A study of cerebellar cortical involvement in motor learning using a new avoidance conditioning paradigm involving limb movement

A study of cerebellar cortical involvement in motor learning using a new avoidance conditioning paradigm involving limb movement

Brain Research, 445 (1988) 171-174 171 Elsevier BRE 22806 A study of cerebellar cortical involvement in motor learning using a new avoidance condit...

409KB Sizes 0 Downloads 27 Views

Brain Research, 445 (1988) 171-174

171

Elsevier BRE 22806

A study of cerebellar cortical involvement in motor learning using a new avoidance conditioning paradigm involving limb movement Jau-Shin Lou and James R. Bloedel Division of Neurob iology , Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, A Z 85013 (U. S.A. ) and Department of Physiology, University of Arizona, Tucson, A Z 85724 (U.S.A.)

(Accepted 8 December 1987) Key words: Cerebellar cortex; Climbing fiber; Motor learning

These experiments were performed to examine the relationship between the simple and complex spike responses of 3-5 simultaneously recorded Purkinje cells during the acquisition, performance and extinction of a conditioned forelimb movement in decerebrate, unanesthetized ferrets. The data demonstrate parallel, correlated changes in simple and complex spike responses throughout the experimental period. Since the evaluated Purkinje cells were examined in the cerebeilar cortical region that contains neurons highly modulated by the intermittent application of the conditioning stimulus, these findings argue against an induction of a long-lasting modification in simple spike responses by the climbing fiber input as the basis for this type of motor learning. Two general categories of hypotheses have been proposed regarding the function of the climbing fiber system. Those in one category suggest that the climbing fiber input produces a long-lasting, persistent modification in the strength of the parallel fiber synapse on Purkinje cell dendrites l'6's'9a6'2°. This argument is one of the bases for considering the cerebellar cortex as a site of the plastic changes required for some types of motor learning 17'2°'2t. Other hypotheses suggest that the climbing fiber input produces a short-lasting, non-persistent effect on Purkinje cells2atas. Our particular view is that these afferents produce a short-term modification in the responsiveness of Purkinje cells to mossy fiber inputs, a view proposed as the gain change hypothesis 2-5. Our most recent studies 14 were verformed to examine the issue of climbing fiber function in a more behavioral preparation. These experiments employed ambulating decerebrate ferrets to examine the activity of up to 6 simultaneously recorded, sagittally oriented Purkinje cells in response to intermittent limb perturbations. For this purpose a new analytical

technique was developed, the real-time postsynaptic response (RTPR). This technique permits a trial-bytrial assessment of the action of the simultaneously recorded neurons on a simulated cerebellar nuclear cell 12"13.The data from these studies indicate that the responsiveness of Purkinje cells located in the paravermal region of lobules V and VI is modulated most dramatically when the step cycle is perturbed, and that this reponsiveness is related to the degree of synchrony among the activated climbing fiber inputs to the cells of that set. In addition to supporting the gain change hypothesis, the results support the general argument that the action of these afferents is short lasting 2a2. In the experiments reported here the perturbed locomotion paradigm l°aaa9 was used as the basis for developing a new conditioning paradigm suitable for evaluating the role of the cerebellar cortex in a type of motor learning ~4'15. Specifically this study examines the relationship between the simple spike and complex spike responses of up to 6 simultaneously recorded Purkinje cells oriented in parasagittal strips

* Present address: Internal Medicine Residency Office, Baylor College of Medicine, Fondren-Brown Building, 6535 Fannin (B501), Houston, TX 77030, U.S.A. Correspondence: J.R. BIoedel, Division of Neurobiology, Barrow Neurological Institute, 350 West Thomas Road. Phoenix, AZ 85013, U.S.A. 0006-8993/88/$03.50© 1988 Elsevier Science Publishers B.V. (Biomedical Division)

172 during the acquisition and extinction of a conditioned limb movement performed during locomotion. It will be shown that this relationship is not modified during the conditioning process in a way suggested by some learning theories of climbing fiber function t'6'8'9'16'2°. No evidence was found for an induced change in simple spike responses that persisted after the complex spike responses returned to control values. Adult ferrets were anesthetized with halothane, a tracheostomy was performed, and each animal was artificially respired. A glue cap was attached to the frontal bone after the animal was placed in the stereotaxic frame. The animal was decerebrated at the rostral level of the superior colliculi using a radiofrequency generator. The anterior lobe of the cerebellum was exposed under 38.5 °C saline drip and covered with agar. The anesthesia an~l artificial respirator were discontinued, and the mfimal was suspended painlessly over a treadmill usiL,g the glue cap to secure the head and a cloth sling under the abdomen. Usually the animal was capable of walking on the treadmill within 1 h after the halothane was discontinued. The right front leg was attached to the lever arm of a potentiometer to measure limb displacement. The output of the potentiometer was fed through a threshold detector to trigger the perturbations. The perturbation of the right forelimb was produced by a bar that could be extended into the trajectory of the limb at specific phases of the step cycle. The position of the right forelimb, the limb receiving the perturbation, was monitored using a WatscopeWatsmart system. Infrared-emitting diodes required by this hardware were placed over the shoulder, elbow and wrist joints. The behavior also was monitored using a standard video camera. The perturbation was applied at the same phase of the locomotor cycle during successive steps over 1- to 3-min periods, depending upon the time course of the acquisition of the conditioned forelimb movement. A fixed array of 5 electrodes was manipulated until the activity of 3-5 Purkinje cells oriented in parasagittal strips could be recorded simultaneously 12.13. Purkinje cells were identified based on the presence of complex spikes 7. All the unitary and analog data were stored on tape for off-line data analysis. The cells were localized in regions of lobules V and VI containing the Purkinje neurons most responsive to intermittently applied perturbations in our previous studies ~3.

The simple and complex spikes of each Purkinje cell were discriminated separately off-line. The simple spike responses of the simultaneously recorded Purkinje cells were analyzed quantitatively on a sweep-by-sweep basis using the RTPR technique, a method previously detailed by Lou and Bloedel la and Lou 12. In brief, the RTPR is calculated from an algorithm based on the assumptions that the recorded cells converge on the same nuclear neuron and each evokes an IPSP of the same amplitude. To quantify the responses evoked by the perturbation, the response onset and termination were selected by visual inspection of the RTPR, and the response was integrated over the time window. The complex spike responses were analyzed quantitatively by calculating the synchrony index (SI), defined as the number of complex spikes/cell/trial 12'13. It should be emphasized that this data analysis technique makes it possible to quantify the synchrony among the climbing fiber inputs as well as the simple spike modulation occurring at each step cycle during the acquisition, performance, and extinction of the conditioned behavior. Responses in the perturbed and unperturbed trials were compared over the same period of the step cycle. The conditioned behavior itself was assessed by measuring the height of the wrist relative to the treadmill during each step cycle 15. The data were analyzed initially by comparing the plots of the SI, RTPR amplitude, and maximal vertical displacement of the wrist for each step cycle during control and conditioning periods (Fig. 1). The plots illustrate the typical variability in these measurements as well as their relationship to the time course of the conditioning and extinction of the elevated swing phase. In this experiment the animal began to clear the bar on the third perturbation trial. Once the conditioned behavior was acquired, the feotpads contacted the bar only intermittently during the continued performance of the conditioned movement 15. When the bar was no longer interjected into the trajectory of the limb (P-), the behavior extinguished rapidly. The force of the limb's contact with the bar was not assessed in these experiments. These plots reveal no observable trend in the plots of SI and RTPR amplitude relative to the conditioning or extinction period. This finding was typical across all of the sets. Next the quantitative relationship between the

173 ~3orA

~ =

...

, 8



, 16

..I 24

278

I 290

284

Steps

I

I 296

P-

Fig. 1. The relationships of step height, synchrony index (SI) and RTPR amplitude of each step cycle during an experimental period for a set of four Purkinje cells. P+ and P- along the X-axis indicate the time the perturbation was started and stopped, respectively. The RTPR amplitude is plotted in relative units. Some data are omitted between steps 28 and 274 in order to optimize the plots during acquisition and extinction. The plot of step height (lowest trace) shows that the animal progressively increased its step height to clear the bar after the perturbation trials started and that when the perturbation trials ended, the conditioned behavior extinguished gradually.

synchrony index and the R T P R response amplitude during the acquisition, performance, and extinction of the conditioned behavior was analyzed statistically in two ways. First, the average synchrony index and average RTPR amplitude during the entire conditioning and control periods for each set were compared. There was no statistically significant difference between these measurements in any of the 9 sets. Second, the average SI and the average R T P R response amplitude were compared in 10-step blocks before and after the initiation of the perturbation for each set of cells. Fig. 2 shows these data for one set of 5 Purkinje cells. Typically there was an increase of both the synchrony index and the RTPR amplitude during the first 10 step cycles after the initiation of the perturbation trials. Furthermore, these two measurements remained correlated over the duration of the conditioning period, even after the conditioned behavior was acquired. In fact, in all 9 sets of cells there was a statistically significant correlation between the SI and the R T P R amplitude when analyzed in this way across all step cycles during the conditioning period (P < 0.0005 in 7 sets and P < 0.005 in the other 2 sets). Neither these data nor the plots similar to those in Fig. 1 revealed any evidence that the climbing fiber input acts as a 'teacher', inducing a change in simple spike responses which persists after the climbing fiber responses decrease. The findings

in both Figs. I and 2 indicate also that the modulation of the simple spike activity during execution of the conditioned behavior is not markedly different than that observed during control locomotion. In summary, the results of this study show that the relationship between the synchronous activation of climbing fiber inputs to Purkinje cells and their simple spike responses did not change during conditioning as predicted by some learning theories of climb-

1 TRIAL

• sI [ ] RTPR

~J) 0.4

~

"13

o . 1 o Ai-iO 1l:2b2~?ai~ before l On

after

Step Number

Fig. 2. The relationship between the synchrony index and RTPR amplitude in 10-step blocks before and after the onset of the perturbation trials. Notice that the changes in RTPR and SI are correlated, indicating that the simple spike response did not undergo a change which persisted after the complex spike responses, as measured using the SI, returned to near control values.

174 ing fiber function 1''~'9"16.This finding also provides indirect support for the view that the climbing fiber input has short-term, non-persistent effects on the responsiveness of Purkinje cells to mossy fiber inputs. These data do not rule out the participation of the climbing fiber system in other types of m o t o r learning 17'2°'21 or the contribution of o t h e r cerebellar regions to this process. However, the study does pro-

vide an assessment of some of the suppositions of the learning theories in a cerebellar region k n o w n to receive climbing fiber afferents activated by the stimulus t3. Consequently, general views proposing the cerebellum as a site of the plastic changes in m o t o r learning must take these findings into account.

1 Albus, J.S., A theory of cerebellar function, Math. Biosci., 10 (1971) 25-61. 2 Bloedel, J.R. and Ebner, T.J., Climbing fiber function: Regulation of Purkinje cell responsiveness. In J.R. Bloedel, J. Dichgans and W. Precht (Eds.), Cerebellar Functions, Springer, Berlin, 1985, pp. 247-251. 3 Bloedel, J.R., Ebner, T.J. and Yu, Q.-X., Increased responsiveness of Purkinje cells associated with climbing fiber inputs to neighboring neurons, J. Neurophysiol., 50 (1983) 220-239. 4 Ebner, T.J. and Bloedel, J.R., Role of climbing fiber afferent input in determining responsiveness of Purkinje cells to mossy fiber inputs, J. Neurophysiol., 45 (1981) 962-971. 5 Ebner, T.J., Yu, Q.-X. and Bloedel, J.R., Increase in Purkinje cell gain associated with naturally activated climbing fiber inputs, J. Neurophysiol., 50 (1983) 205-219. 6 Gilbert, P.F.C., A theory of memory that explains the function and structure of the cerebellum, Brain Research, 70 (1974) 1-18. 7 Granit, R. and Phillips, C.G., Excitatory and inhibitory processes acti~g upon individual Purkinje cells of the cerebellum in the cats, J. Physiol. (Lond.), 133 (1956) 520-547. 8 Ito, M., Cerebellar control of the vestibulo-ocular reflex: Around the flocculus hypothesis, Annu. Rev. Neurosci., 5 (1982) 275-296. 9 Ito, M., The Cerebellum and Neural Control, Raven, New York, 1984. 10 Kim, J.H., Wang, J.-J. and Ebner, T.J., Climbing fiber afferent modulation during treadmill locomotion in the cat, J. Neurophysiol., 57 (1987) 187-202. 11 Llinas, R., Functional significance of the basic cerebellar circuit in motor coordination. In J.R. Bloedel, J. Dichgans and W. Precht (Eds.), Cerebellar Functions, Springer, Berlin, 1985, pp. 247-259. 12 Lou, J.-S., The Responses of Simultaneously Recorded Pur-

kinje Cells to the Perturbations of the Step Cycle in the Walking Ferret: A Study Using a New Analytical Method -- the Real-Time Postsynaptic Response (RTPR), Ph.D. Thesis, University of Minnesota, 1986. 13 Lou, J.-S. and Bloedel, J.R., The responses of simultaneously recorded Purkinje cells to the perturbations of the step cycle in the walking ferret: A study using a new analytical method m the real-time postsynaptic response (RTPR), Brain Research, 365 (1986) 340-344. 14 Lou, J.-S. and Bloedel, J.R., A study of cerebellar cortical involvement in motor learning using a new avoidance conditioning paradigm involving limb movement, Soc. Neurosci. Abstr., 12 (1986) 579. 15 Lou, J.-S. and Bloedel, J.R., A new conditioning paradigm: conditioned limb movement in locomoting decerebrate ferrets, Neurosci. Lett., 84 (1987) 185-190. 16 Marr, E., A theory of cerebellar cortex, J. Physiol. (Lond.), 202 (1969) 437-470. 17 McCormick, D.A. and Thompson, R.F., Neuronal responses of the rabbit cerebellum during acquisition and performance of a classically conditioned nictitating membrane eyelid response, J. Neurosci., 11 (1984) 1811-1822. 18 Pellionisz, A.J., Tensorial brain theory in cerebellar modelling. In J.R. Bloedel, J. Dichgans and W. Precht (Eds.), Cerebellar Functions, Springer, Berlin, 1985, pp. 201-229. 19 Schwartz, A.B., Ebner, T.J. and Bloedel, J.R., Comparison of responses in dentate and interposed nuclei to perturbations of the locomotor cycle, Exp. Brain Res., 67 (1987) 323-328. 20 Thompson, R.F., The neurobiology of learning and memory, Science, 233 (1986) 941-947. 21 Yeo, C.H., Hardiman, M.J. and Glic,kstein, M., Classical conditioning of the nictitating membrane response of the rabbit. II. Lesions of the cerebellar cortex, Exp. Brain Res., 60 (1985) 99-113.

This w o r k was supported by N I H G r a n t NS21958.