of some NMDA receptor sub- units at the Purkinje cell level in adult mice (Thompson et al.,

Ketamine inhibits fast cerebellar oscillation through Purkinje cells synchronicity. (a, b) Averaged traces of two LFPO-phase-locked Purkinje cells (distance between electrodes

250

m) located at the same depth along the parallel fiber axis. (c) Cross-correlogram of the two cells indicating a high degree of synchronicity. (d, e, g, h) Averaged traces of the same cells, 10 (d, e) and 20 (g, h) minutes after a ketamine injection. (f, i) Cross-correlograms of the two cells 10 (f) and 20 (i) minutes after the ketamine injection.

L. Servais and G. Cheron / Neuroscience xx (2005) xxx 8

ARTICLE IN PRESS

2000), it is generally assumed that Purkinje cell excitation by parallel or climbing fiber action is not mediated by NMDA receptors. Therefore, the main inhibiting action of ketamine at the Purkinje cell level is probably related to reduction in parallel fiber activity due to reduced granule cell excitation.

Simple spike firing rate and rhythmicity following pentobarbitone and ketamine injection

Our findings are consistent with previous reports describing a decreased simple spike firing rate following pentobarbitone and ketamine injection (Sinclair and Lo, 1981; Sato et al., 1993).

Fig. 6.

Pentobarbitone increases Purkinje cell synchronicity during LFPO episodes. (a– c) Averaged traces (trigger

simple spike of Purkinje cell #1

(a), #2 (b) and #3 (c)) of three Purkinje cells located at the same depth along the parallel fiber axis (distance between cells

250

m). The recording

is performed in a Cb

Cr

mouse during a LFPO episode. Note that cell #1 is not phase-locked to the LFPO. (d, f, i) Autocorrelograms of Purkinje

cells #1 (d), #2 (f) and #3 (i). (e, g, h) Cross-correlograms of cells #2 (e) and #3 (g) triggered by cell #1, and of cell #3 triggered by cell #2 (h). (j–l)

Averaged traces of the Purkinje cells recordings, 10 min after a pentobarbitone injection. Compare A and J (arrow). (m, o, r) Autocorrelograms of

Purkinje cells #1 (m), #2 (o) and #3 (r) 10 minutes after the injection. (n, p, q) Cross-correlograms of cells #2 (n) and #3 (p) triggered by cell #1, and

of cell #3 triggered by cell #2 (q). Compare n, p and e, g (arrow).

These authors attributed the inhibiting action of pentobarbitone and ketamine on simple spike firing to Purkinje cell hyperpolar- ization by pentobarbitone and to a decrease in parallel fiber activity by ketamine. In the present study, cerebellar microinjec- tions confirm that the action of ketamine and pentobarbitone are

not related to the non-specific effects of general anesthesia and that ketamine action is related to specific NMDA antagonism.

Purkinje cells alternate between tonic, silent and burst- ing modes of firing both in vitro (Womack and Khodakhah, 2002) and in vivo (Servais et al., 2004). During tonic mode

Pentobarbitone increases Purkinje cell rhythmicity, but does not lead to the emergence of fast LFPO or Purkinje cell synchronization. (a– c) Averaged traces triggered by the simple spike of Purkinje cells #1 (a), #2 (b) and #3 (c) of three Purkinje cells located at the same depth along the parallel fiber axis (distance between cells

250

m). The recording is performed in a Cb

Cr

mouse in the absence of LFPO. (d, f, i) Autocorrelograms of Purkinje cells #1 (d), #2 (f) and #3 (i). (e, g, h) Cross-correlogram of cells #2 (e) and #3 (g) triggered by cell #1, and of cell #3 triggered by cell #2 (h). (j–l) Averaged traces of the Purkinje cells recordings, 10 min after a pentobarbitone injection. (m, o, r) Autocorrelograms of Purkinje cells #1 (m), #2 (o) and #3 (r) 10 minutes after the injection. (n, p, q) Cross-correlograms of cells #2 (n) and #3 (p) triggered by cell #1, and of cell #3 triggered by cell #2 (q). Note that no central peak has appeared.

L. Servais and G. Cheron / Neuroscience xx (2005) xxx 10

ARTICLE IN PRESS

periods, Purkinje cells are poorly rhythmic in alert animals (Goossens et al., 2001; Cheron et al., 2004, 2005), more rhythmic in anesthetized animals (Schwarz and Welsh, 2001), and much more so in vitro (Womack and Khoda- khah, 2002). These differences may be explained by the fact that spontaneous inhibitory and excitatory synaptic inputs are strongly decreased or blocked in slice prepara- tion. The effect of ketamine on Purkinje cell rhythmicity might be explained by a similar effect in vivo related to inhibition of parallel fiber activity. In contrast, the increased rhythmicity following pentobarbitone, previously reported in rats and rabbits (Gogolak et al., 1984; Sinclair and Lo, 1981), might be related to the pentobarbitone-induced hy- perpolarization of the cell. Indeed, Chang and colleagues (1993) have proposed that the oscillatory activity of Pur- kinje cells in slice preparation is driven by a hyperpolariza- tion-activated cation current.

Pentobarbitone and ketamine effects on LFPO Cerebellar microinjections of GABA

or NMDA blockers (gabazine and APV, respectively) transiently inhibit LFPO close to the site of injection (Cheron et al., 2004, 2005;

Servais et al., 2005). Therefore, LFPO inhibition by ket- amine and facilitation by pentobarbitone was expected, despite the common effect of these drugs on Purkinje cell rhythmicity. The cerebellar microinjections of pentobarbi- tone and ketamine during LFPO and Purkinje cell record- ings confirmed that these effects were related to a direct action on the cerebellum. The opposite effects of both drugs on fast LFPO further demonstrate that their specific mode of action rather than non-specific effect of general anesthesia was involved. Facilitation of cerebellar fast LFPO and enhancement of Purkinje cell rhythmicity and synchronicity after a pentobarbitone injection despite the decrement in the simple spike firing rate demonstrates that fast cerebellar oscillation is not simply the consequence of an increased simple spike firing rate. Moreover, the inhi- bition of fast LFPO by ketamine despite its significant enhancement of Purkinje cell rhythmicity demonstrates that Purkinje cell rhythmicity is not a sufficient condition to maintain a fast oscillating field.

General anesthetics such as ketamine and pentobar- bitone have been widely used to better understand the mechanisms supporting the emergence of brain oscilla- tions in a wide range of frequencies and in different re- gions, including the red nucleus (Gogolak et al., 1984), piriform sinus (Fontanini et al., 2003), hippocampus (Fisahn et al., 1998; Faulkner et al., 1999; Dickinson et al., 2003), auditory cortex (Cotillon-Williams and Edeline, 2003), and in other parts of the cerebral cortex (Contreras et al., 1997; Mahon et al., 2001). These studies have addressed oscillations in a wide range of frequencies, from 1 to 70 Hz. Therefore, reported effects of the studied drugs have shown marked differences according to the oscilla- tion and the recorded brain region. For example, the ef- fects of barbiturates include waxing and waning (Contreras et al., 1997), facilitation (Gogolak et al., 1984; Cotillon- Williams and Edeline, 2003) and the slowing down of oscillations (Fisahn et al., 1998; Khazipov and Holmes,

2003). Ketamine also presents different effects on oscilla- tions according to the frequency range and the brain region (Contreras et al., 1997; Faulkner et al., 1999; Fontanini et al., 2003). The effects of general anesthetics on oscilla- tions above 100 Hz have been much less documented.

Ketamine does not inhibit neocortical ripples (Grenier et al., 2001). Halothane, a gap junction blocker, strongly in- hibits fast oscillations in the hippocampus (Ylinen et al., 1995), the somatosensory cortex (Jones et al., 2000) and the neocortex (Grenier et al., 2001). As fast cerebellar LFPO was blocked by carbenoxolone (Cheron et al., 2004), we expect a similar effect during halothane anes- thesia. These experiments are beyond the scope of the present study and will be reported in a distinct paper.

Simple spike rhythmicity and LFPO

We found that LFPO frequency is closely correlated to phase-locked Purkinje cell rhythmic frequency. These two parameters concomitantly decrease following pentobarbi- tone injection and remain nearly equal. This could reflect that the slowing down of LFPO by pentobarbitone second- arily slows down phase-locked Purkinje cells. Similar rela- tionship between LFPO and neuronal activity has been described in other regions of the CNS, such as the lateral amygdala and perirhinal cortex, where neurons display modulation in firing probability in relation to LFPO (Collins et al., 2001). Alternatively, pentobarbitone-induced de- crease in rhythmic frequency of Purkinje cells may sec- ondarily slow down the LFPO supported by their synchro- nous firing. A similar mechanism has been demonstrated between hippocampal LFPO and fast synchronous activity of pyramidal cells (Jones et al., 2000). During LFPO epi- sodes, pentobarbitone consistently slows down phase- locked and non-phase-locked Purkinje cells. A similar ef- fect is observed in absence of LFPO and in WT mice where LFPO never occurs. This demonstrates that pento- barbitone has a direct LFPO independent effect on Pur- kinje cell rhythmic frequency. Thus, the slowing down of LFPO following pentobarbitone injections results from the decrease in Purkinje cells rhythmic frequency, which dem- onstrates that LFPO is supported by the synchronous rhythmic firing of Purkinje cells. How the other cerebellar neurons participate in this synchronization remains to be studied. If Golgi cells are definitely not involved in LFPO (Cheron et al., 2004), the role of inhibitory interneurons is suggested by the action of carbenoxolone on the densely gap-junction-connected neurons. Simultaneous recordings of interneurons along with Purkinje cells and fast LFPO are required to further assess this question.

The phase-locking of Purkinje cells with the fast LFPO not only regards their simple but also their complex spike activity. This probably reflects the projections of the oscil- lating activity in the different brain regions targeted by cerebellar projections, one of these being the inferior olive.

The phase-locking of complex spike activity was not af-

fected by pentobarbitone injection, but disappeared con-

comitantly with the simple spike phase-locking following

ketamine injection.

Simple spike synchronicity and LFPO

The main finding of this study is that fast LFPO requires Purkinje cell synchronicity, but also that Purkinje cell syn- chronicity depends on the existence of a fast LFPO. As fast LFPO is not an all-or-none phenomenon, one may sup- pose that its amplitude is governed by the number of Purkinje cells that beat synchronously with the same rhyth- mic frequency. During the facilitation of LFPO by pento- barbitone, synchronicity between pairs distanced by up to 1 mm could be detected. Since the signal/noise ratio is better in Purkinje cells that are not phase-locked with LFPO, many recordings of synchronous Purkinje cell pairs or phase-locked Purkinje cells could not be used because of animal movements during the recording. Therefore, the ratio between phase-locked and not phase-locked Purkinje cells is much higher in the dataset.

Ketamine blocks Purkinje cell synchronicity and inhib- its LFPO. In Cr

, and by extension in Cb

Cr

mice, the enhancement of simple spike firing rate is caused by the increased parallel fiber activity due to increased gran- ule cell excitability (Schiffmann et al., 1999; Gall et al., 2003; Bearzatto et al., 2004). This increased parallel fiber activity may also contribute to Purkinje cell synchronization along their axis. In contrast, the blocking action of ket- amine on NMDA receptors decreases the excitation of granule cells, leading to a decrease in parallel fiber activity.

This could explain the inhibiting action of ketamine on synchronization and LFPO. Non-NMDA-specific action of ketamine is probably not involved here, as local injection of APV had the same effects as ketamine on LFPO and Purkinje cell firing. In contrast with the inhibiting action of ketamine, pentobarbitone facilitates LFPO. As pentobarbi- tone increases Purkinje cell synchronicity during LFPO, it could be hypothesized that pentobarbitone directly in- creases Purkinje cell synchronicity. However, in the ab- sence of LFPO, pentobarbitone did not elicit Purkinje cell synchronicity or LFPO, despite the emergence of the highly rhythmic firing pattern of Purkinje cells. Thus, the following mechanism may be proposed: Pentobarbitone would directly increase Purkinje cell rhythmicity, facilitating LFPO if already present. The increased oscillating field would secondarily recruit rhythmic Purkinje cells. This would constitute a positive feedback loop where the fast oscillation is the cause and the consequence of Purkinje cell synchronicity. The emergence of fast oscillation would thus depend on the synchronization of a critical number of rhythmic Purkinje cells leading to a local oscillating field which would itself synchronize other neighboring Purkinje cells. A similar feedback mechanism between neuronal synchronization and fast oscillation has been recently pro- posed in neocortical ripples, which are probably implicated in seizure initiation (Grenier et al., 2003). As fast cerebellar oscillation in mice has been reported only in ataxic animals to date, our findings suggest that such a mechanism of reciprocal interaction between fast oscillation and synchro- nous neurons may be implicated in the pathophysiological pathways that lead to neuronal network instability or inef- ficiency.

Acknowledgment—L. Servais is research assistant for the Belgian National Fund for Scientific Research (FNRS). This work was funded by the FNRS (Belgium), research funds of ULB and UMH (Belgium). The authors thank P. Demaret and M. P. Dufief for technical assistance; C. Warocquier and P. Kellidis for animal care; Dr. R. Leach, Prof B. Dan and D. Kingwill for their helpful comments on the manuscript.

REFERENCES

Airaksinen MS, Eilers J, Garaschuk O, Thoenen H, Konnerth A, Meyer M (1997) Ataxia and altered dendritic calcium signalling in mice carrying a targeted null mutation of the calbindin D28k gene. Proc Natl Acad Sci U S A 94:1488 –1493.

Baker SN, Kilner JM, Pinches EM, Lemon RN (1999) The role of synchronicity and oscillations in the motor output. Exp Brain Res 128:109 –117.

Bearzatto B, Servais L, Gall G, Roussel C, Baba-Aissa F, de Kerchove d’Exaerde A, Schurmans S, Cheron G, Schiffmann SN (2004) Selective expression of calretinin into cerebellar granule cells is essential for granule cell excitability, Purkinje cells firing, and motor coordination. Society for Neuroscience, Abstract No. 827.9.

Blednov YA, Simpson VJ (1999) Gas chromatography-mass spec- trometry assay for determination of ketamine in brain. J Pharmacol Toxicol Methods 41:91–95.

Casado M, Isope P, Ascher P (2002) Involvement of presynaptic N-methyl-D-aspartate receptors in cerebellar long-term depres- sion. Neuron 33:123–130.

Chang W, Strahlendorf JC, Strahlendorf HK (1993) Ionic contributions to the oscillatory firing activity of rat Purkinje cells in vitro. Brain Res 614:335–341.

Cheron G, Gall D, Servais L, Dan B, Maex R, Schiffmann SN (2004) Inactivation of calcium-binding protein genes induces 160 Hz os- cillations in the cerebellar cortex of alert mice. J Neurosci 24:

434 – 441.

Cheron G, Servais L, Wagstaff J, Dan B (2005) Fast cerebellar oscil- lation associated with ataxia in a mouse model of Angelman syn- drome. Neuroscience 130:631– 637.

Cohen ML, Chan SL, Bhargave HN, Trevor AJ (1974) Inhibition of mammalian brain acetylcholinesterase by ketamine. Biochem Pharmacol 23:1647–1652.

Collins DR, Pelletier JG, Pare D (2001) Slow and fast (gamma) neu- ronal oscillations in the perirhinal cortex and lateral amygdala.

J Neurophysiol 85:1661–1672.

Contreras D, Destexhe A, Steriade M (1997) Spindle oscillations dur- ing cortical spreading depression in naturally sleeping cats. Neu- roscience 77:933–936.

Cotillon-Williams N, Edeline JM (2003) Evoked oscillations in the thalamo-cortical auditory system are present in anesthetized but not in unanesthetized rats. J Neurophysiol 89:1968 –1984.

Dickinson R, Awaiz S, Whittington MA, Lieb WR, Franks NP (2003) The effects of general anaesthetics on carbachol-evoked gamma oscillation in the rat hippocampus in vitro. Neuropharmacology 44:864 – 872.

Duguid IC, Smart TG (2004) Retrograde activation of presynaptic NMDA receptors enhances GABA release at cerebellar interneu- ron-Purkinje cell synapses. Nat Neurosci 7:525–533.

Eckhorn R, Thomas U (1993) A new method for the insertion of multiple microprobes into neural and muscular tissue, including fiber electrodes, fine wires, needles and microsensors. J Neurosci Methods 49:175–179.

Faulkner HJ, Traub RD, Whittington MA (1999) Anaesthetic/amnesic agents disrupt beta frequency oscillations associated with potenti- ation of excitatory synaptic potentials in the rat hippocampal slice.

Br J Pharmacol 128:1813–1825.

Fisahn A, Pike FG, Buhl EH, Paulsen O (1998) Cholinergic induction of network oscillations at 40 Hz in the hippocampus in vitro. Nature 394:186 –189.

L. Servais and G. Cheron / Neuroscience xx (2005) xxx 12

ARTICLE IN PRESS

Fontanini A, Spano P, Bower JM (2003) Ketamine-xylazine-induced slow (

1.5 Hz) oscillations in the rat piriform (olfactory) cortex are functionally correlated with respiration. J Neurosci 23:7993– 8001.

Gall D, Roussel C, Susa I, D’Angelo E, Rossi P, Bearzatto B, Galas MC, Blum D, Schurmans S, Schiffmann SN (2003) Altered neuro- nal excitability in cerebellar granule cells of mice lacking calretinin.

J Neurosci 23:9320 –9327.

Gogolak G, Huck S, Stumpf C (1984) Drug-induced rhythmic burst activity of cerebellar neurons. Naunyn Schmiedebergs Arch Phar- macol 326:227–232.

Goossens J, Daniel H, Rancillac A, van der Steen J, Oberdick J, Crepel F, De Zeeuw CI, Frens MA (2001) Expression of protein kinase C inhibitor blocks cerebellar long-term depression without affecting Purkinje cell excitability in alert mice. J Neurosci 21:

5813–5823.

Grenier F, Timofeev I, Steriade M (2001) Focal synchronization of ripples (80 –200 Hz) in neocortex and their neuronal correlates.

J Neurophysiol 86:1884 –1898.

Grenier F, Timofeev I, Steriade M (2003) Neocortical very fast oscil- lations (ripples 80 –200 Hz) during seizures; intracellular corre- lates. J Neurophysiol 89:841– 852.

Hartmann MJ, Bower JM (1998) Oscillatory activity in the cerebellar hemispheres of unrestrained rats. J Neurophysiol 80:1598 –1604.

Jefferys JG (1995) Nonsynaptic modulation of neuronal activity in the brain: electric currents and extracellular ions. Physiol Rev 75:

689 –723.

Jones MS, MacDonald KD, Choi B, Dudek FE, Barth DS (2000) Intracellular correlates of fast (

200 Hz) electrical oscillations in rat somatosensory cortex. J Neurophysiol 84:1505–1518.

Khazipov R, Holmes GL (2003) Synchronization of kainate-induced epileptic activity via GABAergic inhibition in the superfuse rat hip- pocampus in vivo. J Neurosci 23:5337–5341.

Lang EJ, Sugihara I, Welsh JP, Llinas R (1999) Patterns of spontane- ous Purkinje cell complex spike activity in the awake rat. J Neurosci 19:2728 –2739.

Mahon S, Deniau JM, Charpier S (2001) Relationship between EEG potentials and intracellular activity of striatal and cortico-striatal neurons: an in vivo study under different anesthetics. Cereb Cortex 11:360 –373.

Miura K, Kishino T, Li E, Webber H, Dikkes P, Holmes GL, Wagstaff J (2002) Neurobehavioral and electroencephalographic abnormali- ties in Ube3a maternal-deficient mice. Neurobiol Dis 9:149 –159.

Olsen RW, Snowman AM (1982) Chloride-dependent enhancement by barbiturates of

-aminobutyric acid receptor binding. J Neurosci 2:1812–1823.

Pellerin JP, Lamarre Y (1997) Local field potential oscillations in primate cerebellar cortex during voluntary movement. J Neuro- physiol 78:3502–3507.

Sato Y, Miura A, Fushiki H, Kawasaki T (1993) Barbiturate depresses simple spike activity of cerebellar Purkinje cells after climbing fiber input. J Neurophysiol 69:1082–1190.

Saubermann AJ, Gallagher ML, Hedley-Whyte J (1974) Uptake, dis- tribution, and anesthetic effect of pentobarbital-2-14C after intra- venous injection into mice. Anesthesiology 40:41–51.

Schiffmann S, Cheron G, Lohof A, d’Alcantara P, Meyer M, Parmentier M, Schurmans S (1999) Impaired motor coordination and Purkinje cell excitability in mice lacking calretinin. Proc Natl Acad Sci U S A 96:5257–5262.

Schwarz C, Welsh JP (2001) Dynamic modulation of mossy fiber system throughput by inferior olive synchrony: a multielectrode study of cerebellar cortex activated by motor cortex. J Neuro- physiol 86:2489 –2504.

Servais L, Bearzatto B, Hourez R, Dan B, Schiffmann SN, Cheron G (2004) Effect of simple spike firing mode on complex spike firing rate and waveform in cerebellar Purkinje cells in non-anesthetized mice. Neurosci Lett 367:171–176.

Servais L, Bearzatto B, Schwaller B, Dumont M, De Saedeleer C, Dan B, Barski JJ, Schiffmann SN, Cheron G (2005) Mono- and dual- frequency fast cerebellar oscillation in mice lacking either parval- bumin and/or calbindin D-28k. Eur J Neurosci, in press.

Sinclair JG, Lo GF (1981) The effect of pentobarbital on rat cerebellar Purkinje cells. Gen Pharmacol 12:327–330.

Smith DJ, Azzaro AJ, Zaldivar SB, Palmer S, Lee HS (1981) Proper- ties of the optical isomers and metabolites of ketamine on the high affinity transport and catabolism of monoamines. Neuropharmacol- ogy 120:391–396.

Smith DJ, Pekoe GM, Martin LL, Coalgate B (1980) The interaction of ketamine with the opiate receptor. Life Sci 26:789 –795.

Steriade M, Amzica F, Neckelmann D, Timofeev I (1998) Spike-wave complexes and fast components of cortically generated seizures.

II. Extra- and intracellular patterns J Neurophysiol 80:1456 –1479.

Steriade M, Contreras D, Amzica F, Timofeev I (1996) Synchronization of fast (30 – 40 Hz) spontaneous oscillations in intrathalamic and thalamocortical networks. J Neurosci 16:2788 –2808.

Steriade M, Nunez A, Amzica F (1993) A novel slow (

1Hz) oscillation of neocortical neurons in vivo: depolarising and hyperpolarizing components. J Neurosci 13:3284 –3299.

Stratton SE, Lorden JF, Mays LE, Oltmans GA (1988) Spontaneous and harmaline-stimulated Purkinje cell activity in rats with a genetic movement disorder. J Neurosci 8:3327–3336.

Sugihara I, Lang EJ, Llinas R (1995) Serotonin modulation of inferior olivary oscillations and synchronicity: a multiple-electrode study in the rat cerebellum. Eur J Neurosci 7:521–534.

Thompson CL, Drewery DL, Atkins HD, Stephenson FA, Chazot PL (2000) Immunohistochemical localization of N-methyl-D-aspartate receptor NR1, NR2A, NR2B and NR2C/D subunits in the adult mammalian cerebellum. Neurosci Let 283:85– 88.

Whittington MA, Traub RD, Jefferys JG (1995) Synchronized oscilla- tion in interneuron networks driven by metabotropic glutamate receptors activation. Nature 373:612– 615.

Womack M, Khodakhah K (2002) Active contribution of dendrites to the tonic and trimodal patterns of activity in cerebellar Purkinje neurons. J Neurosci 22:10603–10612.

Ylinen A, Bragin A, Nadasdy Z, Jando G, Szabo I, Sik A, Buzsaki G (1995) Sharp wave-associated high-frequency oscillation (200 Hz) in the intact hippocampus: network and intracellular mechanisms.

J Neurosci 15:30 – 46.

(Accepted 1 June 2005)