Depolarization Block Hypothesis

An early working hypothesis of the mechanism of action of DBS, the depolarization block theory held that DBS inhibits neuronal activity at the site of stimulation (i.e., the DBS electrode target site), leading to decreased output from the stimulated structure. As noted earlier, this hypothesis originated from the observation that the clinical effects of DBS on tremor were near immediate and similar to those of surgically lesioning the same target structure ( ). This similarity suggested that, like surgical lesioning, DBS acts by silencing neurons of the stimulated structure.

Experimental support came from studies in normal and parkinsonian-induced animal models. For example, in the MPTP-treated NHP, lesioning the STN reverses experimental parkinsonism ( ). Evidence from a wide variety of studies on the effects of HFS on the normal and parkinsonian rodent brain also supported this hypothesis. Using ex vivo patch-clamp cell recording techniques in a rat slice preparation, demonstrated that HFS blocked action potential generation in STN neurons in the poststimulation period. This finding suggested that the inhibitory effect of HFS was due to blockage of voltage-sensitive Na ⁺ channels (depolarization block). Similarly, demonstrated that sustained HFS applied to rat brain slices could depolarize the membrane potential and trigger action potentials that subsequently led to total silencing of cells in the STN. They suggested that the silencing effect of HFS is not due to a frequency-dependent presynaptic depression, but rather to the gradual inactivation of Na ⁺ -mediated action potentials. In contrast to this proposed sodium inactivation mechanism, demonstrated in the rat brain slice of the entopeduncular nucleus (human GPi) that elevated K ⁺ concentrations depressed neuronal activity in a way similar to that of HFS.

In vivo HFS of the STN in normal and 6-OHDA-lesioned rats was found to decrease activity in the substantia nigra pars reticulata (SNr) and entopeduncular nucleus and increase activity in the globus pallidus externa (GPe) and ventral lateral nucleus of the thalamus ( ). These very early findings suggested that HFS has principally inhibitory effects in the BGTC network similar to those of STN lesions. In addition, showed that, unlike systemic pretreatment with metabotropic and ionotropic glutamate receptor or γ-aminobutyric acid (GABA) antagonists, in vivo HFS of the STN was able to suppress local spontaneous neuronal activity and spike generation, suggesting that the suppression was mediated by voltage-gated ion channels. placed a recording electrode within 600 μm of the stimulating electrode in the STN of patients with PD undergoing DBS surgery and found that stimulation produced inhibition of most recorded cells. Similarly, it has been shown that HFS of the GPi reduces firing frequency in the MPTP-treated NHP model ( ), as well as in the GPi of patients with PD ( ).

Additional support for the depolarization block hypothesis came from metabolic studies of STN cells during HFS. For example, found a reduction in cytochrome oxidase I (CoI) mRNA levels (a marker of neuronal metabolic activity) in STN cells during STN stimulation in the 6-OHDA-lesioned rodent. Similarly, using in vivo extracellular recordings and histochemistry, found that, relative to normal rats, STN stimulation reduced cellular firing and CoI mRNA in cells in the STN and the substantia nigra pars compacta (SNc) in 6-OHDA-lesioned rats.

In marked contrast, an electrophysiological study of patients with PD by showed that the neuronal firing rate in the STN was unchanged during and upon termination of STN DBS. A subset of cells displayed altered firing patterns with a predominant shift toward random firing. The distinct differences between the findings of and those of and may be attributed to the fact that, unlike the earlier studies, Carlson et al. used clinically relevant stimulation parameters (3–5 V, 80–200 Hz, 90- to 200-μs pulses; 33 neurons/11 patients) delivered through a clinical DBS electrode (Medtronic Lead No. 3389 human DBS four-contact electrode), rather than a microstimulating electrode. Carlson et al. concluded that, rather than inactivating the STN, DBS provides a null signal, or retrograde activation of STN efferents, via the STN to the BGTC that has been altered as a consequence of PD.

Neural Jamming/Neural Modulation Hypothesis of Tremor

The neural jamming or neural modulation hypothesis states that DBS regulates and corrects pathological neural firing in the BGTC network. Important neurophysiological studies on normal and pathological states in the BGTC network reveal highly specific changes in cellular activity during seizures as well as in essential tremor and PD tremor. Computer simulations have been used to model the effects of various stimulation frequencies on the regularity and synchronization of neuronal firing activity and on information transfer between synaptically connected neurons. The results suggest that HFS causes an informational lesion, which either returns pathological signaling to a normal firing pattern or desynchronizes abnormally synchronized or regularized oscillations ( ).

Understanding the fundamental principles of neural jamming requires a detailed knowledge of neuronal ionic conductances as well as normal firing patterns within the BGTC network. For example, STN ( ) and thalamic neurons ( ) are able to fire in both tonic and burst modes because they have intrinsic membrane properties that generate rhythmic oscillations (from 10 to 30+ Hz, β and γ frequencies, respectively) ( ). As described by Bevan, these membrane properties result from several different ionic conductances that include a channel hyperpolarizing-activated current (I _h ) and a depolarizing potassium-activated calcium current (I _CaK ) and sodium current (I _NaP ) ( ). The interplay between I _NaP and I _CaK produces rhythmic activity in the 10 to 30 Hz range. As a result, depending on the input, STN neurons have two preferential frequencies: slow rhythmic firing and high-frequency firing. However, other investigators have shown that STN neurons can exhibit multiple neuronal firing patterns that are predominantly irregular (55%–65%), but also can be in tonic (15%–25%) and in burst (15%–50%) modes ( ). In patients with tremor associated with PD, high-amplitude irregular spike patterns and periodic spiking behavior have been recorded in the STN, which have led to the following classification: (1) tremor cells (2–6 Hz), (2) cells with high (greater than 10 Hz)-frequency periodic activity, and (3) a combination of each ( ). Interestingly, in patients with PD who have limb tremor, many STN neurons display high-frequency oscillations (15–30 Hz) with a high degree of in-phase synchrony ( and references therein). Together, these results suggest that high-frequency synchronized oscillatory activity is associated with the pathology that gives rise to PD tremor.

Importantly, local field potentials recorded in human STN suggest that an increase in oscillatory activity in the β-frequency range may be important in PD tremor ( ). Although the exact relation between oscillatory activity and PD tremor remains to be determined, these results support the previously cited research suggesting that synchronized oscillatory activity may be associated with tremor in patients with PD. Similarly, a study of dopamine-depleted MPTP-treated NHPs suggests that the benefit of STN HFS might be at least partially attributed to the reduction in oscillatory activity in the STN network and consequently in the entire BGTC network ( ).

Additional studies are beginning to elucidate the neural network mechanisms that may be responsible for the pathologic BG–thalamic oscillatory activities that are considered important mechanisms of tremor and certain types of seizures. For example, in vitro recordings of STN neurons have shown them to be part of a neural network that has reciprocal connections with the GP ( ) and generates synchronized oscillations ( ). In this system, the GP releases GABA onto STN neurons, causing inhibitory postsynaptic potentials (IPSPs) and a subsequent rebound Ca ²⁺ spike by activation of a low-threshold calcium channel (I _t ). The rebound calcium spike then allows high-frequency action potential generation in STN neurons that, in turn, release glutamate onto GP neurons. Glutamatergic excitation of GP neurons then sets the stage for the next cycle of oscillation within the STN and GP network. These oscillations may be involved in tremor generation in patients with PD and their disruption may be an important mechanism through which STN HFS diminishes tremor ( ).

A remarkably similar circuitry mechanism also exists in the thalamus that, under pathological conditions, generates network oscillations and spindle waves resulting in tremor and absence epilepsy. Normal spindle waves are 1- to 3-s epochs of synchronized, 7- to 14-Hz oscillations that are generated from interactions between thalamocortical and thalamic nucleus reticularis (nRt) neurons ( ). During a spindle wave, the generation of a burst of action potentials in the GABAergic neurons of the nRt results in 2- to 10-mV IPSPs in thalamocortical neurons ( ). A subset of thalamocortical neurons generates a rebound low-threshold Ca ²⁺ spike that leads to burst firing activity in these neurons. Stimulation of these neurons, in turn, elicits a barrage of excitatory postsynaptic potentials (EPSPs) and activation of low-threshold Ca ²⁺ spikes in nRt neurons, thus initiating the next cycle of the spindle wave ( ).

Spindle waves generalize through the progressive recruitment of neurons into this oscillation, presumably because of axonal interconnections between thalamocortical and nRt neurons ( ). Interestingly, both tremor and absence seizures appear to involve abnormal oscillatory activity in the thalamus, at a frequency of 3 to 6 Hz for tremor ( ) and 3 Hz for absence seizures ( ). Spindle waves are normally mediated through the activation of GABA _A receptors on thalamocortical neurons. Surprisingly, when these receptors are blocked with bicuculline (a GABA _A antagonist), the spindle waves are transformed into events that resemble absence seizures. During normal spindle waves, the IPSPs are about 100 msec in duration. Blockade of GABA _A receptors prolongs the IPSPs to a duration of 300 ms, and the oscillation slows from 6 to 10 Hz to about 3 Hz. Because the intrinsic harmonics of the thalamocortical cells (3 Hz) match that of the thalamocortical–nRt loop (also at 3 Hz), these bursts become very strong, resulting in the generation of a massive synchronized discharge at ∼3 Hz. In this manner, normal spindle waves in vitro can be perverted into those that cause absence seizure–like events ( ).

As has been mentioned earlier, HFS applied to the thalamus leads to immediate tremor arrest and a rapid reversal when stimulation ceases ( ). The depolarization of thalamocortical neurons via HFS is likely to be capable of abolishing spindle waves, tremor, and 3-Hz absence seizure–inducing oscillations owing to the inhibition of rebound responses, which are required for driving nRt/perigeniculate nucleus neurons to discharge in synchrony. Thalamic slice studies support this hypothesis in that application of either neurotransmitters ( ) or HFS ( ) resulted in a marked depolarization of thalamocortical neurons and abolished both spindle and 3-Hz absence seizure–like oscillations. In this manner, HFS-induced neurotransmitter release in the thalamus or STN ( ) may “jam” abnormal oscillations that lead to tremor and absence epilepsy. Thus, in a similar manner, clinical DBS may abolish synchronous oscillatory activities such as those that generate tremor and seizures.

In , de Hemptinne et al. posited that phase–amplitude coupling may be an as-yet unstudied mechanism of STN DBS for patients with PD. Excessive phase–amplitude coupling is thought to suppress cortical information processing, and these investigators found that coupling between low-frequency rhythm phases and the amplitude of broadband activity (between 50 and 200 Hz), often considered a measure of neuronal spiking and synaptic activity, was increased in patients with PD. During acute DBS, while patients performed a movement task, there was a significant reduction in phase–amplitude coupling, suggesting yet another mechanism by which DBS reduces symptoms of PD.

The neural jamming or neural modulation hypothesis remains an important hypothesis toward understanding how DBS regulates and corrects pathological neural firing in the BGTC network. These clinical studies ( ) show promise in the development of closed-loop paradigms using narrowband γ signals as a control signal in patients with PD with dyskinesia using the Medtronic Activa PC+S with the Nexus-D and -E updates that allow for real-time sensing and stimulation updates.

Synaptic Depression Hypothesis

Related to the synaptic modulation hypothesis, the synaptic depression hypothesis posits that a neuron activated by DBS is unable to sustain high-frequency action on its efferent targets owing to depletion of terminal vesicular stores of neurotransmitters, such as glutamate and GABA ( ). For example, patch-clamp cellular recordings from giant synapses in the mouse auditory brainstem have shown that short-term synaptic depression brought about by HFS (greater than 100 Hz) can be largely attributed to rapid depletion of a readily releasable pool of vesicles. In addition, it has been found that HFS of presynaptic terminals significantly enhances the rate of replenishment of the vesicular pool, and that Ca ²⁺ influx through voltage-gated Ca ²⁺ ion channels is the key signal that dynamically regulates the refilling of the releasable vesicular pool in response to differing patterns of inputs ( ). Thus, synaptic depression is usually attributed to depletion of some pool of readily releasable vesicles ( ). used voltage-sensitive dye imaging and field potentials in in vitro studies of thalamocortical afferent axons from mouse brain slices to demonstrate a reduction in cortical activity with incremental increases in thalamic stimulation. A significant decrease in activity was observed above 20 Hz. They concluded that this reduction in cortical activity was a result of synaptic transmission failure by transmitter depletion.

α-Synuclein, a protein that enables neurotransmitter release at presynaptic terminals, is associated with docking vesicles to enable neurotransmitter release, and mutations of this protein have been implicated in the progression of PD ( ). found a deficiency of docked vesicles in α-synuclein-knockout mice, but undocked vesicles were unaffected. Electrophysiological recordings in CA1 synapses of hippocampal slices in these mice demonstrated normal basal synaptic transmission; normal paired-pulse facilitation, an example of synaptic plasticity in which EPSPs are increased when evoked by two impulses in close succession; and normal firing response to a brief train of HFS (100 Hz, 40 pulses). However, an extended train of repetitive HFS (12.5 Hz, 300 pulses) impaired synaptic glutamate release. In these synapses, glutamate was depleted from both docked and reserve pool vesicles, suggesting that α-synuclein is necessary for normal vesicle functioning. Additionally, the replenishment of docked vesicles was slower in mice lacking α-synuclein. These results suggest that the presence of the protein may be a key factor in facilitating the effectiveness of synaptic depression via HFS because the release of the excitatory neurotransmitter glutamate was impaired in mice lacking α-synuclein following extended HFS.

To assess the influence of HFS on axons projecting to the cortex, performed whole-cell recordings in the primary motor cortex (M1) and ventral thalamus of rat brain slices and found that M1 neurons depolarize in response to HFS, but return to prestimulation membrane potential upon cessation of stimulation. HFS depressed excitatory synaptic currents, which produced postsynaptic deafferentation by depleting presynaptic neurotransmitters. These results suggest that HFS of subcortical axonal white matter tracts that project to the motor cortex may affect responses in cortical neurons.

also addressed the effects of HFS on axons. They proposed that orthodromic spikes, which move from cell to terminal, may decouple stimulated nuclei from axons during DBS. This decoupling would prevent neuronal communication as a result of the stimulation-driven depletion of terminal neurotransmitter release. Similarly, found that antidromic spikes, which move from terminal to cell, prevented communication to the nucleus from afferent inputs by colliding with spontaneous orthodromic spikes. These results have been described as a “reversible functional deafferentation” of the nucleus. It is possible that DBS interrupts the pathologic synchronized activity of cortical inputs to the stimulated nucleus characteristic of PD and allows the stimulated nucleus to enter a new resting state. In this sense, the synaptic depression model of DBS could be considered an expansion and deeper explanation of the depolarization block hypothesis.

Synaptic Modulation Hypothesis

The synaptic modulation hypothesis states that, depending on the stimulation parameters, DBS activates excitatory and/or inhibitory neuronal elements, including fibers of passage that are in close proximity to the DBS electrode ( ). Excitation of inhibitory neurons would be expected to result in local synaptic inhibition via the release of inhibitory neurotransmitters such as GABA ( ). Similarly, excitation of excitatory neurons through the release of glutamate, an excitatory neurotransmitter, would be expected to result in local synaptic excitation ( ). Originally, the synaptic modulation hypothesis was limited to local changes in the excitation of inhibitory and excitatory neurons, but was later expanded to include the inhibition as well as the excitation of neurons and to include both local and distal modulation.

Evidence for local stimulation-induced synaptic modulation came from extracellular electrophysiological recordings in rodents undergoing HFS as well as in patients with PD undergoing DBS surgery. These findings also suggested that the effects of HFS may be site specific. For example, the previously mentioned study by found that recordings of the GPi during GPi HFS revealed an inhibition of spontaneous neural activity lasting 10 to 25 m. The duration of this inhibitory activity corresponds to that of a typical GABAergic IPSP. This finding suggests that GPi HFS selectively activates GABAergic axon terminals synapsing in the GPi. These axons probably arise from small spiny GABAergic neurons in the caudate that densely project to the GPi and, to a lesser degree, from GABAergic projections from the GPe ( ). These projections are thought to elicit GABA release to strongly inhibit GPi neurons ( ).

The GPi also receives excitatory glutamatergic inputs from the STN ( ), and thus, excitatory afferent terminal inputs from the STN are also activated by the stimulation. However, the predominance of GABAergic inputs to the GPi probably overcomes this glutamatergic excitation ( ). These results are consistent with the observation that muscimol, a GABA agonist, applied to the VIM of the thalamus ( ) and STN ( ) of patients with PD results in a therapeutic benefit comparable to that of DBS applied to the same regions.

As mentioned in the previous section, the electrophysiological recordings by revealed no change in STN firing rates in patients with PD during bouts of HFS, but did demonstrate random firing in a subset of neurons within the STN, suggesting that a more complete description of the effects of DBS would include both local synaptic inhibition and excitation, depending on the stimulation site (i.e., STN vs. GPi). In line with this hypothesis, using sharp electrode intracellular recording techniques in the rat STN, demonstrated that HFS induced both EPSPs and IPSPs. These postsynaptic potentials were completely blocked by bath application of glutamate and GABA antagonists, suggesting that HFS resulted in excitatory and inhibitory neurotransmitter release in the STN. It is thought that the excitatory inputs to the STN originate in the cerebral cortex ( ) and that the inhibitory inputs derive from the GPi ( ). If so, the EPSPs and IPSPs seen during STN HFS may result from stimulation of both descending cortical inputs to the STN, which generate EPSPs via glutamate release, and descending GPi inputs to the STN, which generate IPSPs via GABA release. A study by confirmed that GABA _A and GABA _B receptors mediate the inhibitory responses evoked by GPi HFS. Altogether, the findings suggest that STN HFS activates predominantly excitatory afferent axons in the STN, an effect that appears dependent on the weighted composition of inhibitory and excitatory inputs to the STN, favoring cortical excitation compared to GPi inhibition.

Similarly, an optogenetics study by demonstrated that parkinsonian symptoms in rats could be mitigated by activating afferent cortico-STN axons without activating STN efferent axons. These results suggest that the efficacy of STN DBS for PD may be predominantly, if not solely, dependent on activation of STN afferent axons originating in the cortex.

In parallel with local synaptic inhibition and excitation, DBS may also affect distal synaptic excitation by activating stimulation site axons that release excitatory amino acid neurotransmitters, such as glutamate or aspartate. Projections from the STN are thought to be glutamatergic ( ), and HFS activation of these axons would be expected to increase glutamate release in STN target structures, such as the GP and SNr ( ). Indeed, the activity of SNr cells has been shown to increase during STN stimulation. Given that the latency of the evoked excitation was consistent with the conduction time of subthalamonigral neurons, the increased SNr activity was probably due to activation of glutamatergic subthalamonigral projections ( ). In addition, found that STN HFS in MPTP-treated NHPs generates a short-latency excitation and an increase in the mean discharge rate of GPe and GPi neurons together with a stimulus-synchronized regular firing pattern. GPi stimulation has also been shown to inhibit thalamic target cells as a result of activation of inhibitory GABAergic GPi efferent axonal projections ( ). However, these responses may be secondary to excitation of efferent fibers from the stimulated nuclei that nonetheless contribute to the therapeutic changes in the temporal firing pattern of the BGTC network.

Metabolic studies of STN HFS concur with findings of distal synaptic excitation and inhibition. For example, when MPTP-treated NHPs received 10 days of STN HFS followed by evaluation for 2-deoxyglucose (2-DG) (synaptic activity) and CoI (metabolic activity) mRNA, found an increase in 2-DG uptake in the STN and a decrease in the GPi, with a concurrent increase in CoI mRNA. Thus, in addition to distal synaptic excitation, HFS may cause distal synaptic inhibition by activating efferents of the stimulated nuclei that release inhibitory amino acid neurotransmitters, such as GABA or glycine.

Microdialysis studies in rats support the theory that the therapeutic effects of STN HFS may be related to the selective increase in inhibitory and excitatory neurotransmission within target nuclei of STN efferents. Unilateral STN HFS was found to induce significant bilateral increases in striatal glutamate and GABA release in both intact and (6-OHDA) lesioned rodents ( ). Similar studies of STN HFS revealed a significant increase in extracellular glutamate concentration in the ipsilateral entopeduncular nucleus and SNr of rodents, whereas GABA was augmented only in the SNr ( ). No modifications of GABA were observed in the GP, regardless of the frequencies applied, whereas in the SNr, GABA increased only when HFS increased from 60 to 350 Hz. Glutamate release in the GP and SNr was maximal at 130 Hz with no change with up to 350 Hz stimulation ( ). Thus, STN HFS produces frequency-dependent release of various excitatory and inhibitory neurotransmitters in efferent target nuclei.

The “distal synaptic modulation hypothesis” of DBS has been explored using computational models of both GPi and STN exposed to clinical DBS stimulation parameters ( ). The results showed an increase in axonal firing independent of the soma. This decoupling of axonal and soma responses helps explain why extracellular recordings can show inhibition of the soma and excitation of efferent targets during DBS. also demonstrated that this separation between axonal and somatic activity could be found in STN and GP cells. Overall, converging electrophysiological evidence and computational models point to a combination of DBS effects, which includes both local and distal inhibitory and excitatory synaptic modulation.

More evidence for synaptic modulation comes from studies of the effects of HFS on the release of dopamine, a key neurotransmitter in the pathology of PD. Degeneration of the nigrostriatal dopaminergic pathway is a well-known cause of PD, and it is thought that the release of dopamine from surviving dopaminergic neurons projecting to the BG may contribute to the therapeutic effects of DBS for PD. As mentioned, the SNc is one of the structures affected by STN DBS. It contains dopaminergic cell bodies comprising the nigrostriatal projection to the BG. Glutamate-containing axonal terminals have been identified that arise from the STN and make synaptic contact with dopaminergic dendrites within the SNr ( ). The finding that STN DBS may increase dopaminergic nigrostriatal activity is supported by electrophysiological studies showing that STN HFS increases firing of identified dopaminergic SNc neurons recorded either extracellularly ( ) or intracellularly ( ). The EPSPs in these neurons are thought to arise from a direct monosynaptic excitatory glutamatergic input from the STN ( ). Similarly, reported that STN HFS resulted in glutamate release in the STN, as well as the generation of EPSPs and action potentials in SNc neurons, as shown in Figs. 17.2 and 17.3 ( ). Thus, STN HFS-evoked glutamate release in the SNc may increase firing of dopaminergic neurons, which, in turn, enhances dopamine release in the BG .

Glutamate release in the subthalamic nucleus (STN).

Positioning of a glutamate sensor adjacent to a bipolar stimulating electrode in the STN (A) permitted the recording of glutamate release at the stimulation site in response to different (B) durations, (C) intensities, and (D) frequencies of electrical stimulation in the STN of anesthetized rats. AP , action potential.

Subthalamic nucleus (STN)-evoked substantia nigra pars compacta firing.

(A) High-frequency stimulation of the STN in rat brain slices evoked excitatory postsynaptic potentials and action potential generation ( red boxes ) in dopaminergic neurons in the substantia nigra pars compacta (A). (B) Enlarged view of a portion of the stimulated response from (A) ( red dashed lines ).

Clinically, bilateral STN stimulation improves the majority of PD motor symptoms, significantly decreases the dosage level of levodopa required to control motor symptoms ( ), and ameliorates fluctuations in motor control and dyskinesias resulting from long-term high dosage use of levodopa in a way that is quantitatively comparable to results obtained with levodopa alone ( ). In addition, in patients who continue the use of levodopa with DBS, albeit at a substantially lower dose, the therapeutic effects of STN stimulation occur in the dopamine-off period (relatively low blood levels of levodopa), but not during the dopamine-on period (peak blood levels of levodopa) ( ). This finding suggests that STN DBS mediates these beneficial effects, in part, via subtle modulation of dopaminergic transmission in the BG in concert with pharmacological elevation of BG extracellular dopamine levels. STN HFS can even cause dyskinesias that resemble those seen with excess or long-term high dosage use of levodopa ( ). Thus, an increase in extracellular levels of striatal dopamine may contribute to the efficacy of STN HFS via modulation of the BGTC network in patients with PD.

In support of this dopaminergic hypothesis, HFS of the rat STN has been shown to activate SNc dopaminergic cells via a glutamate N -methyl- d -aspartate (NMDA) receptor-dependent mechanism ( ), as well as increasing extracellular dopamine concentrations in the SNc, a neurochemical response requiring depolarization of dopaminergic cell bodies ( ). Consistent with these findings, several studies in animals using in vivo microdialysis have shown that STN HFS in normal swine and NHP models and in 6-OHDA-lesioned rat models increases extracellular levels of dopamine in the striatum ( ) or the dopamine metabolites dihydroxyphenylacetic acid and homovanillic acid ( ). With one exception ( ), STN HFS-evoked increases in striatal dopamine dialysate could not be detected without first inhibiting dopamine reuptake with nomifensine, a dopamine reuptake inhibitor, and stimulating for prolonged durations (20 min) ( ).

Although in vivo monitoring of slow (minutes to hours) changes in dopamine release is easily accomplished using conventional microdialysis methods, real-time amperometric monitoring of dopamine permits detection of more rapid changes in extracellular dopamine release in the absence of the dopamine reuptake inhibition that can occur with STN HFS ( ). Fixed potential amperometry is an in vivo electrochemical method for measuring neurotransmitter release in which carbon-based microelectrodes detect the current associated with oxidation of electroactive compounds such as dopamine. The dynamics of dopamine release in the nucleus accumbens and striatum during electrical stimulation of ascending dopaminergic pathways in rats has been described and quantified by a number of investigators using fixed-potential amperometry and other in vivo electrochemical recording techniques, such as fast scan cyclic voltammetry (FSCV) ( ).

Using the in vivo electrochemical method of amperometry, examined striatal dopamine responses in the rat evoked by STN HFS or HFS of dopamine axons of passage in the adjacent nigrostriatal dopaminergic pathway. STN HFS evoked a two-component effect on striatal dopamine release, with the first characterized by a peak increase in dopamine release within ∼0.4 s that decayed back toward prestimulation baseline levels within ∼1 s. The second component was characterized by a steady-state level of dopamine release sustained ∼30% above prestimulation baseline over the course of the application of HFS ( Fig. 17.4A ). In marked contrast, stimulation of tissue immediately dorsal to the STN, containing ascending dopaminergic axons, resulted in a 10-fold greater increase in the dopamine response that plateaued after ∼5 s but remained elevated over the course of HFS. As shown in Fig. 17.4B , using comparable stimulation sites and amperometric recording procedures, similar differences were observed in the magnitude and temporal pattern of dopamine release in the striatum of the awake NHP ( ).

Nonhuman primate (NHP) and rat fixed-potential amperometry response to subthalamic nucleus (STN) high-frequency stimulation (HFS).

(A) Continuous HFS of the STN evoked a transient release of striatal dopamine that peaked within 20 applied pulses at 50 Hz ( red line ), compared to an 11-fold greater and more sustained dopamine release in response to stimulation dorsal to the STN ( blue line ; note the right y axis is 10× greater than the left y axis). (B) STN stimulation (200 Hz) in an awake NHP evoked striatal dopamine release in response to brief (20 pulses, blue line ) and continuous ( red line ) STN stimulation. Inset: Expanded time frame for the mean ± SEM initial responses to continuous and brief STN stimulation.

Together, these in vivo neurochemical data fit well with electrophysiological and microdialysis studies showing that STN HFS increases action potential firing in STN and SNc neurons ( ). The finding that stimulating ascending dopaminergic fibers dorsal to the STN resulted in a greater release of striatal dopamine than STN stimulation suggests that stimulation of tissue immediately dorsal to the STN may provide a better means of enhancing dopamine release in the BG. Indeed, several clinical studies have shown that reduction of PD motor symptoms correlates with the location of the stimulating electrode, with the greatest improvement occurring when the DBS electrode projects onto white matter dorsal to the STN ( ), including the dorsolateral border of the STN ( ). The anatomical correlates of this location may be the pallidothalamic bundle (including Field H of Forel and the thalamic fascicle), the pallidosubthalamic tract, and/or the zona incerta. The axons of SNc dopaminergic neurons are themselves immediately dorsal to the STN and fall within the region of maximum stimulation efficacy ( ). Several retrograde and anterograde tract-tracing studies have shown that ascending dopaminergic axons originating from the SNc have collateral inputs to the STN ( ). In addition, despite the considerable degeneration of the ascending SNc dopaminergic neuronal system in PD, demonstrated that FSCV can be used to quantify extracellular dopamine fluctuations in the striatum of patients undergoing DBS surgery. Taken together these findings suggest that HFS of the STN and tissue dorsal to the STN may activate ascending dopaminergic fibers of passage.

In contrast to these results, studies of patients with unilateral DBS of the STN for PD that used [ ¹¹ C]raclopride positron emission tomography (PET) scanning to assess STN stimulation-evoked dopamine release in the BG have failed to show dopamine release during electrical stimulation of the STN, despite significant symptom improvements on the motor scores of the Unified Parkinson’s Disease Rating Scale ( ). Based on endogenous dopamine competition with postsynaptic receptor binding of [ ¹¹ C]raclopride, stimulation-induced enhancement of dopamine release in the BG would result in a reduction in the PET signal. These PET studies suggest that STN DBS does not mediate its anti-PD effects via the release of dopamine. However, PET scanning with raclopride has relatively poor temporal resolution and sensitivity and it requires an increase of greater than 90% relative to baseline to detect a change in dopamine release ( ). In addition, adaptive changes in dopamine receptor populations (D2 receptor internalization and recycling) occurring over long-term STN HFS have been suggested as interfering with PET quantification of dopamine release in patients with PD ( ). That said, a more recent PET study has shown that baseline synaptic dopamine levels in patients with PD are significantly increased by STN DBS ( ).

A combination of functional magnetic resonance imaging (fMRI) and neurochemical recordings has also been used to track DBS-evoked changes in neural activation. Using DBS-evoked fMRI blood oxygen level-dependent (BOLD) signals to guide the placement of neurochemical recording electrodes in NHPs, was able to successfully record STN DBS-evoked dopamine release in the caudate and putamen ( Fig. 17.5 ). Specific relationships between the site of stimulation within the STN and the changes in neurochemical activity were mapped, and it was found that DBS-evoked dopamine release can be minimized or maximized through subtle changes in the site of STN stimulation.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Fundamentals of Burst Stimulation of the Spinal Cord and Brain Gene-Based Neuromodulation Spinal Cord Stimulation for the Treatment of Low Back Pain Epilepsy : Anatomy, Physiology, Pathophysiology, and Disorders Colonic Electrical Stimulation for Constipation Transcranial Alternating Current Stimulation and Transcranial Random Noise Stimulation

Stay updated, free articles. Join our Telegram channel

Join