Cytoarchitectural disruption of the superior colliculus and an enlarged acoustic startle response in the Tuba1a mutant mouse

Abstract

The Jenna mutant mouse harbours an S140G mutation in Tuba1a that impairs tubulin heterodimer formation resulting in defective neuronal migration during development. The consequence of decreased neuronal motility is a fractured pyramidal cell layer in the hippocampus and wave-like perturbations in the cerebral cortex. Here, we extend our characterisation of this mouse investigating the laminar architecture of the superior colliculus (SC). Our results reveal that the structure of the SC in mutant animals is intact; however, it is significantly thinner with an apparent fusion of the intermediate grey and white layers. Birthdate labelling at E12.5 and E13.5 showed that the S140G mutation impairs the radial migration of neurons in the SC. A quantitative assessment of neuronal number in adulthood reveals a massive reduction in postmitotic neurons in mutant animals, which we attribute to increased apoptotic cell death. Consistent with the role of the SC in modulating sensorimotor gating, and the circuitry that modulates this behaviour, we find that Jenna mutants exhibit an exaggerated acoustic startle response. Our results highlight the importance of Tuba1a for correct neuronal migration and implicate postnatal apoptotic cell death in the pathophysiological mechanisms underlying the tubulinopathies.

Fig. 1

The structure of the SC in Jna/+ mutants. (A) Dorsal view of wild-type (WT) and Jna/+ brains. The red arrow highlights the exposed SC in mutant animals. (B) Coronal Nissl stain of WT and mutant animals illustrating the thinner SC in Jna/+ mutants. (C) Quantification of the thickness of the SC in WT and mutant animals at P0 and 12 wk reveals a significant difference between controls and Jna/+ animals (F=18.1, P<0.001, F=102.4, P<0.0001). (D) Diagram illustrating the laminar structure of the SC adapted from Paxinos and Watson's Brain Atlas (Paxinos et al., 2007). The SC consists of seven layers: the zonal layer (Zn), the superficial grey layer (SuG), the optic layer (Op), the intermediate grey layer (InG), the intermediate white layer (InW), the deep grey layer (DpG) and the deep white (DpW) layer. Other abbreviation: PAG, periaqueductal grey. (E) Calretinin, TPD52I1 and ER81 staining (top to bottom) reveal that the laminar structure of the SC is intact in Jna/+ mutants. (F, G) Sagittal sections of the SC in WT littermates and mutant animals reveal that it is elongated in Jna/+ mutants (G). Red arrow between dotted black lines shows the SC. Scale bars 5 mm (A), 1 mm (B, F, G) and 250 μm (E).

Fig. 2

Myeloarchitectural investigation of the SC in Jna/+ mutants. (A) Diagram illustrating the laminar structure of the SC adapted from Paxinos and Watson's Brain Atlas (Paxinos et al., 2007). (B) Diagram showing the layers of the SC aligned with Gallyas-stained WT and mutant sections. Equally proportioned analysis boxes (shown in white) were drawn from the aqueduct to the dorsal surface of Zn layer. These boxes map to similar anatomical regions. Boxes 1–5 align with the superficial and intermediate layers (Zn, SuG, Op InG, InW), Boxes 6, 7 the deep layers (DpG, DpW) and Boxes 8–10 the PAG. (C, D) Enlarged images of the superficial and intermediate layers of Gallyas-stained sections in WT and mutant animals. The superficial SC of the mutant appears compacted and less defined with a fusion of the InG and InW (white arrow). (E, F) Enlarged images of the deep layers (DpG and DpW) of Gallyas-stained sections in WT and mutant animals. Sections appear similar, with crossing of the collicular commissure observed in both mutants and WT animals. Scale bars show 250 μm.

Fig. 3

Impaired neuronal migration in the SC of Jna/+ mutants. (A, B) Staining for BrdU in the SC of littermate controls and Jna/+ mutants harvested at P0 after injection of BrdU at E12.5 (n=4) and E13.5 (n=3). The SC was divided into 10 equal bins (shown in white) extending from the aqueduct to the dorsal surface of Zn layer and BrdU-positive cells were counted. (C, D) There was no significant difference in the average number of cells labelled per section between WT littermates and Jna/+ mutants. (E, F) The distribution of BrdU-positive cells per zone in WT controls and Jna/+ mutants after injections at E12.5 (E) and E13.5 (F). The percentage of BrdU cells in the upper two zones was significantly reduced in Jna/+ mutants when injected at E12.5 (layer 1 [F=20.3, P<0.001], layer 2 [F=27, P<0.001]), and higher in layer 6 (F=11.78, P<0.01). Similarly, when injected at E13.5, the percent of BrdU-labelled cells in zones 2 (F=10.56, P<0.05) and 3 (F=18.32, P<0.001) was reduced in Jna/+ animals (n=3), complemented by a higher portion observed in zones 6 (F=20.9, P<0.0001) and 7 (F=15.39, P<0.01) (n=3). (G) Staining of postmitotic neurons with the marker NeuN reveals no significant difference in the percentage of neurons per a zone at P21. Scale bars in A and B show 250 μm.

Fig. 4

Loss of neurons and increased apoptosis in the SC of Jna/+ mutants. (A, B) The SC was divided into 10 equal bins and DAPI (blue) and NeuN (green)-labelled cells counted at P21 and 12 wk of age for both WT littermates and Jna/+ mutants (n=4). This enabled the percentage of neurons to be calculated for each zone at each time point (C, D). We compared the percentage of neurons in WT littermates and Jna/+ mutants and found no significant difference between genotypes for any zone in the P21 cohort (C). In contrast, there was a massive reduction in the percentage of neurons throughout the SC in 12-wk-old Jna/+ mutants (D). The loss of neurons reached significance in all zonal boxes, particularly those mapping to the DpG/DpW layers (zone 6 [F=33.07, P<0.001], zone 7 [F=41.57, P<0.001]). (E, F) Immunostaining of the SC of 7-wk-old animals revealed a significant increase in the total number (F=21.2, P<0.01) and density (F=72.1, P<0.0001) of caspase-positive cells (black arrows) in Jna/+ mutants (n=5) when compared with WT littermates (G, H). Scale bars show 250 μm.

Fig. 5

Exaggerated acoustic startle response in Jna/+ mutants. (A, B) Assessment of the acoustic startle response in male (n=7) and female Jna/+ mutants (n=8) in response to randomised 40 ms bursts of 90 dB, 100 db, 110 db and 120 db white noise. In the absence of a weight correction, the acoustic startle response is significantly higher in both males and female Jna/+ mutants when exposed to a 120 dB stimulus (male [F=56.1, P<0.0001], female [F=48.08, P<0.0001]). (C, D) When these data are normalised to take into account the differences in weight between genotypes, mutant animals of both sexes display a significantly larger acoustic startle response to 110 dB (male [F=17.85, P<0.001], female [F=15.5, P<0.001]) and 120 dB stimuli (male [F=134.8, P<0.0001], female [F=96.8, P<0.0001]).