Large-scale reorganization of the tonotopic map in mouse auditory midbrain revealed by MRI

Abstract

The cortex is thought to be the primary site of sensory plasticity, particularly during development. Here, we report that large-scale reorganization of the mouse auditory midbrain tonotopic map is induced by a specific sound-rearing environment consisting of paired low- (16 kHz) and high-frequency (40 kHz) tones. To determine the potential for plasticity in the mouse auditory midbrain, we used manganese-enhanced MRI to analyze the midbrain tonotopic maps of control mice during normal development and mice reared in the two-tone (16 + 40 kHz) environment. We found that the tonotopic map emerged during the third postnatal week in normal mice. Before 3 weeks, a larger percentage of auditory midbrain responded to each of the suprathreshold test frequencies, despite the fact that the primary afferent projections are in place even before hearing onset. By 3 weeks, the midbrain tonotopic map of control mice was established, and manganese-enhanced MRI showed a clear separation between the 16- and 40-kHz responses. Two-tone rearing dramatically altered the appearance of these discrete frequency-specific responses. A significant volume of the auditory midbrain became responsive to both rearing frequencies, resulting in a large-scale reorganization of the tonotopic map. These results indicate that developmental plasticity occurs on a much greater scale than previously appreciated in the mammalian auditory midbrain.

Fig. 1.

Pure-tone stimulation induced tonotopic signal enhancement in the IC. Averaged coronal IC images (n ≥ 8 in each group) demonstrated spatial differences in MEMRI enhancement between mice kept in a quiet environment (A) and mice exposed to either 16 kHz (B) or 40 kHz (C). Pseudocolor-coded images (D–F) made it easier to appreciate the enhancement patterns (color scale included). t-test analysis, comparing either the 16-kHz-exposed group to the quiet controls (green) (G) or the 40-kHz-exposed group to the quiet controls (red) (H), shows the regions of significant differences in MEMRI enhancement (P < 0.05) in general agreement with the established electrophysiological tonotopic map (Inset) (29).

Fig. 2.

Volumetric analysis is used to quantify the developmental differences in activity. (A) MEMRI 2D coronal images and 2D p-maps show the developmental changes in tone-induced activity (color scale for p-maps is shown). (B) Three-dimensional p-maps (P < 0.05) show the 16-kHz (green) and 40-kHz (red) activity patterns at P13 (Top), P16 (Middle), and P19 (Bottom). At each stage, the 3D p-maps are displayed in both coronal (Left) and sagittal (Center) views. The region of overlap between 16 and 40 kHz (yellow) (Right) is also shown for each postnatal age. Quantitative analysis demonstrated a decrease in the overlap volume (C) and an increase in the centroid-to-centroid separation (D) with increasing developmental stage.

Fig. 3.

Two-tone (16 + 40 kHz) rearing induced changes in both the 16- and 40-kHz IC activity patterns. Three-dimensional p-maps of normal (A–D) and two-tone-reared mice (E–H) demonstrate marked differences in both the 16-kHz (green) (A and E) and 40-kHz (red) (B and F) activity patterns (P < 0.05; n ≥ 8 in each group). The superimposed activity maps show differences in overlap (yellow) between 16- and 40-kHz patterns in two-tone-reared mice (G and H) compared with normal controls (C and D). In two-tone-reared mice, the regions of overlap between the 16- and 40-kHz patterns (yellow) (H) clearly separate the altered distribution into dorsal (putative 16 kHz) and ventral (putative 40 kHz) IC. (I–L) Quantitative ROI analysis (dorsal ROI, green; ventral ROI, red) (Inset) shows significant differences between dorsal and ventral MEMRI signal enhancement in normal mice at both 16-kHz (I) and 40-kHz (J) (∗, P < 0.01). In contrast, the two-tone-reared mice show no dorsal–ventral differences at either 16 kHz (K) or 40 kHz (L).

Fig. 4.

The changes in IC activity patterns were specific to the rearing tones. Similar activity patterns (blue) were observed after 32-kHz stimulation of normal (A) (P < 0.05, n ≥ 8) and two-tone-reared mice (B) (P < 0.05, n ≥ 8). Superimposed p-maps (P < 0.05; n ≥ 8 in each group) show the relative positions of the 16-kHz (green), 32-kHz (blue), and 40-kHz (red) activity patterns in normal (C and E) and two-tone-reared mice (D and F). The corresponding electrophysiological tonotopic map is also shown for reference (Inset).