A rotarod test for evaluation of motor skill learning

Abstract

The rotarod test is widely used to evaluate the motor coordination of rodents, and is especially sensitive in detecting cerebellar dysfunction. However, mice with striatal dopamine depletion show only mild or no motor deficit on the typical accelerating rotarod. This suggests that dopamine-depleted mice are useful as animal models for non-motor symptoms, because the influence of motor deficit is minimum and easy to discriminate from cognitive aspects of the behavioral change. The typical accelerating rotarod test is designed to evaluate maximal motor performance and is not optimized to detect motor skill learning. In an attempt to make the test more selective to motor skill learning rather than maximal gait performance, we modified the rotarod test by using a slowly rotating large drum to obtain a steep learning curve. Furthermore, administration of nomifensine, a dopamine uptake inhibitor, improved the learning. On the other hand, apomorphine, an agonist of dopamine autoreceptor, a dopaminergic toxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) impaired the learning. These pharmacological profiles fit the involvement of the so-called phasic dopamine neurotransmission. Using our modified procedure, we found impaired learning of Parkin-deficit mice, which has not been detected in typical accelerating rotarod. The modified rotarod test would be useful for evaluation of dopamine involvement in the acquisition of motor skill learning.

Introduction

Parkinson's disease (PD) is a neurodegenerative disease characterized by akinesia, rigidity, resting tremor, and postural instabilities. In addition, neuropsychiatric, perceptual and cognitive deficits are increasingly recognized as non-motor manifestations of Parkinson's disease (Carbon and Eidelberg, 2006, Frank et al., 2004, Owen et al., 1992, Taylor et al., 1990). It is generally difficult to discriminate motor and cognitive aspects in behavioral tests of Parkinson's disease patients, because impaired movement can influence all of behavioral performance. If there is an animal model which has Parkinson-like pathological brain degeneration but has no motor deficit, it would be an ideal model for the non-motor symptoms of Parkinson's disease.

The major neurochemical hallmark of Parkinson's disease is the degeneration of dopaminergic neurons in the substantia nigra pars compacta. In animal models, nigral degeneration can be produced by the selective toxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydroxypyridine (MPTP). Although nigral neurons are degenerated in MPTP-treated mice, several studies have failed to detect motor deficit (Jackson-Lewis and Przedborski, 2007, Przedborski et al., 2001). While MPTP-treated primates show Parkinsonism-like motor deficits (Arai et al., 1990, Fisher et al., 2004, Sedelis et al., 2001, Tillerson et al., 2002), the behavior of MPTP-treated mice can be hardly be mentioned as “Parkinsonism”.

Among several behavioral tests that measure motor performance, the rotarod is a suitable test for evaluation of cerebellar deficits in rodents (Caston et al., 1995, Lalonde et al., 1995). The motor performance on the rotarod can be influenced by several factors, such as motor coordination, learning and cardiopulmonary endurance. Since several studies have shown that basal ganglia are essential in motor skill learning of serial motor sequence (Hikosaka et al., 1999), we tried to extract the acquisition of motor skill from the original procedure. In the previous study, deficit in the acquisition of rotarod learning was not obvious in MPTP-treated C57 BL/6 mice (Sedelis et al., 2000). Some studies have detected impairment of rotarod performance in dopamine-depleted mice and rats (Monville et al., 2006, Rozas et al., 1998). Monville have shown that the sensitivity to motor disability improved as rotation speed became higher. But these studies focused on behavior after overtraining whereas the learning curve at the initial acquisition phase was not presented. The learning effect appears as the elongated falling latency along with the trial numbers, but the typical accelerating rotarod does not seem ideal for evaluating acquisition of motor skill learning, because the learning curve is shallow and the performance after training for several days is no more than twice of the first day (Perez and Palmiter, 2005). Various rod sizes and speeds have been tried on a considerable amount of the literature evaluating daily changes in the motor performance on the rotarod (Akita et al., 2006, Caston et al., 1995, Jeljeli et al., 1999, Ogura et al., 2005, Sedelis et al., 2000). But if researchers decide to look for changes in performance over successive testing bouts where “motor learning” can be demonstrated, the rotarod test should have a distinction separating “motor learning” from basal gait/postural ability. Therefore, in the present study, we designed a modified rotarod protocol to evaluate the acquisition of motor skill learning selectively. Our non-accelerating rotarod test employs a wide drum with a hard surface, on which naive mice find it difficult to stay, but its low rotating speed leaves room for improvement after training.

A recent study reported dynamic reorganization of striatal circuits during the acquisition of motor skill on the accelerating rotarod (Yin et al., 2009). Furthermore, Akita et al. (2006) reported that a decrease in synaptic dopamine release induced by blocking the expression of synaptophysin in the nigrostriatal neurons resulted in impairment of acquisition of the rotarod task. Transgenic mice with striatal degeneration could walk on the rotarod but lacked the ability to learn (Kishioka et al., 2009). Based on these findings, one can postulate that the rotarod task reflects the striatum-based motor skill learning. To test this hypothesis, first, we tried our modified rotarod test in MPTP-treated mice. Next, we examined the effects of dopamine uptake inhibitors and a dopamine agonist. Finally, we examined our modified rotarod test in Parkin-deficient mice to compare with past studies using accelerating rotarod. Several studies have failed to detect impairment in gait performance of Parkin-deficient mice (Goldberg et al., 2003, Von Coelln et al., 2004, Perez and Palmiter, 2005, Sato et al., 2006), but impairment of motor skill learning was evident in our modified rotarod test. The results showed that our modified rotarod test is suitable for the selective evaluation of acquisition of motor skill, and cognitive involvement of the nigrostriatal dopamine system.