Deep brain stimulation of the subthalamic nucleus (STN) is an effective treatment option for Parkinson's disease1 and other movement disorders. The underlying mechanisms are still poorly understood and multifactorial, but a key feature is the modulation of neuronal network activity by repetitive depolarization of axons in the vicinity of the stimulating electrode2-4. High-frequency (>100 Hz) stimulation is required for a beneficial effect in most brain targets and for most indications of DBS. Side effects of deep brain stimulation result from inadvertent coactivation of other fibers, which are covered by the stimulation volume and which subserve different functions, such as the pyramidal tract. Hence, it would be desirable to develop stimulation parameters, which preferentially activate beneficial neural elements, while avoiding the coactivation of side effect elements5,6. Although neurophysiology may offer such fine tuning options of DBS, the scientific progress has been minimal during the past two decades, because programming strategies have primarily been assessed by "trial and error" in patients and restricted by the limited programming options of commercially available DBS devices, rather than using neurophysiological insight and defined experimental settings to systematically explore the full parameter space.
To overcome the translational roadblock in DBS research we are proposing a protocol to screen alternative stimulation parameters in rodent models of Parkinsonism prior to clinical exploration. Unilateral Parkinson's disease in rats is modeled using 6-hydroxydopamine injections into the right medial forebrain bundle7,8. The resulting lesion, described further as hemiparkinsonian, is assessed in the apomorphine test by evaluation of rotation score after low dose apomorphine injection and confirmed post mortem by tyrosine hydroxylase immunohistochemistry. The method is easy to apply and highly reproducible, while bearing a low mortality and morbidity. The resulting motor deficits are very discrete7,8; the animals exhibit a slight impairment of the contralateral left paw during both spontaneous exploration and complex grasping behavior9,10.
To assess the effectiveness of deep brain stimulation protocols tests are required which allow measuring a quick and reliable change in motor performance and can be repeated over time with different neurostimulation settings. Several groups have proposed different stimulation approaches and different tests to assess the motor functions in rats11 with highly variable and inconsistent outcomes11-14. This forced us to choose a set of tests with high predict validity and complementarity. Additionally, for assessment of motor outcome under deep brain stimulation conditions, tests were favored which could be performed by animals connected via cable to the stimulus generator. For these purposes we established our test battery consisting of one test for paw use asymmetry and one test for skilled reaching. The study design is illustrated in Figure 1.
For spontaneous paw use we performed the cylinder test described by Shallert15, which is a widely used test for paw use during vertical exploration. No training of the animal is required. For the assessment of more complex grasping behavior we established the staircase test according to Montoya16. Our protocol is modified according to Kloth17. The rats are trained for a period of twelve days in reaching pellets from the test box. After the training period the test can be applied to measure the complex grasping behavior by counting the success rate described as number of pellets eaten. The article presents the detailed training in the staircase box as well as the performance of both behavioral tests under naïve, hemiparkinsonian and deep brain stimulation conditions.