- Email: [email protected]
Sensorimotor learning and retention during equilibrium tests in Purkinje cell degeneration mutant mice
Brain Research 768 Ž1997. 310–316
Research report
Sensorimotor learning and retention during equilibrium tests in Purkinje cell degeneration mutant mice N. Le Marec a
a,b
, R. Lalonde
a,b,)
Neurology SerÕice, Unit of BehaÕioral Neurology, Neurobiology, and Neuropsychology, Hotel-Dieu Hospital Research Center, Montreal, Canada ˆ b Depts ´ de Medecine et Psychologie et Centre de Recherche en Sciences Neurologiques, UniÕersite´ de Montreal, ´ Montreal, Canada Accepted 13 May 1997
Abstract In order to determine the consequences of atrophy to the cerebellar cortex on postural sensorimotor learning and performance, a natural mutation, Purkinje cell degeneration Žpcd., was used. The homozygous mutants were compared to heterozygous non-ataxic controls on three static beams, two grids Žvertical and tilted., a mobile beam Žaccelerating rotorod., and a coat-hanger. Although their posture was less stable than that of controls, the pcd mutants were not impaired in distance travelled or in latencies before falling on the static beams. Mutant performance on the grids was not impaired in comparison to controls, while a reduction of latencies before falling on the coat-hanger occurred only during the early part of training. On the accelerating rotorod, pcd mutants fell far sooner than controls and spent more time in passive rotation. By contrast to controls, pcd mutants were not able to improve with practice. Both mutants and controls were deficient during a retention test conducted 8 days after acquisition. The cerebellar cortex is critically involved in timing whole body movements during postural adjustments to a mobile beam but not to four types of immobile apparatus. q 1997 Elsevier Science B.V. Keywords: Cerebellum; Purkinje cell; Mutant mouse; Sensorimotor learning; Balance; Equilibrium
1. Introduction The cerebellum has been suggested to play a role in sensorimotor learning and memory w11–13,17,18,39x. Cerebellar modulation of classical conditioning of the eyeblink response has especially been well investigated w2,39,41,42x. The specific role played by deep cerebellar nuclei, particularly the interpositus, in this kind of conditioning has been emphasized w26,29–31,38,39,42x. In addition, the cerebellum is necessary for adaptation of the vestibulo-ocular reflex w10,18,37x and is also involved in complex sensorimotor learning requiring whole body movements w3,6,7,21–24,28,43x. The participation of the cerebellum in behavior has been extended to higher cognitive functions in humans w1,4,5,27,32x. In view of the well-known postural disorders after cerebellar damage, we wished to determine the specific
)
Corresponding author. Present address: Universite´ de Rouen, Faculte´ des Sciences, Laboratoire de Neurobiologie de l’Apprentissage, 76821 Mont-Saint-Aignan Cedex, France. Fax: q33 Ž2. 3514-6349; E-mail: [email protected] 0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 6 6 6 - 5
role of the cerebellar cortex in various equilibrium tests, more particularly in regard to sensorimotor learning and long-term retention. For this purpose, Purkinje cell degeneration Žpcd. mice were used. In this spontaneous mutation, there is a total loss during the third and fourth postnatal weeks of Purkinje cells, with relative sparing of cerebellar granule cells and deep nuclei, the latter degenerating much later, possibly due to anterograde transsynaptic degeneration secondary to the Purkinje cell loss w14,15,25,35,40x. There is also a retinal degeneration of slower onset w34,35x. Because Purkinje cells are the only output neuron from the cerebellar cortex w17x, the deep nuclei receive no input from the cerebellar cortex. However, two important afferent pathways from brainstem nuclei Žclimbing fibres and parallel fibres., conveying peripheral information to normal cerebellar nuclei by means of collateral fibres w8,33x, may be still functional in this mutation. It is already known that pcd mutant mice are severely impaired during acquisition of classical eyeblink conditioning w9x and spatial orientation w16x. The purpose of the present series of studies was to determine whether the same mutant is equally impaired in sensorimotor tests
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
requiring balance. A series of static beam and grid tests were conducted w20,21,23,24x, together with the rotorod test w6,7x and the coat-hanger test w23x. The coat-hanger test differs from standard beam tests in requiring greater muscle strength, since the animals are suspended upside down from a horizontally positioned string.
2. Materials and methods
311
body positions Žsee w23x for additional details. were measured, namely position 1 s the body of the mice is parallel to the beam with all four paws on it; position 2 s the body of the mice is perpendicular to the beam with all four paws on it; position 3 s the mice have only one front paw and one hind paw on the beam; position 4 s the mice have only two front paws on the beam. Positions 1 and 2 represent stable postures, whereas positions 3 and 4 represent unstable postures. The body position of the mice was recorded every 15 s.
2.1. Animals Female pcd mutant mice Ž pcdrpcd . and non-ataxic littermate controls Žqrpcd . of the B6C3Fe-ara strain, approximately 2 months old Žrange: 1.5–2.5 months. during the first day of testing, were purchased from Jackson Laboratory ŽBar Harbor, ME, USA. and housed in a temperature- and humidity-controlled room. Before testing, the animals were handled for a few days in order to facilitate adaptation. 2.2. Apparatus and procedures The order of presentation of the tests was as follows: large round beam, narrow round beam, rectangular beam, vertical grid, inclined grid, rotorod, and coat-hanger. The interval between the end of one test and the start of another was 2–4 days. 2.2.1. Wooden beams In the beam tests and the grid tests, 7 pcd mutants and 7 controls were used. There were two round static wooden beams covered with surgical tape. The first beam measured 8.5 cm in diameter and 80 cm in length and was divided into 8 equal segments. The second, of equal length, measured 4.6 cm in diameter. In order to prevent any escape, the endpoint of each beam was blocked by a piece of cardboard. The floor was covered by a layer of soft cloth in case of any fall. The mice were positioned in the middle part of the beams and allowed to remain on them for 60 min. There were 4 trials in a single day of testing with an intertrial interval of 10–15 min. The time spent on the beams and the number of segments crossed were measured. After these two round beam tests, a rectangular beam test was used, consisting of four interconnected wooden boards forming a rectangle Ž90 = 70 cm, width 1 cm, height from the floor 34 cm. and a fifth board inserted within the rectangle. The beam was separated into 10 cm segments, yielding a total of 39 segments Ž9 on two sides and 7 on the other three sides.. The mice were placed in the middle part of one of the boards. There were 4 trials per day, during 2 consecutive days, and an intertrial interval of 10–15 min, with a cut-off period of 60 s per trial. The time spent on the beam, the number of segments crossed, and the frequencies of occurrence of 4 specific
2.2.2. Grids Two grids were used. The mice were first placed in the middle part of a vertically positioned steel grid Ž25 = 18 cm, spacing between bars: 1.2 cm.. The number of falls and the time needed before reaching the top of the grid were measured. There were 4 trials on a single testing day with a cut-off point of 60 s per trial and an intertrial interval of 10–15 min. The second grid consisted of a wiremesh Ž38 = 38 cm, 5 squares per cm. surrounded by a wooden frame and an inclination of 308 or 608 from the horizontal. This test was conducted on the same day as the previous test. The mice were placed in the middle part of the grid facing downward. There were 4 trials Ž2 at 308; 2 at 608 in alternating order, starting at 308. on a single testing day with a cut-off point of 60 s and an intertrial interval of 10–15 min. The time needed before turning upward Žusing the whole body of the animal as the criterion, i.e. a 908 head-turn was not considered to be sufficient., the time needed before reaching the top of the grid Žsnout criterion., and the number of falls were measured. Animals successful in reaching the top were immediately removed and the trial ended. 2.2.3. Accelerating rotorod The accelerating rotorod Žmodel 7650, Stoelting, Wood Dale, IL, USA. consisted of a horizontal mast Ž3 cm in diameter., turning around its longitudinal axis, each divided into 5 sections of 5.5 cm. The mast permitted the mice to locomote without slipping because of the knurled material. Each section was separated by a piece of plastic in order to prevent mice from jumping to the adjacent section. The speed of the rotation increased gradually Žwithout any jerk. at each 30 s interval by 4 revolutions per minute Žrpm. from 4 rpm to 40 rpm Žcut-off point: 5 min per trial.. The mice were gently dropped while the apparatus was in motion. After 3–5 s, the beam rotated on the opposite side of where the mice were facing, promoting forward locomotion in order to avoid a fall. There were 4 trials per day during 10 consecutive days, with an intertrial interval of 15–20 min. On some trials, the mice clung to the rod without moving and were thus passively rotated. The time spent on the rotorod, the latency before onset of passive rotation, and the time spent in passive rotation were measured. Passive rotation was tabulated only on those trials where the mice used this strategy until
312
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
the end of the trial. Eight days after the end of the training session, a retention test was given under identical conditions. Two animals from the control group were eliminated from the study because of a tendency to jump deliberately from the mast and one suffered an injury to its front paw during the training session, leaving n s 4. To replace those mice, 4 new mice of the same age were added. None of the mice from the pcd group was observed to jump deliberately while the mast was in motion. 2.2.4. Coat-hanger During the coat-hanger test, a shipment of 4 pcd mutants arrived and were thereby added to the group, yielding a total of n s 11, in comparison to the 6 controls that had been exposed to every previous test. The triangular shaped coat-hanger consisted of a horizontally positioned string measuring 40 cm in length and 2 mm in diameter flanked by two side-bars of equal diameter measuring 19 cm in length Ž358 angle.. The coat-hanger was placed 78 cm above a cushioned table. The mice were placed upside down in the middle part of the horizontal string, care being taken to release the mice when all four paws clung to it. Latencies before falling and the movement time ŽMT. necessary to reach with two front paws one of the diagonal bars were measured. There were 4 trials per day during 10
consecutive days and the intertrial interval was 15–20 min. Eight days after the end of the training session, a retention test was given under identical conditions. 2.3. Statistical analyses The unpaired t-test was used to compare pcd and control performances on the beams and grids, except for the occurrence of body positioning in the rectangular beam test, where the Mann-Whitney U-test was used because of unequal variances. A 2 = 10 ANOVA with repeated measures on the second factor was used on the rotorod and on the coat-hanger tests, with 2 independent groups Žpcd vs. controls. and 10 days of testing, together with one-way ANOVAs as needed to evaluate the performance of each group. The paired t-test was used to compare performances between day 10 of acquisition and the retention test.
3. Results 3.1. Static beams There was no difference Ž P ) 0.05. between pcd mutant mice and controls on any measure for the round wide
Fig. 1. Means and S.E.M. of pcd mutant mice Ž n s 7. and controls Ž n s 8. for 10 days of training and one day of retention Žday 11. on the rotorod. A: time spent on the rod; B: time spent before onset of passive rotation; C: time spent in passive rotation; D: percentage of trials in passive rotation.
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316 Table 1 Mean ŽS.E.M.. values of pcd mutants and controls in three beam tests of motor coordination Measures Round wide beam Time on beam Žs. Segments Round narrow beam Time on beam Žs. Segments Rectangular beam Time on beam Segments Position 1 Position 2 Position 3 Position 4 a
313
the inclined grid. The pcd mutants turned upward more rapidly Ž P - 0.01. and reached the top of the grid more quickly Ž P - 0.01. than did controls at both inclinations ŽTable 2..
pcd mutants
Controls
240 Ž0. 29.3 Ž6.3.
240 Ž0. 28.4 Ž5.8.
3.3. Rotorod
240 Ž0. 28.6 Ž6.2.
240 Ž0. 21.1 Ž2.7.
480 Ž0. 55.6 Ž12.4. 14.0 Ž4.0. 0.4 Ž0.2. 7.6 Ž2.1. 10.0 Ž3.8.
480 Ž0. 70.6 Ž25.9. 23.0 Ž3.4. 8.1 Ž2.8. a 0 Ž0. a 0.9 Ž0.6. a
A 2 = 10 ANOVA revealed significant group Ž P 0.001. and day Ž P - 0.001. effects but no significant interaction Ž P ) 0.05. for latencies before falling and latencies before the onset of passive rotation. The pcd mutants had lower values than controls on both measures ŽFig. 1A,B.. While the control mice improved their performances on these measures Ž P - 0.001., the slight increase in pcd mutants did not reach significance Ž P ) 0.05.. A 2 = 10 ANOVA on time spent in passive rotation and the percentage of trials with passive rotation revealed a significant interaction term Ž P - 0.05.. The pcd mutants spent more time in passive rotation and had a higher number of trials with passive rotation later in training, whereas the opposite tendency was found in controls ŽFig. 1C,D.. The pcd mutants and controls had lower values Ž P 0.05. for time spent on the mast during the retention test in comparison to day 10 of acquisition. The time spent before
P - 0.05 vs. pcd mutants.
beam or the round narrow beam ŽTable 1.. Because of the lack of difference between performances on day 1 and day 2 on the rectangular beam, the results of the 2 days were combined. There was no difference Ž P ) 0.05. between pcd mutants and control mice for the time spent on the beam and for the number of segments crossed. By contrast, the positions adopted by pcd mutants and controls to maintain their equilibrium on the beam were different ŽTable 1.. The pcd mice were found more frequently than controls in two unstable positions, namely 3 and 4 Ž P 0.05 and P - 0.01, respectively. and less frequently in one stable position, namely position 2 Ž P - 0.01.. The effect on position 1 did not reach significance Ž P ) 0.05.. 3.2. Grids As only one mutant fell from the vertical grid, there was no group difference Ž P ) 0.05. in latencies before falling ŽTable 2.. Paradoxically, the time spent before reaching the top part of the grid was shorter Ž P - 0.01. in pcd mutants than in controls. None of the mice fell from Table 2 Mean ŽS.E.M.. values of pcd mutants and controls on the vertical grid and on the inclined grid Measures Vertical grid Falls Top latencies Inclined grid 308 Falls Turning time Žs. Top latencies Žs. 608 Falls Turning time Žs. Top latencies Žs. a
P - 0.01 vs. pcd mutants.
pcd mutants
Controls
0.1 Ž0.1. 45.0 Ž14.8.
0 Ž0. 131.1 Ž23.3. a
0 Ž0. 7.9 Ž2.4. 17.1 Ž3.4.
0 Ž0. 21.0 Ž4.3. a 40.9 Ž7.1. a
0 Ž0. 7.0 Ž2.0. 18.6 Ž2.6.
0 Ž0. 33.4 Ž9.4. a 73.0 Ž12.1. a
Fig. 2. Means and S.E.M. of pcd Ž ns11. and controls Ž ns6. for 10 days of training and one day of retention Žday 11. on the coat-hanger. A: latencies before reaching the side-bar with two paws; B: latencies before falling.
314
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
onset of passive rotation was shorter for controls Ž P - 0.05. but not for the mutants Ž P ) 0.05.. Conversely, the time spent in passive rotation was lower in pcd mutants Ž P 0.05. but not in controls Ž P ) 0.05.. There was no difference for either group in the percentage of trials with passive rotation Ž P ) 0.05.. 3.4. Coat-hanger A two-way ANOVA revealed no significant group, day or interaction effect Ž P ) 0.05. for the MT measure ŽFig. 2A.. A two-way ANOVA revealed a significant interaction for latencies before falling Ž P - 0.05.. As shown in Fig. 2B, the latencies before falling increased as a function of time in pcd mutants Ž P - 0.001. but not in controls Ž P ) 0.05., as their values were nearly maximal throughout the testing period. There was no difference between the retention score and the score on the final day of acquisition for either group Ž P ) 0.05..
4. Discussion The pcd mutants in the present study were tested at a young age, at a time when the major neuropathological change is a loss of Purkinje cells w14,15,25,35,40x, although the functional consequences of retrograde and anterograde degeneration cannot be excluded. By contrast to other cerebellar mutant mice with more widespread cerebellar pathology w20,21x, pcd mutants were not impaired on static beam and grid tests. Irrespective of beam width or shape, pcd mutants did not fall sooner than controls. Moreover, in spite of ataxia, the distance travelled by pcd mutants on the three beams equalled that of controls. In a previous study w23x, pcd mutants had normal activity levels in a T-maze, an indication of the mildness of their locomotor deficits. The only measure where they differed from controls was the stability of their posture on the rectangular beam. This result is similar to that found in lurcher mutant mice, another cerebellar mutant with degeneration of the cerebellum and the inferior olive w24x. On the two grid tests, the performance of the pcd mutants was not deficient either. Paradoxically, the time taken before reaching the top of the grids was shorter in pcd mutants than in controls. Moreover, the time taken before turning upward on the inclined grid was shorter in the mutant group at both inclinations. The latter result may be explained by the poorer stability of the mutants. Whereas control mice explored the bottom and the top parts of the grid in a similar manner, the mutants were probably able to attain a more stable body position by facing upward on the inclined slope. This interpretation is supported by the finding that lurcher mutants were also observed to turn upward more quickly than normal mice on the inclined grid w19x. The pcd mutants also reached the top of the grid more quickly than the control group. Because of the
instability of their body position, the mutants probably attempted to reach a horizontal position or escape from the apparatus altogether. However, by contrast to the results in pcd mutants, lurcher mutants fell sooner from the vertically positioned grid than normal mice, a result that may be due to their more severe cerebellar pathology w23x. By contrast to lurcher mutants w23x, pcd mutants were not deficient on the coat-hanger in terms of MT. The mutants fell from the bar sooner than controls, but this result only occurred early in training. Neither mutants nor controls were impaired after an 8-day retention interval for either measure. Only on the rotorod did large mutant–nonmutant differences emerge. Control mice were far superior to pcd mutants on latencies before falling from the rod and were slower before adopting the passive rotation strategy. Moreover, whereas the control group decreased the time spent in passive rotation across days, on the contrary this behavior was increased with time in pcd mutants. This result indicates that normal mice were able to improve their latencies before falling by learning a complex sensorimotor skill, walking forward in conjunction with increasing rotation speed. By contrast, pcd mutants were unable to improve significantly the time spent on the mast despite an increase in the time spent in passive rotation. This lack of improvement is in contrast to their increase in latencies before falling on the coat-hanger. These results indicate improved motor function in a test where postural adjustments can be self-initiated but no improvement in a test where postural adjustments must be made in response to an external stimulus. Anderson and Steinmetz w2x and Raymond et al. w36x have suggested that the cerebellar cortex plays a crucial role in the timing of learned movements. This hypothesis is supported by the results of the present study. The pcd mutants had a severe deficit in synchronizing their movements in time to that of the rotating mast. Although no learning was demonstrated in the mutants within the confines of this study, it is possible that sensorimotor learning can occur with additional trials. We have found an improvement of latencies before falling in lurcher mutants under the same experimental conditions, an indication that even with massive cerebellar cortical pathology some learning can occur w28x. The poorer performance of pcd mutants is paradoxical because they have less extensive cerebellar pathology. The falling latencies of lurcher mutants increased significantly during the 10-day period Žmeans and S.E.M.. from 391 Ž44. to 749 Ž44. s, while those of pcd mutants were only slightly Žnon-significantly. higher, from 401 Ž34. to 615 Ž100.. This lack of a significant increase is in part due to the high intragroup variability. Thus, the initial performance of both groups was approximately the same. However, the terminal performance of lurcher mutants was higher and less variable. Another difference between the mutants is that whereas latencies before the onset of passive rotation increased
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
considerably in lurchers, from 347 Ž37. on day 1 to 632 Ž38. on day 10, those of pcds remained approximately the same, from 371 Ž28. to 462 Ž42.. Thus, the enhanced latencies before falling in lurchers are not due to passive clinging, whereas in pcds there was no significant increase in latencies before falling in spite of enhanced passive clinging. We conclude that a more widespread degeneration of cerebellar cell populations does not necessarily lead to poorer motor performance. Metabolic changes in cerebellar afferent or efferent structures in lurcher mutants may compensate for loss in cerebellar function. In the present study, the performance of control mice on the retention test was slightly lower than their performance on day 10 of acquisition in terms of latencies before falling and latencies before onset of passive rotation. Thus, some loss of memory of the acquired sensorimotor skills occurred in normal mice. The performance of pcd mutants also worsened after the retention interval in terms of time spent on the rod. By contrast to controls, pcd mice spent less time in passive rotation during the retention test. Thus, each group lost during the retention interval their preferred strategy. Control mice were less able to walk forward in synchrony with the moving beam, whereas pcd mice appeared to forget, at least in part, their previous tendency of clinging passively to the beam. Chen et al. w9x reported an impairment in pcd mutant mice during classical eyeblink conditioning. Our results show that selective atrophy of Purkinje cells impaired acquisition of a task requiring precise forward walking in response to predictable rotation speeds of a mast. Further testing of this mutant on other sensorimotor tasks will delineate the role of the cerebellar cortex on learning.
w8x w9x
w10x w11x
w12x
w13x w14x
w15x
w16x
w17x w18x
w19x w20x w21x
Acknowledgements
w22x
This research was funded by a grant from the Natural Sciences and Engineering Research Council of Canada.
w23x w24x
References w1x N.A. Akshoomoff, E. Courchesne, G.A. Press, V. Iraqui, Contributions of the cerebellum to neuropsychological functioning, Neuropsychologia 30 Ž1992. 315–338. w2x B.J. Anderson, J.E. Steinmetz, Cerebellar and brainstem circuits involved in classical eyeblink conditioning, Rev. Neurosci. 5 Ž1994. 251–273. w3x N. Auvray, J. Caston, A. Reber, T. Stelz, Role of the cerebellum in the ontogenesis of the equilibrium behavior in the young rat: a behavioral study, Brain Res. 505 Ž1989. 291–301. w4x T. Botez-Marquard, M.I. Botez, Cognitive behavior in heredodegerative ataxias, Eur. Neurol. 33 Ž1993. 351–357. w5x T. Botez-Marquard, I. Routhier, Reaction time and intelligence in patients with olivopontocerebellar atrophy, Neuropsychiatry Neuropsychol. Behav. Neurol. 8 Ž1995. 168–175. w6x J. Caston, N. Delhaye-Bouchaud, J. Mariani, Motor behavior of heterozygous staggerer mutant Žqr sg . versus normal Žqrq. mice during aging, Behav. Brain Res. 72 Ž1996. 97–102. w7x J. Caston, F. Vasseur, T. Stelz, C. Chianale, N. Delhaye-Bouchaud,
w25x
w26x
w27x w28x
w29x
w30x
315
J. Mariani, Differential role of cerebellar cortex and deep cerebellar nuclei in learning the equilibrium behavior: studies in intact and cerebellectomized lurcher mutant mice, Dev. Brain Res. 86 Ž1995. 311–316. V. Chan-Palay, Cerebellar Dentate Nucleus: Organization, Cytology and Transmitters, Springer, Berlin, 1977. L. Chen, S. Bao, J.M. Lockard, J.J. Kim, R.F. Thompson, Impaired classical eyeblink conditioning in cerebellar-lesioned and Purkinje cell degeneration Žpcd. mutant mice, J. Neurosci. 16 Ž1996. 2829– 2838. H. Collewijn, The modifiability of the adult vestibulo-ocular reflex, Trends Neurosci. 2 Ž1979. 98–102. M. Dahhaoui, J. Caston, N. Auvray, A. Reber, Role of the cerebellum in an avoidance conditioning task in the rat, Physiol. Behav. 47 Ž1990. 1175–1180. M. Dahhaoui, J. Caston, J. Lannou, S. Avenel, Role of the cerebellum in habituation exploration behavior in the rat, Physiol. Behav. 52 Ž1992. 339–344. R.S. Gelman, F.A. Miles, A new role for the cerebellum in conditioning, Trends Neurosci. 8 Ž1985. 181–182. B. Ghetti, C.J. Alyea, J. Muller, Studies on the Purkinje cell degeneration Žpcd. mutant mice: primary pathology and transneuronal changes, J. Neuropathol. Exp. Neurol. 37 Ž1978. 617. B. Ghetti, J. Norton, L.C. Triarhou, Nerve cell atrophy and loss in the inferior olivary complex of ‘Purkinje cell degeneration’ mutant mice, J. Comp. Neurol. 260 Ž1987. 409–422. R.C. Goodlett, K.M. Hamre, J.R. West, Dissociation of spatial navigation and visual guidance performance in Purkinje cell degeneration Žpcd. mutant mice, Behav. Brain Res. 47 Ž1992. 129–141. M. Ito, The Cerebellum and Neural Control, Raven, New York, 1984. M. Ito, Synaptic plasticity in the cerebellar cortex that may underlie the vestibulo-ocular adaptation, in: A. Berthoz, G.M. Melvill-Jones ŽEds.., Adaptative Mechanisms in Gaze Control: Facts and Theories, Elsevier, Amsterdam, 1985, pp. 213–221. R. Lalonde, Motor learning in lurcher mutant mice, Brain Res. 639 Ž1994. 351–353. R. Lalonde, Motor abnormalities in staggerer mutant mice, Exp. Brain Res. 68 Ž1987. 417–420. R. Lalonde, Motor abnormalities in weaver mutant mice, Exp. Brain Res. 65 Ž1987. 479–481. R. Lalonde, M. Manseau, M.I. Botez, Spontaneous alternation and exploration in Purkinje cell degeneration mutant mice, Brain Res. 411 Ž1987. 187–189. R. Lalonde, M.I. Botez, C.C. Joyal, M. Caumartin, Motor abnormalities in lurcher mutant mice, Physiol. Behav. 51 Ž1992. 523–525. R. Lalonde, C.C. Joyal, J.-M. Guastavino, M.I. Botez, Hole poking and motor coordination in lurcher mutant mice, Physiol. Behav. 54 Ž1993. 41–44. S.C. Landis, R.J. Mullen, The development and degeneration of Purkinje cells in pcd mutant mice, J. Comp. Neurol. 177 Ž1978. 125–144. D.G. Lavond, T. Hembree, R.F. Thompson, Effect of kainic acid lesions of the cerebellar interpositus nucleus on eyelid conditioning in the rabbit, Brain Res. 326 Ž1985. 179–182. H.C. Leiner, A.L. Leiner, R.S. Dow, Cognitive and language functions of human cerebellum, Trends Neurosci. 11 Ž1993. 444–447. N. Le Marec, J. Caston, R. Lalonde, Impaired motor skills on static and mobile beams in lurcher mutant mice, Exp. Brain Res. Ž1997. in press. D.A. McCormick, R.F. Thompson, Neuronal responses of the rabbit cerebellum during acquisition and performance of a classically conditioned nictitating membrane-eyelid response, J. Neurosci. 4 Ž1984. 2811–2822. D.A. McCormick, R.F. Thompson, Cerebellum: essential involvement in the classically conditioned eyelid response, Science 223 Ž1984. 296–299.
316
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
w31x D.A. McCormick, D.G. Lavond, G.A. Clark, R.E. Keetner, C.A. Rising, R.F. Thompson, The engram found? Role of the cerebellum in classical conditioning of nictating membrane and eyelid response, Bull. Psychon. Soc. 18 Ž1981. 103–105. w32x F.A. Middleton, P.L. Strick, Anatomical evidence for cerebellar and basal ganglia involvement in higher cognitive function, Science 266 Ž1994. 456–461. w33x G.A. Mihailoff, Identification of pontocerebellar axon collateral synaptic boutons in the rat cerebellar nuclei, Brain Res. 48 Ž1994. 313–318. w34x R.J. Mullen, M.M. LaVail, Two new types of retinal degeneration in cerebellar mutant mice, Nature 258 Ž1975. 528–530. w35x R.J. Mullen, E.M. Eicher, R.L. Sidman, Purkinje cell degeneration: a new neurological mutation in the mouse, Proc. Natl. Acad. Sci. USA 73 Ž1976. 208–212. w36x J.L. Raymond, S.G. Lisberger, M.D. Mauk, The cerebellum: a neuronal learning machine?, Science 272 Ž1996. 1126–1131. w37x D.A. Robinson, Adaptative gain control of vestibulo-ocular reflex by the cerebellum, J. Neurophysiol. 39 Ž1976. 954–969.
w38x J.E. Steinmetz, S.F. Logue, J.J. Steinmetz, Rabbit classically conditioned eyelid responses do not reappear after interpositus nucleus lesion and extensive post-lesion training, Behav. Brain Res. 51 Ž1992. 103–114. w39x R.F. Thompson, The neurobiology of learning and memory, Science 233 Ž1986. 941–947. w40x L.C. Triarhou, J. Norton, B. Ghetti, Anterograde transsynaptic degeneration in the deep cerebellar nuclei of Purkinje cell degeneration mutant mice, Exp. Brain Res. 66 Ž1987. 577–588. w41x C.H. Yeo, M.J. Hardiman, M. Glickstein, Discrete lesions of the cerebellar cortex abolish the classically conditioned nictating membrane response in the rabbit, Behav. Brain Res. 13 Ž1984. 261–266. w42x C.H. Yeo, M.J. Hardiman, M. Glickstein, Classical conditioning of the nictitating membrane response of the rabbit. II. Lesions of the cerebellar nuclei, Exp. Brain Res. 60 Ž1985. 87–98. w43x C. Zion, N. Auvray, J. Caston, A. Reber, T. Stelz, Effects of cerebellectomy at day 15 on the ontogenesis of the equilibrium behavior in the rat, Brain Res. 515 Ž1990. 104–110.
Research report
Sensorimotor learning and retention during equilibrium tests in Purkinje cell degeneration mutant mice N. Le Marec a
a,b
, R. Lalonde
a,b,)
Neurology SerÕice, Unit of BehaÕioral Neurology, Neurobiology, and Neuropsychology, Hotel-Dieu Hospital Research Center, Montreal, Canada ˆ b Depts ´ de Medecine et Psychologie et Centre de Recherche en Sciences Neurologiques, UniÕersite´ de Montreal, ´ Montreal, Canada Accepted 13 May 1997
Abstract In order to determine the consequences of atrophy to the cerebellar cortex on postural sensorimotor learning and performance, a natural mutation, Purkinje cell degeneration Žpcd., was used. The homozygous mutants were compared to heterozygous non-ataxic controls on three static beams, two grids Žvertical and tilted., a mobile beam Žaccelerating rotorod., and a coat-hanger. Although their posture was less stable than that of controls, the pcd mutants were not impaired in distance travelled or in latencies before falling on the static beams. Mutant performance on the grids was not impaired in comparison to controls, while a reduction of latencies before falling on the coat-hanger occurred only during the early part of training. On the accelerating rotorod, pcd mutants fell far sooner than controls and spent more time in passive rotation. By contrast to controls, pcd mutants were not able to improve with practice. Both mutants and controls were deficient during a retention test conducted 8 days after acquisition. The cerebellar cortex is critically involved in timing whole body movements during postural adjustments to a mobile beam but not to four types of immobile apparatus. q 1997 Elsevier Science B.V. Keywords: Cerebellum; Purkinje cell; Mutant mouse; Sensorimotor learning; Balance; Equilibrium
1. Introduction The cerebellum has been suggested to play a role in sensorimotor learning and memory w11–13,17,18,39x. Cerebellar modulation of classical conditioning of the eyeblink response has especially been well investigated w2,39,41,42x. The specific role played by deep cerebellar nuclei, particularly the interpositus, in this kind of conditioning has been emphasized w26,29–31,38,39,42x. In addition, the cerebellum is necessary for adaptation of the vestibulo-ocular reflex w10,18,37x and is also involved in complex sensorimotor learning requiring whole body movements w3,6,7,21–24,28,43x. The participation of the cerebellum in behavior has been extended to higher cognitive functions in humans w1,4,5,27,32x. In view of the well-known postural disorders after cerebellar damage, we wished to determine the specific
)
Corresponding author. Present address: Universite´ de Rouen, Faculte´ des Sciences, Laboratoire de Neurobiologie de l’Apprentissage, 76821 Mont-Saint-Aignan Cedex, France. Fax: q33 Ž2. 3514-6349; E-mail: [email protected] 0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 6 6 6 - 5
role of the cerebellar cortex in various equilibrium tests, more particularly in regard to sensorimotor learning and long-term retention. For this purpose, Purkinje cell degeneration Žpcd. mice were used. In this spontaneous mutation, there is a total loss during the third and fourth postnatal weeks of Purkinje cells, with relative sparing of cerebellar granule cells and deep nuclei, the latter degenerating much later, possibly due to anterograde transsynaptic degeneration secondary to the Purkinje cell loss w14,15,25,35,40x. There is also a retinal degeneration of slower onset w34,35x. Because Purkinje cells are the only output neuron from the cerebellar cortex w17x, the deep nuclei receive no input from the cerebellar cortex. However, two important afferent pathways from brainstem nuclei Žclimbing fibres and parallel fibres., conveying peripheral information to normal cerebellar nuclei by means of collateral fibres w8,33x, may be still functional in this mutation. It is already known that pcd mutant mice are severely impaired during acquisition of classical eyeblink conditioning w9x and spatial orientation w16x. The purpose of the present series of studies was to determine whether the same mutant is equally impaired in sensorimotor tests
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
requiring balance. A series of static beam and grid tests were conducted w20,21,23,24x, together with the rotorod test w6,7x and the coat-hanger test w23x. The coat-hanger test differs from standard beam tests in requiring greater muscle strength, since the animals are suspended upside down from a horizontally positioned string.
2. Materials and methods
311
body positions Žsee w23x for additional details. were measured, namely position 1 s the body of the mice is parallel to the beam with all four paws on it; position 2 s the body of the mice is perpendicular to the beam with all four paws on it; position 3 s the mice have only one front paw and one hind paw on the beam; position 4 s the mice have only two front paws on the beam. Positions 1 and 2 represent stable postures, whereas positions 3 and 4 represent unstable postures. The body position of the mice was recorded every 15 s.
2.1. Animals Female pcd mutant mice Ž pcdrpcd . and non-ataxic littermate controls Žqrpcd . of the B6C3Fe-ara strain, approximately 2 months old Žrange: 1.5–2.5 months. during the first day of testing, were purchased from Jackson Laboratory ŽBar Harbor, ME, USA. and housed in a temperature- and humidity-controlled room. Before testing, the animals were handled for a few days in order to facilitate adaptation. 2.2. Apparatus and procedures The order of presentation of the tests was as follows: large round beam, narrow round beam, rectangular beam, vertical grid, inclined grid, rotorod, and coat-hanger. The interval between the end of one test and the start of another was 2–4 days. 2.2.1. Wooden beams In the beam tests and the grid tests, 7 pcd mutants and 7 controls were used. There were two round static wooden beams covered with surgical tape. The first beam measured 8.5 cm in diameter and 80 cm in length and was divided into 8 equal segments. The second, of equal length, measured 4.6 cm in diameter. In order to prevent any escape, the endpoint of each beam was blocked by a piece of cardboard. The floor was covered by a layer of soft cloth in case of any fall. The mice were positioned in the middle part of the beams and allowed to remain on them for 60 min. There were 4 trials in a single day of testing with an intertrial interval of 10–15 min. The time spent on the beams and the number of segments crossed were measured. After these two round beam tests, a rectangular beam test was used, consisting of four interconnected wooden boards forming a rectangle Ž90 = 70 cm, width 1 cm, height from the floor 34 cm. and a fifth board inserted within the rectangle. The beam was separated into 10 cm segments, yielding a total of 39 segments Ž9 on two sides and 7 on the other three sides.. The mice were placed in the middle part of one of the boards. There were 4 trials per day, during 2 consecutive days, and an intertrial interval of 10–15 min, with a cut-off period of 60 s per trial. The time spent on the beam, the number of segments crossed, and the frequencies of occurrence of 4 specific
2.2.2. Grids Two grids were used. The mice were first placed in the middle part of a vertically positioned steel grid Ž25 = 18 cm, spacing between bars: 1.2 cm.. The number of falls and the time needed before reaching the top of the grid were measured. There were 4 trials on a single testing day with a cut-off point of 60 s per trial and an intertrial interval of 10–15 min. The second grid consisted of a wiremesh Ž38 = 38 cm, 5 squares per cm. surrounded by a wooden frame and an inclination of 308 or 608 from the horizontal. This test was conducted on the same day as the previous test. The mice were placed in the middle part of the grid facing downward. There were 4 trials Ž2 at 308; 2 at 608 in alternating order, starting at 308. on a single testing day with a cut-off point of 60 s and an intertrial interval of 10–15 min. The time needed before turning upward Žusing the whole body of the animal as the criterion, i.e. a 908 head-turn was not considered to be sufficient., the time needed before reaching the top of the grid Žsnout criterion., and the number of falls were measured. Animals successful in reaching the top were immediately removed and the trial ended. 2.2.3. Accelerating rotorod The accelerating rotorod Žmodel 7650, Stoelting, Wood Dale, IL, USA. consisted of a horizontal mast Ž3 cm in diameter., turning around its longitudinal axis, each divided into 5 sections of 5.5 cm. The mast permitted the mice to locomote without slipping because of the knurled material. Each section was separated by a piece of plastic in order to prevent mice from jumping to the adjacent section. The speed of the rotation increased gradually Žwithout any jerk. at each 30 s interval by 4 revolutions per minute Žrpm. from 4 rpm to 40 rpm Žcut-off point: 5 min per trial.. The mice were gently dropped while the apparatus was in motion. After 3–5 s, the beam rotated on the opposite side of where the mice were facing, promoting forward locomotion in order to avoid a fall. There were 4 trials per day during 10 consecutive days, with an intertrial interval of 15–20 min. On some trials, the mice clung to the rod without moving and were thus passively rotated. The time spent on the rotorod, the latency before onset of passive rotation, and the time spent in passive rotation were measured. Passive rotation was tabulated only on those trials where the mice used this strategy until
312
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
the end of the trial. Eight days after the end of the training session, a retention test was given under identical conditions. Two animals from the control group were eliminated from the study because of a tendency to jump deliberately from the mast and one suffered an injury to its front paw during the training session, leaving n s 4. To replace those mice, 4 new mice of the same age were added. None of the mice from the pcd group was observed to jump deliberately while the mast was in motion. 2.2.4. Coat-hanger During the coat-hanger test, a shipment of 4 pcd mutants arrived and were thereby added to the group, yielding a total of n s 11, in comparison to the 6 controls that had been exposed to every previous test. The triangular shaped coat-hanger consisted of a horizontally positioned string measuring 40 cm in length and 2 mm in diameter flanked by two side-bars of equal diameter measuring 19 cm in length Ž358 angle.. The coat-hanger was placed 78 cm above a cushioned table. The mice were placed upside down in the middle part of the horizontal string, care being taken to release the mice when all four paws clung to it. Latencies before falling and the movement time ŽMT. necessary to reach with two front paws one of the diagonal bars were measured. There were 4 trials per day during 10
consecutive days and the intertrial interval was 15–20 min. Eight days after the end of the training session, a retention test was given under identical conditions. 2.3. Statistical analyses The unpaired t-test was used to compare pcd and control performances on the beams and grids, except for the occurrence of body positioning in the rectangular beam test, where the Mann-Whitney U-test was used because of unequal variances. A 2 = 10 ANOVA with repeated measures on the second factor was used on the rotorod and on the coat-hanger tests, with 2 independent groups Žpcd vs. controls. and 10 days of testing, together with one-way ANOVAs as needed to evaluate the performance of each group. The paired t-test was used to compare performances between day 10 of acquisition and the retention test.
3. Results 3.1. Static beams There was no difference Ž P ) 0.05. between pcd mutant mice and controls on any measure for the round wide
Fig. 1. Means and S.E.M. of pcd mutant mice Ž n s 7. and controls Ž n s 8. for 10 days of training and one day of retention Žday 11. on the rotorod. A: time spent on the rod; B: time spent before onset of passive rotation; C: time spent in passive rotation; D: percentage of trials in passive rotation.
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316 Table 1 Mean ŽS.E.M.. values of pcd mutants and controls in three beam tests of motor coordination Measures Round wide beam Time on beam Žs. Segments Round narrow beam Time on beam Žs. Segments Rectangular beam Time on beam Segments Position 1 Position 2 Position 3 Position 4 a
313
the inclined grid. The pcd mutants turned upward more rapidly Ž P - 0.01. and reached the top of the grid more quickly Ž P - 0.01. than did controls at both inclinations ŽTable 2..
pcd mutants
Controls
240 Ž0. 29.3 Ž6.3.
240 Ž0. 28.4 Ž5.8.
3.3. Rotorod
240 Ž0. 28.6 Ž6.2.
240 Ž0. 21.1 Ž2.7.
480 Ž0. 55.6 Ž12.4. 14.0 Ž4.0. 0.4 Ž0.2. 7.6 Ž2.1. 10.0 Ž3.8.
480 Ž0. 70.6 Ž25.9. 23.0 Ž3.4. 8.1 Ž2.8. a 0 Ž0. a 0.9 Ž0.6. a
A 2 = 10 ANOVA revealed significant group Ž P 0.001. and day Ž P - 0.001. effects but no significant interaction Ž P ) 0.05. for latencies before falling and latencies before the onset of passive rotation. The pcd mutants had lower values than controls on both measures ŽFig. 1A,B.. While the control mice improved their performances on these measures Ž P - 0.001., the slight increase in pcd mutants did not reach significance Ž P ) 0.05.. A 2 = 10 ANOVA on time spent in passive rotation and the percentage of trials with passive rotation revealed a significant interaction term Ž P - 0.05.. The pcd mutants spent more time in passive rotation and had a higher number of trials with passive rotation later in training, whereas the opposite tendency was found in controls ŽFig. 1C,D.. The pcd mutants and controls had lower values Ž P 0.05. for time spent on the mast during the retention test in comparison to day 10 of acquisition. The time spent before
P - 0.05 vs. pcd mutants.
beam or the round narrow beam ŽTable 1.. Because of the lack of difference between performances on day 1 and day 2 on the rectangular beam, the results of the 2 days were combined. There was no difference Ž P ) 0.05. between pcd mutants and control mice for the time spent on the beam and for the number of segments crossed. By contrast, the positions adopted by pcd mutants and controls to maintain their equilibrium on the beam were different ŽTable 1.. The pcd mice were found more frequently than controls in two unstable positions, namely 3 and 4 Ž P 0.05 and P - 0.01, respectively. and less frequently in one stable position, namely position 2 Ž P - 0.01.. The effect on position 1 did not reach significance Ž P ) 0.05.. 3.2. Grids As only one mutant fell from the vertical grid, there was no group difference Ž P ) 0.05. in latencies before falling ŽTable 2.. Paradoxically, the time spent before reaching the top part of the grid was shorter Ž P - 0.01. in pcd mutants than in controls. None of the mice fell from Table 2 Mean ŽS.E.M.. values of pcd mutants and controls on the vertical grid and on the inclined grid Measures Vertical grid Falls Top latencies Inclined grid 308 Falls Turning time Žs. Top latencies Žs. 608 Falls Turning time Žs. Top latencies Žs. a
P - 0.01 vs. pcd mutants.
pcd mutants
Controls
0.1 Ž0.1. 45.0 Ž14.8.
0 Ž0. 131.1 Ž23.3. a
0 Ž0. 7.9 Ž2.4. 17.1 Ž3.4.
0 Ž0. 21.0 Ž4.3. a 40.9 Ž7.1. a
0 Ž0. 7.0 Ž2.0. 18.6 Ž2.6.
0 Ž0. 33.4 Ž9.4. a 73.0 Ž12.1. a
Fig. 2. Means and S.E.M. of pcd Ž ns11. and controls Ž ns6. for 10 days of training and one day of retention Žday 11. on the coat-hanger. A: latencies before reaching the side-bar with two paws; B: latencies before falling.
314
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
onset of passive rotation was shorter for controls Ž P - 0.05. but not for the mutants Ž P ) 0.05.. Conversely, the time spent in passive rotation was lower in pcd mutants Ž P 0.05. but not in controls Ž P ) 0.05.. There was no difference for either group in the percentage of trials with passive rotation Ž P ) 0.05.. 3.4. Coat-hanger A two-way ANOVA revealed no significant group, day or interaction effect Ž P ) 0.05. for the MT measure ŽFig. 2A.. A two-way ANOVA revealed a significant interaction for latencies before falling Ž P - 0.05.. As shown in Fig. 2B, the latencies before falling increased as a function of time in pcd mutants Ž P - 0.001. but not in controls Ž P ) 0.05., as their values were nearly maximal throughout the testing period. There was no difference between the retention score and the score on the final day of acquisition for either group Ž P ) 0.05..
4. Discussion The pcd mutants in the present study were tested at a young age, at a time when the major neuropathological change is a loss of Purkinje cells w14,15,25,35,40x, although the functional consequences of retrograde and anterograde degeneration cannot be excluded. By contrast to other cerebellar mutant mice with more widespread cerebellar pathology w20,21x, pcd mutants were not impaired on static beam and grid tests. Irrespective of beam width or shape, pcd mutants did not fall sooner than controls. Moreover, in spite of ataxia, the distance travelled by pcd mutants on the three beams equalled that of controls. In a previous study w23x, pcd mutants had normal activity levels in a T-maze, an indication of the mildness of their locomotor deficits. The only measure where they differed from controls was the stability of their posture on the rectangular beam. This result is similar to that found in lurcher mutant mice, another cerebellar mutant with degeneration of the cerebellum and the inferior olive w24x. On the two grid tests, the performance of the pcd mutants was not deficient either. Paradoxically, the time taken before reaching the top of the grids was shorter in pcd mutants than in controls. Moreover, the time taken before turning upward on the inclined grid was shorter in the mutant group at both inclinations. The latter result may be explained by the poorer stability of the mutants. Whereas control mice explored the bottom and the top parts of the grid in a similar manner, the mutants were probably able to attain a more stable body position by facing upward on the inclined slope. This interpretation is supported by the finding that lurcher mutants were also observed to turn upward more quickly than normal mice on the inclined grid w19x. The pcd mutants also reached the top of the grid more quickly than the control group. Because of the
instability of their body position, the mutants probably attempted to reach a horizontal position or escape from the apparatus altogether. However, by contrast to the results in pcd mutants, lurcher mutants fell sooner from the vertically positioned grid than normal mice, a result that may be due to their more severe cerebellar pathology w23x. By contrast to lurcher mutants w23x, pcd mutants were not deficient on the coat-hanger in terms of MT. The mutants fell from the bar sooner than controls, but this result only occurred early in training. Neither mutants nor controls were impaired after an 8-day retention interval for either measure. Only on the rotorod did large mutant–nonmutant differences emerge. Control mice were far superior to pcd mutants on latencies before falling from the rod and were slower before adopting the passive rotation strategy. Moreover, whereas the control group decreased the time spent in passive rotation across days, on the contrary this behavior was increased with time in pcd mutants. This result indicates that normal mice were able to improve their latencies before falling by learning a complex sensorimotor skill, walking forward in conjunction with increasing rotation speed. By contrast, pcd mutants were unable to improve significantly the time spent on the mast despite an increase in the time spent in passive rotation. This lack of improvement is in contrast to their increase in latencies before falling on the coat-hanger. These results indicate improved motor function in a test where postural adjustments can be self-initiated but no improvement in a test where postural adjustments must be made in response to an external stimulus. Anderson and Steinmetz w2x and Raymond et al. w36x have suggested that the cerebellar cortex plays a crucial role in the timing of learned movements. This hypothesis is supported by the results of the present study. The pcd mutants had a severe deficit in synchronizing their movements in time to that of the rotating mast. Although no learning was demonstrated in the mutants within the confines of this study, it is possible that sensorimotor learning can occur with additional trials. We have found an improvement of latencies before falling in lurcher mutants under the same experimental conditions, an indication that even with massive cerebellar cortical pathology some learning can occur w28x. The poorer performance of pcd mutants is paradoxical because they have less extensive cerebellar pathology. The falling latencies of lurcher mutants increased significantly during the 10-day period Žmeans and S.E.M.. from 391 Ž44. to 749 Ž44. s, while those of pcd mutants were only slightly Žnon-significantly. higher, from 401 Ž34. to 615 Ž100.. This lack of a significant increase is in part due to the high intragroup variability. Thus, the initial performance of both groups was approximately the same. However, the terminal performance of lurcher mutants was higher and less variable. Another difference between the mutants is that whereas latencies before the onset of passive rotation increased
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
considerably in lurchers, from 347 Ž37. on day 1 to 632 Ž38. on day 10, those of pcds remained approximately the same, from 371 Ž28. to 462 Ž42.. Thus, the enhanced latencies before falling in lurchers are not due to passive clinging, whereas in pcds there was no significant increase in latencies before falling in spite of enhanced passive clinging. We conclude that a more widespread degeneration of cerebellar cell populations does not necessarily lead to poorer motor performance. Metabolic changes in cerebellar afferent or efferent structures in lurcher mutants may compensate for loss in cerebellar function. In the present study, the performance of control mice on the retention test was slightly lower than their performance on day 10 of acquisition in terms of latencies before falling and latencies before onset of passive rotation. Thus, some loss of memory of the acquired sensorimotor skills occurred in normal mice. The performance of pcd mutants also worsened after the retention interval in terms of time spent on the rod. By contrast to controls, pcd mice spent less time in passive rotation during the retention test. Thus, each group lost during the retention interval their preferred strategy. Control mice were less able to walk forward in synchrony with the moving beam, whereas pcd mice appeared to forget, at least in part, their previous tendency of clinging passively to the beam. Chen et al. w9x reported an impairment in pcd mutant mice during classical eyeblink conditioning. Our results show that selective atrophy of Purkinje cells impaired acquisition of a task requiring precise forward walking in response to predictable rotation speeds of a mast. Further testing of this mutant on other sensorimotor tasks will delineate the role of the cerebellar cortex on learning.
w8x w9x
w10x w11x
w12x
w13x w14x
w15x
w16x
w17x w18x
w19x w20x w21x
Acknowledgements
w22x
This research was funded by a grant from the Natural Sciences and Engineering Research Council of Canada.
w23x w24x
References w1x N.A. Akshoomoff, E. Courchesne, G.A. Press, V. Iraqui, Contributions of the cerebellum to neuropsychological functioning, Neuropsychologia 30 Ž1992. 315–338. w2x B.J. Anderson, J.E. Steinmetz, Cerebellar and brainstem circuits involved in classical eyeblink conditioning, Rev. Neurosci. 5 Ž1994. 251–273. w3x N. Auvray, J. Caston, A. Reber, T. Stelz, Role of the cerebellum in the ontogenesis of the equilibrium behavior in the young rat: a behavioral study, Brain Res. 505 Ž1989. 291–301. w4x T. Botez-Marquard, M.I. Botez, Cognitive behavior in heredodegerative ataxias, Eur. Neurol. 33 Ž1993. 351–357. w5x T. Botez-Marquard, I. Routhier, Reaction time and intelligence in patients with olivopontocerebellar atrophy, Neuropsychiatry Neuropsychol. Behav. Neurol. 8 Ž1995. 168–175. w6x J. Caston, N. Delhaye-Bouchaud, J. Mariani, Motor behavior of heterozygous staggerer mutant Žqr sg . versus normal Žqrq. mice during aging, Behav. Brain Res. 72 Ž1996. 97–102. w7x J. Caston, F. Vasseur, T. Stelz, C. Chianale, N. Delhaye-Bouchaud,
w25x
w26x
w27x w28x
w29x
w30x
315
J. Mariani, Differential role of cerebellar cortex and deep cerebellar nuclei in learning the equilibrium behavior: studies in intact and cerebellectomized lurcher mutant mice, Dev. Brain Res. 86 Ž1995. 311–316. V. Chan-Palay, Cerebellar Dentate Nucleus: Organization, Cytology and Transmitters, Springer, Berlin, 1977. L. Chen, S. Bao, J.M. Lockard, J.J. Kim, R.F. Thompson, Impaired classical eyeblink conditioning in cerebellar-lesioned and Purkinje cell degeneration Žpcd. mutant mice, J. Neurosci. 16 Ž1996. 2829– 2838. H. Collewijn, The modifiability of the adult vestibulo-ocular reflex, Trends Neurosci. 2 Ž1979. 98–102. M. Dahhaoui, J. Caston, N. Auvray, A. Reber, Role of the cerebellum in an avoidance conditioning task in the rat, Physiol. Behav. 47 Ž1990. 1175–1180. M. Dahhaoui, J. Caston, J. Lannou, S. Avenel, Role of the cerebellum in habituation exploration behavior in the rat, Physiol. Behav. 52 Ž1992. 339–344. R.S. Gelman, F.A. Miles, A new role for the cerebellum in conditioning, Trends Neurosci. 8 Ž1985. 181–182. B. Ghetti, C.J. Alyea, J. Muller, Studies on the Purkinje cell degeneration Žpcd. mutant mice: primary pathology and transneuronal changes, J. Neuropathol. Exp. Neurol. 37 Ž1978. 617. B. Ghetti, J. Norton, L.C. Triarhou, Nerve cell atrophy and loss in the inferior olivary complex of ‘Purkinje cell degeneration’ mutant mice, J. Comp. Neurol. 260 Ž1987. 409–422. R.C. Goodlett, K.M. Hamre, J.R. West, Dissociation of spatial navigation and visual guidance performance in Purkinje cell degeneration Žpcd. mutant mice, Behav. Brain Res. 47 Ž1992. 129–141. M. Ito, The Cerebellum and Neural Control, Raven, New York, 1984. M. Ito, Synaptic plasticity in the cerebellar cortex that may underlie the vestibulo-ocular adaptation, in: A. Berthoz, G.M. Melvill-Jones ŽEds.., Adaptative Mechanisms in Gaze Control: Facts and Theories, Elsevier, Amsterdam, 1985, pp. 213–221. R. Lalonde, Motor learning in lurcher mutant mice, Brain Res. 639 Ž1994. 351–353. R. Lalonde, Motor abnormalities in staggerer mutant mice, Exp. Brain Res. 68 Ž1987. 417–420. R. Lalonde, Motor abnormalities in weaver mutant mice, Exp. Brain Res. 65 Ž1987. 479–481. R. Lalonde, M. Manseau, M.I. Botez, Spontaneous alternation and exploration in Purkinje cell degeneration mutant mice, Brain Res. 411 Ž1987. 187–189. R. Lalonde, M.I. Botez, C.C. Joyal, M. Caumartin, Motor abnormalities in lurcher mutant mice, Physiol. Behav. 51 Ž1992. 523–525. R. Lalonde, C.C. Joyal, J.-M. Guastavino, M.I. Botez, Hole poking and motor coordination in lurcher mutant mice, Physiol. Behav. 54 Ž1993. 41–44. S.C. Landis, R.J. Mullen, The development and degeneration of Purkinje cells in pcd mutant mice, J. Comp. Neurol. 177 Ž1978. 125–144. D.G. Lavond, T. Hembree, R.F. Thompson, Effect of kainic acid lesions of the cerebellar interpositus nucleus on eyelid conditioning in the rabbit, Brain Res. 326 Ž1985. 179–182. H.C. Leiner, A.L. Leiner, R.S. Dow, Cognitive and language functions of human cerebellum, Trends Neurosci. 11 Ž1993. 444–447. N. Le Marec, J. Caston, R. Lalonde, Impaired motor skills on static and mobile beams in lurcher mutant mice, Exp. Brain Res. Ž1997. in press. D.A. McCormick, R.F. Thompson, Neuronal responses of the rabbit cerebellum during acquisition and performance of a classically conditioned nictitating membrane-eyelid response, J. Neurosci. 4 Ž1984. 2811–2822. D.A. McCormick, R.F. Thompson, Cerebellum: essential involvement in the classically conditioned eyelid response, Science 223 Ž1984. 296–299.
316
N. Le Marec, R. Lalonder Brain Research 768 (1997) 310–316
w31x D.A. McCormick, D.G. Lavond, G.A. Clark, R.E. Keetner, C.A. Rising, R.F. Thompson, The engram found? Role of the cerebellum in classical conditioning of nictating membrane and eyelid response, Bull. Psychon. Soc. 18 Ž1981. 103–105. w32x F.A. Middleton, P.L. Strick, Anatomical evidence for cerebellar and basal ganglia involvement in higher cognitive function, Science 266 Ž1994. 456–461. w33x G.A. Mihailoff, Identification of pontocerebellar axon collateral synaptic boutons in the rat cerebellar nuclei, Brain Res. 48 Ž1994. 313–318. w34x R.J. Mullen, M.M. LaVail, Two new types of retinal degeneration in cerebellar mutant mice, Nature 258 Ž1975. 528–530. w35x R.J. Mullen, E.M. Eicher, R.L. Sidman, Purkinje cell degeneration: a new neurological mutation in the mouse, Proc. Natl. Acad. Sci. USA 73 Ž1976. 208–212. w36x J.L. Raymond, S.G. Lisberger, M.D. Mauk, The cerebellum: a neuronal learning machine?, Science 272 Ž1996. 1126–1131. w37x D.A. Robinson, Adaptative gain control of vestibulo-ocular reflex by the cerebellum, J. Neurophysiol. 39 Ž1976. 954–969.
w38x J.E. Steinmetz, S.F. Logue, J.J. Steinmetz, Rabbit classically conditioned eyelid responses do not reappear after interpositus nucleus lesion and extensive post-lesion training, Behav. Brain Res. 51 Ž1992. 103–114. w39x R.F. Thompson, The neurobiology of learning and memory, Science 233 Ž1986. 941–947. w40x L.C. Triarhou, J. Norton, B. Ghetti, Anterograde transsynaptic degeneration in the deep cerebellar nuclei of Purkinje cell degeneration mutant mice, Exp. Brain Res. 66 Ž1987. 577–588. w41x C.H. Yeo, M.J. Hardiman, M. Glickstein, Discrete lesions of the cerebellar cortex abolish the classically conditioned nictating membrane response in the rabbit, Behav. Brain Res. 13 Ž1984. 261–266. w42x C.H. Yeo, M.J. Hardiman, M. Glickstein, Classical conditioning of the nictitating membrane response of the rabbit. II. Lesions of the cerebellar nuclei, Exp. Brain Res. 60 Ž1985. 87–98. w43x C. Zion, N. Auvray, J. Caston, A. Reber, T. Stelz, Effects of cerebellectomy at day 15 on the ontogenesis of the equilibrium behavior in the rat, Brain Res. 515 Ž1990. 104–110.