Gene expression alterations in the cerebellum and granule neurons of Cstb(-/-) mouse are associated with early synaptic changes and inflammation

Abstract

Progressive myoclonus epilepsy of Unverricht-Lundborg type (EPM1) is an autosomal recessively inherited neurodegenerative disease, manifesting with myoclonus, seizures and ataxia, caused by mutations in the cystatin B (CSTB) gene. With the aim of understanding the molecular basis of pathogenetic events in EPM1 we characterized gene expression changes in the cerebella of pre-symptomatic postnatal day 7 (P7) and symptomatic P30 cystatin B -deficient (Cstb(-/-) ) mice, a model for the disease, and in cultured Cstb(-/-) cerebellar granule cells using a pathway-based approach. Differentially expressed genes in P7 cerebella were connected to synaptic function and plasticity, and in cultured cerebellar granule cells, to cell cycle, cytoskeleton, and intracellular transport. In particular, the gene expression data pinpointed alterations in GABAergic pathway. Electrophysiological recordings from Cstb(-/-) cerebellar Purkinje cells revealed a shift of the balance towards decreased inhibition, yet the amount of inhibitory interneurons was not declined in young animals. Instead, we found diminished number of GABAergic terminals and reduced ligand binding to GABAA receptors in Cstb(-/-) cerebellum. These results suggest that alterations in GABAergic signaling could result in reduced inhibition in Cstb(-/-) cerebellum leading to the hyperexcitable phenotype of Cstb(-/-) mice. At P30, the microarray data revealed a marked upregulation of immune and defense response genes, compatible with the previously reported early glial activation that precedes neuronal degeneration. This further implies the role of early-onset neuroinflammation in the pathogenesis of EPM1.

Figure 1. Differential expression of Gabra6 and Gabrd.

qPCR shows increased expression of Gabra6 and Gabrd in P7 Cstb^−/− cerebellum compared to the control mice. The data are expressed as a fold change relative to controls ± SE. *, p<0.05.

Figure 2. Spontaneous postsynaptic currents in Purkinje cells.

(A) Representative traces of spontaneous EPSCs and IPSCs in control and Cstb^−/− Purkinje cells. EPSCs are seen as inward currents (downward deflections) and IPSCs are seen as outward currents (upward deflections). Insets show single IPSCs and EPSCs taken from the time points indicated by asterisks. An arrow indicates a synchronous burst of IPSCs seen in control but not in Cstb^−/− Purkinje cells. (B) The occurrence of IPSCs and synchronous burst of IPSCs on control and Cstb^−/− Purkinje cells. Individual cells measured are shown as spheres. (C) The frequency of EPSCs was significantly higher in Cstb^−/− cells compared to controls (p = 0.034). The data are expressed as mean frequency (Hz) ± SE. *, p<0.05; **, p<0.01.

Figure 3. The expression of synapsin 1, VGAT and gephyrin positive synaptic puncta.

(A) The number of synapsin 1 positive puncta in the molecular layer of Cstb^−/− cerebellum was significantly lower compared to controls at P14 (p = 0.0035) and P20 (p = 0.0447). P7 and P30 animals did not show significant difference. (B) Molecular layer of P14 and P20 Cstb^−/− cerebellum shows less synapsin 1 positive puncta compared to control. (C) The number of VGAT positive puncta in the molecular layer of Cstb^−/− cerebellum was significantly lower compared to controls at P14 (p = 0.0115). P7, P20 and P30 animals did not show significant difference. (D) Molecular layer of P14 Cstb^−/− cerebellum shows less VGAT positive puncta compared to control. (E) The number of gephyrin positive puncta in the molecular layer of Cstb^−/− cerebellum was significantly lower compared to controls at P20 (p = 0.0448) and P30 (p = 0.0131). P7 and P14 animals did not show significant difference. (F) Molecular layer of P20 and P30 Cstb^−/− cerebellum shows less gephyrin positive puncta compared to control. The data are expressed as mean amount of positive puncta relative to control ± SE. *, p<0.05; **, p<0.01.

Figure 4. Autoradiography of GABA_A receptors.

(A) Binding of [³H]muscimol was significantly reduced in P30 Cstb^−/− cerebellum (p = 0.037). No change was seen at P7. (B) Binding of [³H]Ro15-4513 did not show significant changes at P7 or at P30. However, when diazepam (DZ) was added to reveal the diazepam-insensitive α6 subunit-dependent GABA_A receptor subtype, decreased binding for [³H]Ro15-4513 (p = 0.012) was seen. The data are expressed as mean radioactivity levels (nCi/mg) ± SE. *, p<0.05. (C) Representative images of [³H]muscimol binding to P30 Cstb^−/− and control brain. (D) Representative images of [³H]Ro15-4513 binding to P30 Cstb^−/− and control brain without and (E) in presence of diazepam.