Nature Medicine 14, 17 - 19 (2008)
doi:10.1038/nm0108-17

The movers and shakers of deep brain stimulation

Wael Asaad1 & Emad Eskandar1

Deep brain stimulation is increasingly used in the treatment of Parkinson's disease, essential tremor and other disorders, yet its mechanism of action remains unknown. New findings suggest that at least some of its action involves the release of adenosine, dampening tremors (pages 75–80).
Introduction

Deep brain stimulation (DBS) has been used successfully in the treatment of neurologic illnesses such as Parkinson's disease, tremor and dystonia, and it is currently being explored as a treatment for major depression and obsessive-compulsive disorder. Yet, empirical efficacy aside, little is known of the mechanisms by which the electrical stimulation of deep brain structures alleviates these conditions. In fact, there is not even widespread agreement as to whether such stimulation facilitates, impedes or 'overwrites' information passing through the stimulated nuclei1.

Nevertheless, the reversible and adjustable nature of DBS has made it an attractive, if rather blunt, tool for treating an increasingly large number of problems—when the only surgical tool available is a hammer, every disorder starts to look like a nail. Hence, it is of immense importance to understand the mechanistic basis of DBS in order to improve, revise or expand its application in a more rational fashion. Toward this end, Bekar et al.2 show that, at least in the case of tremor, adenosine may have a major role in mediating the therapeutic efficacy of DBS.

Adenosine, a neuromodulator found throughout the brain, exerts its postsynaptic effect through G protein–coupled receptors3, 4. In particular, the A1 class of receptors tends to decrease the activity of adenylyl cyclase and open potassium channels, thus hyperpolarizing neurons and rendering them less active. This ability to dampen neural activity is at the heart of the findings by Bekar et al.2, demonstrating that adenosine is a product of DBS and inhibits tremor.

Tremor is an involuntary, rhythmic movement that can occur in isolation, for example in benign or 'essential' tremor, or as part of a more elaborate movement disorder, such as Parkinson's disease. Benign tremor is a 5–8-Hz oscillation that is most pronounced during purposeful movement. It generally affects the hands or feet, but can involve other parts of the body such as the head. The tremor of Parkinson's disease, in contrast, is usually slower—closer to the 4–5-Hz range—and is most prominent at rest; it lessens with movement. Furthermore, whereas the cerebellum is believed to have a central role in the etiology of benign tremor, it is probably only indirectly involved in parkinsonian tremor.

Despite these differences, DBS of the thalamic cerebellar relay nucleus (the ventral intermediate nucleus) has been successfully applied to the treatment of both benign and Parkinson's disease tremors5, 6.

The onset of DBS dampens tremor nearly immediately. When the stimulation is turned off, the tremor resumes without delay. The stimulation itself consists of a continuous train of relatively high-frequency (>130 Hz) pulses delivered through wires implanted in the target brain nucleus and powered by a small pulse generator implanted under the clavicle. The system is open loop, meaning that stimulation is delivered without feedback modulation. Moreover, the effects of electrical stimulation are complex, influencing cell bodies, dendrites and axons in potentially different ways and probably causing the release of a myriad of neurotransmitters and neuromodulators. Bekar et al.2 illuminate one path through this thicket of possible mechanisms by studying the effects of DBS in mouse brain slices and then in an in vivo mouse model of tremor (Fig. 1).


Two electrodes planted into the thalamus of a transgenic mouse expressing yellow fluorescent protein (in axons) under the Thy1 promotor. Astrocytes were stained against glial fibrillary acidic protein (red). Note the intense gliosis around the electrode tips. Bekar et al.2 implicate adenosine in the effects of deep brain stimulation, a treatment with a mysterious mechanistic basis.


First, in thalamic slices, the authors showed that high-frequency stimulation (HFS, the experimental analog of therapeutic DBS) impedes synaptic transmission via adenosine2. Specifically, HFS transiently reduced the amplitude of excitatory postsynaptic potentials both in the stimulated (homosynaptic) pathway and, more prominently, in separate converging (heterosynaptic) pathways. Complementing this electrophysiological effect, HFS resulted in a calcium-independent (nonvesicular) release of ATP into the extracellular space; this ATP was then metabolized to adenosine by an ecto-ATPase. Importantly, application of an adenosine (A1) receptor antagonist or an ecto-ATPase inhibitor blocked the HFS-induced reduction in excitatory postsynaptic potential amplitude.

Next, Bekar, et al.2 tested the effect of A1 antagonists on a mouse model of tremor. They administered the MAO-A inhibitor harmaline systemically, resulting in the development of a generalized tremor that, although higher in frequency and evident at rest as well as during movement, is believed to mimic key aspects of human essential tremor. Local application of adenosine into the thalamus decreased the magnitude of the tremor with an efficacy comparable to that of HFS of the thalamus. Conversely, infusion of an adenosine A1 receptor antagonist augmented the tremor. In addition, in the presence of this antagonist, low electric currents produced side effects that otherwise were seen at only higher stimulation levels. Unfortunately, although it would have been interesting to see whether DBS could alleviate tremor in A1-knockout mice, these mice were prone to seizures, preventing the adequate testing of HFS at therapeutic levels.

As with any good experiment, unanswered questions remain. Specifically, how does adenosine activation of thalamic A1 receptors result in decreased tremor without apparently affecting normal movement? Is the neural activity that produces tremor at a certain frequency that is more susceptible to adenosine-derived inhibition? Or is there some more global influence on the metrics of movement—of which tremor is only the most obvious example? Answers to such questions will help refine our attempts at therapy.

Of course, as the authors themselves note, the therapeutic efficacy of DBS probably depends on multiple mechanisms (for instance, they show that the gamma-aminobutyric acid agonist muscimol also reduces tremor in harmaline-treated mice2). These mechanisms are likely to include the accumulation of neuroactive substances, such as adenosine, as well as circuit-level effects, such as, perhaps, the activation of inhibitory pathways to dampen oscillatory neuronal activity. To further complicate the matter, DBS using uniform electrical parameters may act via distinct mechanisms when applied to different parts of the brain and may result in long-term plastic changes that are difficult to study acutely.

Within the next several years, experiments such as those of Nedergaard and colleagues2 and the incorporation of new techniques will undoubtedly further elucidate those mechanisms. For example, the substitution of optical for electrical stimulation may allow the precise targeting of specific cell types, rather than generic geometric brain volumes, to more carefully control the target of DBS (Aravanis, A.M., Meltzer, L.A., Zhang, F., Mogri, M.Z., Wang, L.P. et al., personal communication). Moreover, there is increasing interest in developing closed-loop stimulators that are responsive to electrical or neurochemical changes. These approaches may provide not only experimental utility, but also significant therapeutic benefit. Ultimately, the expansion of our toolbox for repairing the brain will depend on such advances.

another article:

http://www.nature.com/nm/journal/v14...ll/nm1693.html