Asymmetric recovery in cerebellar-deficient mice following unilateral labyrinthectomy

Abstract

The term "vestibular compensation" refers to the resolution of motor deficits resulting from a peripheral vestibular lesion. We investigated the role of the cerebellum in the compensation process by characterizing the vestibuloocular reflex (VOR) evoked by head rotations at frequencies and velocities similar to those in natural behaviors in wild-type (WT) versus cerebellar-deficient Lurcher (Lc/+) mice. We found that during exploratory activity, normal mice produce head rotations largely consisting of frequencies < or =4 Hz and velocities and accelerations as large as 400 degrees/s and 5,000 degrees/s2, respectively. Accordingly, the VOR was characterized using sinusoidal rotations (0.2-4 Hz) as well as transient impulses (approximately 400 degrees/s; approximately 2,000 degrees/s2). Before lesions, WT and Lc/+ mice produced similar VOR responses to sinusoidal rotation. Lc/+ mice, however, had significantly reduced gains for transient stimuli. After unilateral labyrinthectomy, VOR recovery followed a similar course for WT and Lc/+ groups during the first week: gain was reduced by 80% for ipsilesionally directed head rotations on day 1 and improved for both strains to values of approximately 0.4 by day 5. Moreover, responses evoked by contralesionally directed rotations returned to prelesion in both strains within this period. However, unlike WT, which showed improving responses to ipsilesionally directed rotations, recovery plateaued after first week for Lc/+ mice. Our results show that despite nearly normal recovery in the acute phase, long-term compensation is compromised in Lc/+. We conclude that cerebellar pathways are critical for long-term restoration of VOR during head rotation toward the lesioned side, while noncerebellar pathways are sufficient to restore proper gaze stabilization during contralesionally directed movements.

FIG. 1.

Dynamics of natural head movements. Head movements were recorded in unrestrained wild-type (WT) mice while they explored their environment. A: Fourier analysis of angular head-velocity profiles reveals little frequency content >5 Hz. B and C: analysis of corresponding velocities (B) and accelerations (C) shows that majority (>85%) consisted of movements with peak velocities <100°/s and peak accelerations <5,000°/s². Note, movements were symmetrically distributed between left and right. Shaded areas represent the range of stimuli used in the study to probe the vestibuloocular reflex (VOR).

FIG. 2.

VOR in wild-type and Lurcher mice before lesion. A: example eye and head (table) velocities during sinusoidal rotations in the dark at 0.5 (A1) and 2 Hz (A2) for a WT mouse. B: gain (B1) and phase (B2) plots of responses to 0.2- to 4 -Hz rotations averaged across the groups of WT (▵ and ▴) and Lc/+ (□ and ▪) mice. No significant differences were observed between the 2 strains before lesion. C: responses to transient impulses. 1: example of a WT mouse's response to a transient rotation applied in the dark. 2: gains of VOR responses to abrupt stimulations responses averaged across the groups of WT (▵ and ▴) and Lc/+ (□ and ▪) mice; WT produced almost perfect compensatory responses, whereas Lc/+ mice showed relatively lower gains. No difference between right and left responses was observed before lesions. *, significant differences (P < 0.05) between WT and Lc/+ mice. Note, that in this and the following figures: ↑, the onset of the stimulation; head velocity traces have been inverted to facilitate comparison with eye velocity; and error bars indicate means ± SE.

FIG. 3.

VOR evoked by sinusoidal rotations in wild-type mice 1 day after unilateral labyrinthectomy. A: example eye and head velocity traces, during sinusoidal rotations of the head-on-body in the dark for a WT mouse 1 day after lesion. An asymmetry between ipsi- and contralesionally directed responses is more evident at higher frequencies of rotations [compare A1 (0.5 Hz) and A2 (2 Hz)]. B: gain (1) and phase (2) plots of responses to 0.2- to 4-Hz rotations averaged across the group of WT mice. Note that ipsilesional responses are virtually absent and that contralesional as well as ipsilesional gains were significantly reduced compared with control. *, values are significantly different from prelesion values (P < 0.05).

FIG. 4.

Time course of VOR compensation for wild-type vs. lurcher mice: sinusoidal rotations. A: gain plots of the VOR response evoked 5 days after lesion by 0.2- to 4-Hz rotations. The response gains evoked by both ipsilesionally (A1) and contralesionally (A2) directed rotations are comparable for WT and Lc/+ mice. Also note that in both strains, the gain of the VOR evoked by contralesionally directed movements is comparable to prelesion values. B: gain plots of the VOR response evoked 20 days after the lesion for ipsilesionally (1) and contralesionally (2) directed movements. While the response of WT animals showed additional recovery (relative to day 5), Lc/+ gains showed no significant improvement. Significant differences compared with prelesion values (black asterisk) and between lesioned WT and Lc/+ values (gray asterisk) are indicated.

FIG. 5.

Time course of VOR compensation for wild-type vs. lurcher mice: transient stimulation. A: example of a WT mouse's response to a transient rotation applied in the dark 1 day after lesion. A, 1 and 2: the VOR evoked by ipsi- and contralesionally directed rotations, respectively. B: time course of the mean VOR gain measured in WT vs. Lc/+ mice before and after lesion. Gains are shown for ipsilesionally (B1) and contralesionally (B2) directed abrupt impulses. While response recovery in both strains was comparable for contralesionally directed rotations, responses to ipsilesionally directed rotations plateaued around day 5 for Lc/+ mice. In contrast to responses evoked by sinusoidal stimulation, responses to ipsilesionally directed transients remained significantly reduced relative to prelesion values for both WT and Lc/+ mice—even 3 wk after lesion. Significant differences (P < 0.05) are denoted by asterisk.

FIG. 6.

Summary of the recovery: wild-type vs. Lurcher mice. Time course of percentage recovery of the VOR evoked by ipsilesionally (A) and contralesionally (B) directed rotations for WT (open symbols) vs. Lc/+ mice (closed symbols). For Lc/+ mice, average values are shown for testing using low (circles, 0.2 Hz) and higher (squares, 4 Hz) frequency sinusoidal stimulation as well as using transient impulses (triangles). Response gains were normalized to prelesion during each condition to compute the percentage recovery. For WT mice, mean response gains were computed across paradigms (see text). The shaded areas represent the SE across the values measured values at days 1, 5, 10 and 20.