| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TRANSLATIONAL PHYSIOLOGY
1Laboratoire de Neurophysiologie, Centre National de la Recherche Scientifique Unté Mixte de Recherche 5543, Université de Bordeaux 2, Bordeaux Cedex; and 2Institut de Neurobiologie de la Méditerranée, Institut National de la Recherche Médicale U 29, Marseille Cedex 9, France; and 3School of Mathematics, University of Southampton, Southampton, United Kingdom
Submitted 12 May 2005; accepted in final form 6 August 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
100 Hz (130–185 Hz), with pulses of short duration (60–100 µs) and 1- to 3-mA amplitude. Other parameter values produce no effect or aggravate the symptoms. To gain a better understanding of the mechanisms that underlie the therapeutic action of HFS, we have compared the effects of different combinations of parameter values delivered by clinical stimulators on the activity of STN neurons recorded in whole cell patch-clamp configuration in slices. We showed that none of tested combinations of parameters silenced the neurons. Non-therapeutic combinations i.e., low-frequency pulses (10–50 Hz), even at large amplitude or width, further excited the STN neurons with respect to their spontaneous activity. In contrast, combinations in the therapeutic range (80–185 Hz, 90–200 µs, 500–800 µA) replaced the preexisting activity by spikes, time-locked to the stimuli and thus presenting a striking regularity. When increasing pulse width or amplitude in this high-frequency range, the dual effect was still present but the activity generated became more irregular. We propose that during HFS at clinically relevant parameters, STN neurons behave as stable oscillators entirely driven by the stimulation, giving an average stable STN output that overrides spontaneous activity and introduces high-frequency regular spiking in the basal ganglia network. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Limousin et al. 1995). HFS operates with a certain combination of parameter values: the best effects on motor behavior are obtained with regular pulses of 100–185 Hz with a narrow width (60–100 µs) and an amplitude of 1–3 mA. In contrast low-frequency stimulation (LFS) at 5–10 Hz has no effect or worsens parkinsonism (Moro et al. 2002; Rizzone et al. 2001). The mechanisms by which electrical stimulation of a structure deep in the brain reinstates motor function are still not elucidated. According to studies in humans, suffering from Parkinson's disease and in monkeys treated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), loss of dopaminergic projections to the striatum results in hyperactivity of subthalamic neurons, aberrant temporal couplings, and abnormal oscillations in cortico-basal ganglia circuits with increased coherence in the 10- to 30-frequency band (Bergman et al. 1998; Hassani et al. 1996; Levy et al. 2000, 2002; Miller and De Long 1987; Williams et al. 2002). The hyperactivity of STN in the parkinsonian state and the amelioration of motor symptoms by lesion or chemical inactivation of STN neurons led to the hypothesis that STN-HFS silences STN neurons (Aziz et al. 1991; Bergman 1990; Gill and Heywood 1997; Levy et al. 2001). In agreement with this hypothesis, extracellular recordings during HFS in patients and animal models reported a reduction of STN activity (Filali et al. 2004; Tai et al. 2003; Welter et al. 2004). The silencing hypothesis was first challenged by recordings in MPTP-treated monkeys where HFS, at parameters that decreased akinesia, changed the spontaneous irregular firing of target cells of the STN (GPe and GPi) to a high-frequency regular pattern of discharge, time-locked to the stimulation (Hashimoto et al. 2003). Using rat STN slices, we have recently shown that HFS at therapeutic frequencies generates a stable pattern of bursts of spikes time-locked to the stimulation that overrides ongoing activity (Garcia et al. 2003). Based on these observations, we suggested that the beneficial actions of HFS are related to both the suppression of the spontaneous activity and its replacement by a novel pattern of activity. If this hypothesis is correct, only STN-HFS at "therapeutic" parameters should have this dual effect. In the present study, we have varied the values of each of the parameters one at a time and determined how it modified the pattern of STN activity in slices from dopamine-depleted rats. We report that therapeutic and non-therapeutic combinations of parameters have totally different impacts on STN activity. |
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
We performed experiments in coronal STN slices from reserpine-treated (Hernandez-Lopez et al. 1996; Moody and Spear 1992) Nineteen- to 22-day-old Wistar rats using a preparation technique that has been described in detail previously (Garcia et al. 2003). We preferred this model to the 6-hydroxydopamine-treated rats because we primarily aimed at obtaining dopamine depletion inside the STN. Also in the latter procedure, because of the time required for the degeneration of dopaminergic neurons, the rats were older and thus less convenient for in vitro recordings. Five experiments were also performed in 6-wk-old rats (140–160 g) to verify that the results obtained were not due to the immaturity of the preparation. We do not present these results separately since they were similar to those obtained in younger (juvenile) rats. After a 2-h recovery period, we placed the slices in a submersion-type recording chamber and perfused them (1.5–2 ml/min) at room temperature with a bicarbonate-buffered solution saturated with 95% O2-5% CO2 containing (in mM) 124 NaCl, 3.6 KCl, 1.3 MgCl2, 1.25 N-[2-hydroxyethyl]piperazin-N'-2-ethanesulfonic acid] (HEPES), 2.4 CaCl2, 26 HCO3, and 10 glucose.
Stimulation and artifact suppression
The stimulating electrode used in the present study (NEX-100; Phymep, Paris) had a contact length of 500 µm and a contact area of 0.3 mm2. We positioned the electrode in the middle of the STN and connected it in the monopolar configuration. We applied the stimulation between the tip of the electrode (negative pole) and a platinum wire surrounding the slice (ground). The electrical stimulation consisted of negative square current pulses the amplitude of which was displayed continuously. First we fixed the width of the pulses and then we adjusted manually their frequency. Once these two parameters were adjusted, we increased the current amplitude gradually to avoid sudden variations of the baseline. We tested different durations (60, 90, 200, 400 µs), intensities (100–1,000 µA) and frequencies (10, 50, 80, 130, 185 Hz). For a current of 500 µA, the current density delivered at the tip of the electrode was 1.6 mA/mm2. However, stimulation applied in a submerged recording chamber places the exposed portion of the stimulating electrode in contact with the bath as well as with the tissue. This introduces a bath resistance (50 
-cm) in parallel with the tissue resistance (300–500 -cm) (Nowak and Bullier 1998) and thus reduces the current seen by the tissue. We have estimated that with the NEX-100 electrode, 2/3 of the applied current leaked in the bath. In the present study, we always report the current intensity applied to the slices, though in the tissue the current was 3 times smaller. To verify this estimation, we conducted some experiments with a smaller tipped electrode having a contact length of 250 µm and a contact area of 0.2 mm2 (RNEX-300, Phymep, France). When this electrode was positioned in the STN, its contact surface was entirely inside the 400-µm-thick slice. The application of a current two to three times weaker than in the previous experiment produced similar electrophysiological responses of the STN neurons. Although pulse frequency and pulse width stay the same whatever the electrode and recording configuration, the differences in surface area of the electrode contact and the configuration of the extracellular medium renders the comparison of pulse amplitude between in vivo and in vitro studies difficult. In patients, the electrode delivers a current of around 500 µA/mm2 (3 mA across 6 mm2) (Lozano et al. 2002), i.e., approximately three times less than in the present study. However, the current density generated by a monopolar point source falls off rapidly in space. For example at a distance of 200–400 µm from the stimulating electrode, the current density generated in vivo by a current intensity of 1 mA is equal to 500-2,000 µA/cm2 (Grill 2003).
Artifacts in current-clamp recordings had a longer duration than that applied (60–400 µs) due to membrane capacitance, but all the same much shorter than that of spikes (1–2 ms). A sample and hold electronic device (built in the laboratory) synchronized with the stimulator allowed the removal of artifacts (Minzly et al. 1993). This artifact suppressor held the recorded signal at the level before the current pulse and restored it a few microseconds later. After the suppression procedure, stimulation artifacts appeared in recordings as positive capacitative rebounds of 3- to 15-mV amplitude, surmounted or not by evoked spikes (see Fig. 1D). We did not suppress this rebound to visualize the stimuli. Comparison of recording traces before and after artifact suppression showed that the sample-and-hold procedure did not consistently affect spike recordings.
|
Recorded STN cells were located at a distance of 100–400 µm from the central stimulating electrode. We used an Axoclamp 2A amplifier (Axon Instruments, Foster City, CA) to perform whole cell recordings in current-clamp mode. The electrodes (8–12 M) contained (in mM) 120 Kgluconate, 10 KCl, 10 NaCl, 1 CaCl2, 1.1 or 10 EGTA, 10 HEPES, 0 or 2 MgCl2, 2 MgATP, and 0.5 NaGTP, pH 7.2–7.3. We recorded only one cell per stimulated STN to avoid posteffects of long-term stimulation (1–2 h). Current output was displayed on a four-channel chart recorder (Gould Instruments, Longjumeau, France), digitized (Digidata, Axon Instruments), stored on a computer using axoscope software (Axon Instruments) and videotaped for subsequent off-line analysis. We differentiated spikes evoked by the stimulation from the spontaneous ones on the following basis: they were always triggered by artifacts and were not preceded by the spontaneous slow depolarization known to result from the persistent Na+ current (Beurrier et al. 2000; Bevan et al. 1999) (compare Fig. 1, C and D, left).
Mathematical analysis
We analyzed the recordings in a semi-automatic manner using purpose-written software in Yorick. The software detected automatically spikes and bursts. Spikes are a region of the trace with high variation and were identified by the time of occurrence of the maximum. The software identified bursts as plateaus of the lower envelope of the trace: it considered all regions of the envelope above a certain threshold as bursts and the corresponding spikes as part of the same burst. We inspected all the traces visually to ensure the correct identification of all their features. In those cases where the default threshold parameters identified spikes or bursts incorrectly, we chose other sets of parameters. To ascertain the nature of the output activity of the STN neurons, for each recording, we have produced smooth histograms of the interspike intervals (ISIs), burst durations, and interburst intervals (IBIs). We have also plotted the power spectrum and the autocorrelation function of the trace and the ISI return map. A return map or Poincaré map illustrates the serial dependence of ISIs. It consists of a point plot, where each point corresponds to a spike. The abscissa and ordinate of each point are, respectively, the ISI preceding and following the corresponding spike. The 45° line indicates the position of identical values. For a strictly periodic signal of period T, the points of the ISI return map have all the same coordinates, (T,T). If, instead, a spike of a regular spike train is missing, then the return map has points at coordinates (T,T), (T,2T) and (2T,T) (see Fig. 5). To detect slow variations in the spiking frequency within a single burst, we have also plotted the ISI as a function of the time of occurrence of the spike within a burst. This is also a point plot, but in this case, the horizontal coordinate of each point is the time of occurrence within the burst while the vertical coordinate is the interval with the previous spike (examples in Fig. 1, C and D). Finally, we have computed a set of "synchronization indices": we define the synchronization index of order j as the percentage of ISIs equal to j times the stimulation interval within a range of ±1 ms. For example, in a neuron with stimulation indices n1 = 30% and n2 = 60%, 30% of the ISIs are equal to the stimulation interval, while 60% are equal to twice the stimulation interval. Clearly in this example where 90% of the ISIs are locked to the external stimuli, the dynamics is very much driven by the stimulation. The sum of the synchronization indices of order j, and their distribution provides a good quantitative measure of whether the neuron is or is not driven by the external stimulation.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
|
Figure 1 shows the typical response of STN neurons to HFS of the STN with pulses at frequency and duration values commonly used in clinical applications, 130 Hz and 90 µs. Before stimulation, the neuron had a spontaneous bursting activity (Fig. 1, A and C). The interval between spikes (ISI) increased within the burst (Fig. 1C, left). Therefore the plot of the ISI as a function of the time of occurrence of the spikes within the bursts gave a graph where the points were on a curve sloping upwards (see METHODS, Fig. 1C, right). HFS (550 µA) had a dual effect: it forced STN neurons to discharge bursts of evoked spikes and thus totally blocked their spontaneous activity (Fig. 1, B and D). Both these effects lasted the duration of HFS and reversed at the end of HFS after a period of post-HFS silence (not shown). During bursts, all spikes were evoked by a stimulus: the synchronization indices for the recordings in Fig. 1D are n1 = 5%, n2 = 88%, n3 = 4%. The first spikes inside bursts were evoked at 130 Hz but then "missed" one stimulus out of two or three stimuli and were thus separated by an interval that was twice or three times that of the stimulus signal as shown by the plot of the ISI as a function of the time of occurrence of the spike within bursts (Fig. 1D, right). If an interval T regularly separated the spikes inside bursts, the graph would consist of a single horizontal line at height T. Here some points had a vertical coordinate equal to T (7.7 ms), but most of them had it equal to 2T (15.4 ms), 3T (23.1 ms), or 4T (30.8 ms). In this example, 88% of the intraburst spikes were synchronized to one out of two stimuli (Fig. 1, D and E). Bursts, unlike spikes, did not appear at predictable times and among neurons, the periods of silence between bursts were highly variable (Table 1 and Fig. 1F, left). Bursts had also highly variable durations between cells (Table 1, Fig. 1F, right). In summary, HFS at 130 Hz generated a new pattern of STN activity consisting of spikes time-locked to the stimulation and organized in bursts. This HFS-driven activity totally replaced the spontaneous one.
|
Low-frequency stimulation (LFS) of the STN at 10 Hz evoked a completely different response compared with 130 Hz. In all STN neurons tested (n = 57), 10 Hz stimuli either had no effect (at low intensity) or evoked single spikes at 10 Hz (mean ISI: 103 ± 56 ms; Fig. 2). These HFS-evoked spikes were intermingled with the spontaneous ones when spontaneous activity had a frequency >10 Hz. In the opposite case, spontaneous spikes disappeared during LFS (Fig. 2, A and C). This observation is confirmed by the synchronization indices: n1 = 29%, n2 = 19%, n3 = 4% for 2B (top), indicating that only 52% of the ISIs where synchronous with the stimulation, while n1 = 93%, n2 = 1%, n3 = 0% for Fig. 2D (top). In this latter case, nearly all recorded spikes are exactly synchronous with the external stimulation. Moreover, the depolarization of the membrane by high-intensity pulses at 10 Hz increased the probability of spontaneous firing during LFS (Fig. 2A, bottom). Therefore LFS at 10 Hz had a simple excitatory effect: it evoked single spikes at 10 Hz or its subharmonics (Fig. 2, B and D), and it did not block their spontaneous activity (sometimes even increased it), even with high-intensity (200% above the threshold) and long duration (400 µs) pulses (Fig. 2, A–D, bottom). Increasing smoothly the frequency of stimulation 
185 Hz at constant pulse duration (90 µs) and amplitude (200–1000 µA) progressively changed the HFS-evoked response from evoked single spikes to the dual effect described above. At 50 Hz, HFS induced STN neurons to discharge single spikes at 50 Hz or doublets-triplets or trains of spikes at 50 Hz (n = 52). The percentage of each type of response was roughly 30% (data not shown). In contrast, from 80 to 185 Hz, HFS had consistently its dual effect in 85–90% of the tested cells: the generation of a new activity consisting in spikes time-locked to the stimulation and organized in bursts together with the suppression of spontaneous activity (n = 99; Fig. 1B).
Variation of pulse amplitude
We increased the pulse amplitude from 100 to 1,000 µA and kept its duration constant (90 µs). LFS at 10 Hz, as shown in Fig. 2, evoked single spikes at 10 Hz provided that the product of the pulse amplitude and duration was above the spiking threshold. Further increasing the LFS pulse amplitude never switched the pattern of the response to bursting mode. In contrast the activity evoked by HFS at 80, 130, and 185 Hz, evolved with pulse amplitude. HFS at 130 or 185 Hz had no effect or evoked single spikes for intensities 300 µA (Fig. 3, A1 and B1, 300 µA). When we progressively increased the pulse amplitude to 500 µA, spontaneous spikes and evoked single spikes disappeared and were replaced by bursts of evoked spikes (Fig. 3, A1 and B1, 500 µA). At these parameters, HFS-evoked bursting activity was stable in the sense that bursts had a regular temporal structure with most of the spikes evoked 1 of 1–2 or 1–3 stimuli (Fig. 3, A2 and B2, synchronization indices n1 = 42%, n2 = 47%, n3 = 2%, and n1 = 11%, n2 = 68%, n3 = 12%, n4 = 3%, respectively). At amplitudes >500 µA, HFS still evoked bursts but with a far less regular structure. Either the intraburst temporal structure was disorganized, with, for example, spikes only present at the beginning and end of membrane depolarizations (Fig. 3, A1 and B1, 600 and 700 µA) or the cell depolarized and long trains of spikes replaced bursts (Fig. 3 B1, 700 µA). In the former case, interruption of spiking revealed a plateau of 30–35 mV amplitude. At such high intensities, intraburst spikes were evoked one out of 1–4 or 2–7 stimuli in this cell (Fig. 3 A3, B3, synchronization indices n1 = 11%, n2 = 61%, n3 = 2%, n4 = 6%, and n1 = 1%, n2 = 15%, n3 = 18%, n4 = 14%, n5 = 8%, n6, = 7%, n7 = 5%, respectively), and we thus observed many more ISIs outside the displayed range than with lower intensities (see legend). We re-obtained a stable HFS-evoked bursting state by decreasing the pulse amplitude.
|
We increased stimulus duration and kept constant its intensity (400–500 µA). LFS at 10 Hz, as already shown in Fig. 2, evoked single spikes at 10 Hz whatever the stimulus duration, as long as the product of pulse amplitude and duration was above the spiking threshold. In contrast, activity evoked by HFS at 80, 130, and 185 Hz evolved with pulse duration. HFS with pulses of 60 µs duration mostly evoked single spikes (Fig. 4, A1 and B1, 60 µs). When we increased the duration to 90 µs, it generated bursts of evoked spikes with a mean frequency of 65 ± 12 Hz (intraburst frequency defined as the inverse of the average ISI; Fig. 4, A1 and B1, 90 µs). At these parameters, the HFS-evoked bursting activity was stable in the sense that bursts had a regular temporal structure with most of the spikes evoked one out of 1–2 or 2–4 stimuli (Fig. 4, A2 and B2, synchronization indices n1 = 10%, n2 = 80%, and n3 = 4%, and n1 = 12%, n2 = 69%, n3 = 9%, and n4 = 2%, respectively). Figures 3A2 and 4A2 are not identical though studying the same set of HFS parameters because they were obtained from two different populations of STN cells, one being much larger than the other. At widths >300 µs, HFS still evoked bursts but with different characteristics. Either the intraburst temporal structure was disorganized with interruptions of spiking (Fig. 4B1, 400 µs) or long trains of spikes replaced bursts (Fig. 4A1, 400 µs). At such durations, most of the spikes were evoked in this cell every 2–5 or 2–7 stimuli, respectively (Fig. 4, A3 and B3, synchronization indices n1 = 3%, n2 = 41%, n3 = 22%, n4 = 9%, n5 = 4%, n6 = 2% and n1 = 3%, n2 = 66%, n3 = 17%, n4 = 5%, n5 = 3%, n6 = 2%, respectively). ISIs outside the displayed range were much more numerous than with shorter pulse durations (see legend). We re-obtained a stable HFS-evoked bursting state by decreasing the pulse width.
ISI signatures of STN neurons according to stimulation parameters
The return maps of Fig. 5 summarize the results of the present study. To construct the return maps of Fig. 5, A1, A2, and B1, B2, we used the ISIs between all recorded spikes. For Fig. 5, C1, C2, and D1, D2, we used the ISIs between spikes belonging to the same bursts (intraburst ISIs). In the absence of stimulation, the spontaneous firing pattern consisted of isolated spikes (5A1) or bursts (5A2). In both cases, the points that represent the ISIs were clustered around the diagonal. In the former case, this indicates that the spikes were approximately periodic within a single recording, but different recordings contained spikes with different firing periods. In the latter case, an analysis of the ISIs as a function of the time of occurrence within the burst (Fig. 1, A and C) indicated that the ISIs varied slowly within a single burst. Therefore in the case of Fig. 5A2, the points within a single burst move up and down the diagonal as the ISI slowly changes. The ISI patterns referred to as ISIs signatures (Szucs et al. 2005) were totally different under LFS and HFS. In response to 10-Hz stimulation (Fig. 5B), the firing pattern was periodic irrespective of the width of the stimulation pulse (100 ms for B1 and 400 ms for B2) but with many intermingled spontaneous spikes, thus giving rise to the characteristic triangular structure of the return map. The triangular pattern can be understood by considering a succession of periodic spikes P (evoked at 10 Hz) of period T (100 ms) with an intermingled random spontaneous spike S at a random interval R after a periodic spike. Spikes in a periodic train each give rise to a point with coordinates (T, T) on the return map, whereas the spontaneous spike S and its two neighboring periodic spikes give rise to three points with coordinates (T,R), (R,T-R), and (T-R,T) that lie on the three sides of a triangle with vertex (T,T). Different spontaneous spikes will happen at different intervals R after a periodic spike: therefore their corresponding points in the return map cover the three sides of the triangle (Faure et al. 2000). In contrast, in response to 130- and 185 Hz stimulations the ISIs were concentrated at approximately integer multiples of 7.7 or 5.4 ms (intervals between HFS pulses), and the return maps showed distinct clusters around these values with a decreasing probability toward higher values. This indicates that for parameter values close to the therapeutic ones (Fig. 5, C1 and D1), the firing pattern followed the stimulation with most of the spikes evoked every second stimulation pulse. When we increased the stimulation intensity (Fig. 5, C2 and D2) the firing pattern, while still being dominated by the stimulation period, became more irregular and the return map showed spikes separated by 20 times the stimulation period.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
There are several interpretations for the absence of spontaneous spikes inside and between HFS-driven bursts. Spontaneous and evoked afferent synaptic responses (EPSPs and IPSPs) may decrease due to synaptic plasticity (receptor desensitization or transmitter depletion) during high-frequency and long-term stimulation of intranuclear afferents. Also, HFS depresses subthreshold voltage-dependent currents underlying intrinsic pacemaker activity (Beurrier et al. 2001). Between bursts, periods of hyperpolarization that last 0.3–2 s prevent the generation of spontaneous and evoked spikes. We propose that HFS produces regular bursts by resetting the STN somatic membrane in a reproducible region of its parameter space where burst-generating currents undergo voltage-dependent activation or de-inactivation with appropriate kinetics (Elson et al. 1999).
HFS spikes have a regular pattern
HFS-evoked spikes have a totally different pattern during HFS-driven bursts than during bursts recorded in vitro on membrane hyperpolarization in control STN slices (Beurrier et al. 1999; Bevan et al. 2002) or at resting membrane potential in dopamine-depleted slices. Spontaneous spikes inside spontaneous bursts are spaced by highly variable ISIs, whereas HFS-evoked spikes inside HFS-evoked bursts, have a high-frequency (60–80 Hz), are time-locked to the stimulation, spaced by intervals that are multiples of the inter stimuli interval and thus evoked at predictable times (see Fig. 5). The pattern of HFS-evoked spikes may also differ from that during bursts recorded in slow-wave sleep (Urbain et al. 2000), during a conditioned movement (Georgopoulos et al. 1983; Wichmann et al. 1994), or in the parkinsonian state (Bergman et al. 1994; Perier et al. 2000) in the rat and monkey, but a precise analysis of the structure of these activities is lacking.
Comparison of HFS-evoked firing patterns in the STN in vitro and in vivo
Margarinos-Ascone et al. (2002) and Lee et al. (2004) performed in vitro experiments similar to ours but used very short periods of stimulation (10–60 s or 0.1–2 s, respectively). They report excitation of STN neurons (Lee et al. 2004) or a first period of bursting activity before silence (Margarinos-Ascone et al. 2002). However, comparison with our results is rendered difficult by the absence of precise analysis of the relationship between stimuli and recorded spikes. Extracellular recordings in vivo, in patients, in MPTP-treated monkeys or 6-OHDA-treated rats, all report that STN-HFS (100–140 Hz) totally silences STN neurons (Filali et al. 2004; Tai et al. 2003; Welter et al. 2004; Wu et al. 2001). During LFS also, results obtained with extracellular (in vivo) and intracellular (in vitro) recording techniques are discordant. Thus STN-LFS (1–10 Hz) evoked IPSPs, EPSPs, and spikes in STN neurons in slices in vitro (Garcia et al. 2003; Lee et al. 2003, 2004), whereas similar experiments in vivo showed no effect in the STN (Tai et al. 2003; Welter et al. 2004) or inhibition of STN (Filali et al. 2004) activity. In these extracellular studies, LFS-evoked spikes should have been recorded. Stimulus artifacts have a long duration (around 2 ms) in studies relying on extracellular recordings, in particular when the stimulation is close to the recording site and thus probably mask short-latency (1 ms) action potentials evoked by direct stimulation of nearby cell bodies or axons. As a result, experimentalists record only the silencing effect of HFS, i.e., the absence of spikes between artifacts that we also recorded in the present study. This silencing effect persists for several seconds or minutes once HFS is stopped and is thus recorded alone at the end of HFS (Beurrier et al. 2001; Filali et al. 2004; Garcia et al. 2003; Lee et al. 2004; Tai et al. 2003; Welter et al. 2004).
Physiological relevance of the HFS-driven new bursting pattern
Studies on the impact of electrical parameters settings on the clinical signs of Parkinson's disease, bradykinesia, rigidity, and tremor (Moro et al. 2002; Rizzone et al. 2001) indicated that the most beneficial effects are obtained at high frequencies and that voltage is the most critical factor to obtain adequate alteration in STN activity. Effectively, we have shown here that stimulations below and above 50–80 Hz act differently on the spontaneous activity and the firing pattern of STN neurons, provided that the product of the pulse amplitude and duration was largely above spiking threshold. The relevance of the new STN bursting activity imposed by HFS at therapeutic parameters is under question: does it propagate to target neurons and have a role by itself or does it represent a regular output with a lack of information content (Grill et al. 2004). Recordings in target nuclei during STN-HFS showed that at least some of the spikes evoked by HFS in STN neurons propagate to and excite target neurons. In awake MPTP-monkeys, HFS at parameters that ameliorated the motor signs of monkeys generated an excitation in the GPe and GPi time-locked to the stimuli (Hashimoto et al. 2003). Also, HFS of the STN activate some SNr neurons in control rats (Maurice et al. 2003). However, HFS-evoked spikes recorded in target neurons do not appear to be organized in bursts or trains. One explanation could be that STN neurons fire in unsynchronized bursts thus giving a continuous high-frequency STN output. Sequential recordings of different STN neurons in the same slice showed that STN neurons respond with bursts of different durations and frequencies to the same HFS (Garcia et al. 2003), thus suggesting asynchronization of HFS-driven bursts. Also somatic recordings may not be representative of the axonal output because axons are more susceptible to extracellular stimulation than somas as they are activated with pulses of shorter width (Holsheimer et al. 2000; McIntyre and Grill 2002; McIntyre et al. 2004; Nowack and Bullier 1998; Ranck 1975). Thus HFS may depolarize first the STN efferent axons and generate there a regular tonic activity that propagates to the terminals whereas it evokes bursts of regularly spaced spikes in STN somas. Montgomery and Baker (2000) proposed in their modeling study that the therapeutic effect of HFS results from driving neurons at higher and perhaps more importantly regular frequencies. The high frequency and the regularity of HFS both clearly shape the output of the stimulated STN neurons. HFS generates, with a high degree of temporal precision, high-frequency spikes in the upper gamma range. We propose that the signature of the pattern imposed by HFS complements its silencing action and is instrumental in the consequences of HFS in bringing the basal ganglia circuit to operate (Garcia et al. 2005).
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address of P.-O. Fernagut: UCLA, Dept. of Neurology, Reed Neurological Research Center B117, 710 Westwood Plaza, Los Angeles, CA 90095-1769.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: C. Hammond, INMED Inserm, BP 13, 163 route de Luminy, 13273 Marseille Cedex 09 France (E-mail: hammond{at}inmed.univ-mrs.fr)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Benazzouz A, Gross C, Feger J, Boraud T, and Bioulac B. Reversal of rigidity and improvement in motor performance by subthalamic high-frequency stimulation in MPTP-treated monkeys. Eur J Neurosci 5: 382–389, 1993.[CrossRef][ISI][Medline]
Bergman H, Feingold A, Nini A, Raz A, Slovin H, Abeles M, and Vaadia E. Physiological aspects of information processing in the basal ganglia of normal and parkinsonian primates. Trends Neurosci 21: 32–38, 1998.[CrossRef][ISI][Medline]
Bergman H, Wichmann T, and DeLong MR. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 249: 1436–1438, 1990.
Bergman H, Wichmann T, Karmon B, and DeLong MR. The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 72: 507–520, 1994.
Beurrier C, Bioulac B, Audin J, and Hammond C. High-frequency stimulation produces a transient blockade of voltage-gated currents in subthalamic neurons. J Neurophysiol 85: 1351–1356, 2001.
Beurrier C, Bioulac B, and Hammond C. Slowly inactivating sodium current (INap) underlies single-spike activity in rat subthalamic neurons. J Neurophysiol 83: 1951–1957, 2000.
Beurrier C, Congar P, Bioulac B, and Hammond C. Subthalamic nucleus neurons switch from single-spike activity to burst- firing mode. J Neurosci 19: 599–609, 1999.
Bevan MD, Magill PJ, Hallworth NE, Bolam JP, and Wilson CJ. Regulation of the timing and pattern of action potential generation in rat subthalamic neurons in vitro by GABA-A IPSPs. J Neurophysiol 87: 1348–1362, 2002.
Bevan MD, and Wilson CJ. Mechanisms underlying spontaneous oscillation and rhythmic firing in rat subthalamic neurons. J Neurosci 19: 7617–7628, 1999.
Elson RC, Huerta R, Abarbanel HD, Rabinovich MI, and Selverston AI. Dynamic control of irregular bursting in an identified neuron of an oscillatory circuit. J Neurophysiol 82: 115–122, 1999.
Faure P, Kaplan D, and Korn H. Synaptic efficacy and the transmission of complex firing patterns between neurons. J Neurophysiol 84: 3010–3025, 2000.
Filali M, Hutchison WD, Palter VN, Lozano AM, and Dostrovsky JO. Stimulation-induced inhibition of neuronal firing in human subthalamic nucleus. Exp Brain Res 156: 274–281, 2004.[CrossRef][ISI][Medline]
Garcia L, Audin J, D'Alessandro G, Bioulac B, and Hammond C. Dual effect of high-frequency stimulation on subthalamic neuron activity. J Neurosci 23: 8743–8751, 2003.
Garcia L, D'Alessandro G, Bioulac B, and Hammond C. High-frequency stimulation in Parkinson's disease: more or less? Trends Neurosci 28: 209–216, 2005.[CrossRef][ISI][Medline]
Georgopoulos AP, DeLong MR, and Crutcher MD. Relations between parameters of step-tracking movements and single cell discharge in the globus pallidus and subthalamic nucleus of the behaving monkey. J Neurosci 3: 1586–1598, 1983.[Abstract]
Gill SS and Heywood P. Bilateral dorsolateral subthalamotomy for advanced Parkinson's disease. Lancet 350:1224, 1997.[CrossRef][ISI][Medline]
Grill W. Electrically excitable nerve elements: excitation sites in peripheral and central stimulation. In: Brain Stimulation and Epilepsy, edited by H. Lüders. London: Martin Dunitz, 2003, p. 55–66.
Grill WM, Snyder AN, and Miocinovic S. Deep brain stimulation creates an informational lesion of the stimulated nucleus. Neuroreport 15: 1137–1140, 2004.[CrossRef][ISI][Medline]
Hashimoto T, Elder CM, Okun MS, Patrick SK, and Vitek JL. Stimulation of the subthalamic nucleus changes the firing pattern of pallidal neurons. J Neurosci 23: 1916–1923, 2003.
Hassani OK, Mouroux M, and Féger J. Increased subthalamic neuronal activity after nigral dopaminergic lesion independent of disinhibition via the globus pallidus. Neuroscience 72: 105–115, 1996.[CrossRef][ISI][Medline]
Hernandez-Lopez S, Flores G, Rosales MG, Sierra A, Martinez-Fong D, and Aceves J. Muscarinic antagonists microinjected into the subthalamic nucleus decrease muscular rigidity in reserpinized rats. Neurosci Lett 213: 157–160, 1996.[ISI][Medline]
Holsheimer J, Dijkstra EA, Demeulemeester H, and Nuttin B. Chronaxie calculated from current-duration and voltage-duration data. J Neurosci Methods 97: 45–50, 2000.[CrossRef][ISI][Medline]
Lee KH, Chang SY, Roberts DW, and Kim U. Neurotransmitter release from high-frequency stimulation of the subthalamic nucleus. J Neurosurg 101: 511–517, 2004.[ISI][Medline]
Lee KH, Roberts DW, and Kim U. Effect of high-frequency stimulation of the subthalamic nucleus on subthalamic neurons: an intracellular study. Stereotact Funct Neurosurg 80: 32–36, 2003.[CrossRef][ISI][Medline]
Levy R, Dostrovsky JO, Lang AE, Sime E, Hutchison WD, and Lozano AM. Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinson's disease. J Neurophysiol 86: 249–260, 2001.
Levy R, Hutchison WD, Lozano AM, and Dostrovsky JO. High-frequency synchronization of neuronal activity in the subthalamic nucleus of parkinsonian patients with limb tremor. J Neurosci 20: 7766–7775, 2000.
Levy R, Hutchison WD, Lozano AM, and Dostrovsky JO. Synchronized neuronal discharge in the basal ganglia of parkinsonian patients is limited to oscillatory activity. J Neurosci 22: 2855–2861, 2002.
Limousin P, Pollak P, Benazzouz A, Hoffmann D, Le Bas JF, Broussolle E, Perret JE, and Benabid AL. Effect of parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet 345: 91–5, 1995.[CrossRef][ISI][Medline]
Lozano AM, Dostrovsky J, Chen R, and Ashby P. Deep brain stimulation for Parkinson's disease: disrupting the disruption. Lancet Neurol 1: 225–231, 2002.[CrossRef][ISI][Medline]
Magarinos-Ascone C, Pazo JH, Macadar O, and Buno W. High-frequency stimulation of the subthalamic nucleus silences subthalamic neurons: a possible cellular mechanism in Parkinson's disease. Neuroscience 115: 1109–1117, 2002.[CrossRef][ISI][Medline]
Maurice N, Thierry AM, Glowinski J, and Deniau JM. Spontaneous and evoked activity of substantia nigra pars reticulata neurons during high-frequency stimulation of the subthalamic nucleus. J Neurosci 23: 9929–9936, 2003.
McIntyre CC and Grill WM. Extracellular stimulation of central neurons: influence of stimulus waveform and frequency on neuronal output. J Neurophysiol 88: 1592–1604, 2002.
McIntyre CC, Grill WM, Sherman DL, and Thakor NV. Cellular effects of deep brain stimulation: model-based analysis of activation and inhibition. J Neurophysiol 91: 1457–1469, 2004.
Miller WC and DeLong MR. Altered tonic activity of neurons in the globus pallidus and subthalamic nucleus in the primate MPTP model parkinsonian. In: The Basal Ganglia. II. Structure and Function: Current Concepts, edited by Carpenter MB and Jayaraman A. New York: Plenum, 1987, p. 415–427.
Minzly J, Mizrahi J, Hakim N, and Liberson A. Stimulus artefact suppressor for EMG recording during FES by a constant-current stimulator. Med Biol Eng Comput 31: 72–75, 1993.[CrossRef][ISI][Medline]
Montgomery EB Jr and Baker KB. Mechanisms of deep brain stimulation and future technical developments. Neurol Res 22: 259–266, 2000.[ISI][Medline]
Moody CA and Spear LP. Effects of acute dopamine depletion on responsiveness to D1 and D2 receptor agonists in infant and weanling rat pups. Psychopharmacology 107: 39–49, 1992.[CrossRef][Medline]
Moro E, Esselink RJ, Xie J, Hommel M, Benabid AL, and Pollak P. The impact on Parkinson's disease of electrical parameter settings in STN stimulation. Neurology 59: 706–713, 2002.
Nowak LG, and Bullier J. Axons, but not cell bodies, are activated by electrical stimulation in cortical gray matter. I. Evidence from chronaxie measurements. Exp Brain Res 118: 477–488, 1998.[CrossRef][ISI][Medline]
Perier C, Agid Y, Hirsch EC, and Feger J. Ipsilateral and contralateral subthalamic activity after unilateral dopaminergic lesion. Neuroreport 11: 3275–3278, 2000.[ISI][Medline]
Ranck JB. Which elements are excited in electrical stimulation of mammalian central nervous system: a review. Brain Res 98: 417–440, 1975.[CrossRef][ISI][Medline]
Rizzone M, Lanotte M, Bergamasco B, Tavella A, Torre E, Faccani G, Melcarne A, and Lopiano L. Deep brain stimulation of the subthalamic nucleus in Parkinson's disease: effects of variation in stimulation parameters. J Neurol Neurosurg Psychiatry 71: 215–219, 2001.
Szucs A, Abarbanel HD, Rabinovich MI, and Selverston AI. Dopamine modulation of spike dynamics in bursting neurons. Eur J Neurosci 21: 763–772, 2005.[CrossRef][ISI][Medline]
Tai CH, Boraud T, Bezard E, Bioulac B, Gross C, and Benazzouz A. Electrophysiological and metabolic evidence that high-frequency stimulation of the subthalamic nucleus bridles neuronal activity in the subthalamic nucleus and the substantia nigra reticulata. FASEB J 17: 1820–1830, 2003.
Urbain N, Gervasoni D, Souliere F, Lobo L, Rentero N, Windels F, Astier B, Savasta M, Fort P, Renaud B, Luppi PH, and Chouvet G. Unrelated course of subthalamic nucleus and globus pallidus neuronal activities across vigilance states in the rat. Eur J Neurosci 12: 3361–3374, 2000.[CrossRef][ISI][Medline]
Welter ML, Houeto JL, Bonnet AM, Bejjani PB, Mesnage V, Dormont D, Navarro S, Cornu P, Agid Y, and Pidoux B. Effects of high-frequency stimulation on subthalamic neuronal activity in parkinsonian patients. Arch Neurol 61: 89–96, 2004.
Wichmann T, Bergman H, and DeLong MR. The primate subthalamic nucleus. I. Functional properties in intact animals. J Neurophysiol 72: 494–506, 1994.
Williams D, Tijssen M, Van Bruggen G, Bosch A, Insola A, Di L, V, Mazzone P, Oliviero A, Quartarone A, Speelman H, and Brown P. Dopamine-dependent changes in the functional connectivity between basal ganglia and cerebral cortex in humans. Brain 125: 1558–1569, 2002.
Wu YR, Levy R, Ashby P, Tasker RR, and Dostrovsky JO. Does stimulation of the GPi control dyskinesia by activating inhibitory axons? Mov Disord 16: 208–216, 2001.[CrossRef][ISI][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP |
, middle). interspike interval (ISI) as a function of the time of occurrence within a burst averaged over 89 spontaneous bursts (right). D: expanded traces of an HFS-evoked burst from B (*), an HFS-evoked spike (e) inside the burst (middle, 
