Sustained overexpression of spliced X-box-binding protein-1 in neurons leads to spontaneous seizures and sudden death in mice

Abstract

The underlying etiologies of seizures are highly heterogeneous and remain incompletely understood. While studying the unfolded protein response (UPR) pathways in the brain, we unexpectedly discovered that transgenic mice (XBP1s-TG) expressing spliced X-box-binding protein-1 (Xbp1s), a key effector of UPR signaling, in forebrain excitatory neurons, rapidly develop neurologic deficits, most notably recurrent spontaneous seizures. This seizure phenotype begins around 8 days after Xbp1s transgene expression is induced in XBP1s-TG mice, and by approximately 14 days post induction, the seizures evolve into status epilepticus with nearly continuous seizure activity followed by sudden death. Animal death is likely due to severe seizures because the anticonvulsant valproic acid could significantly prolong the lives of XBP1s-TG mice. Mechanistically, our gene profiling analysis indicates that compared to control mice, XBP1s-TG mice exhibit 591 differentially regulated genes (mostly upregulated) in the brain, including several GABA_A receptor genes that are notably downregulated. Finally, whole-cell patch clamp analysis reveals a significant reduction in both spontaneous and tonic GABAergic inhibitory responses in Xbp1s-expressing neurons. Taken together, our findings unravel a link between XBP1s signaling and seizure occurrence.

Fig. 1. Sustained Xbp1s expression in neurons results in spontaneous seizures and death.

a Survival of adult XBP1s-TG mice after induction of Xbp1s expression. XBP1s-TG and control mice were maintained with Dox in the drinking water prior to induction. When the mice were 2–3 months old, Dox was removed. The survival rate was recorded over 4 weeks. n = 17–18/group. The highlighted area labeled “seizure” refers to the presence of seizure behaviors. Inset: a representative picture of XBP1s-TG mice showing a seizure-like phenotype. b–d Progression of seizures with EEG monitoring. XBP1s-TG and control littermates were subjected to continuous EEG monitoring, immediately after removing Dox from the drinking water. b Seizure occurrence and survival rate. c Examples of EEG-recorded activity and seizures. The behavioral seizure activity is color-coded with blue boxes representing freezing, yellow boxes denoting rearing with automatisms of the forelimbs, and red boxes representing bounding behaviors. The period in the green box is a post-ictal period showing little to no movement. All the XBP1s-TG EEG recordings were from the same mouse. Of note, the two EEG traces at the bottom depict that the profoundly abnormal low amplitude EEG that was seen in several mice just hours before death was completely reversed to an essentially normal EEG pattern after adding Dox back to drinking water. d EEG characteristics of seizures. In the first 6 hours after seizure onset (early), the number, duration, and burden (time spent in seizures per hour) of seizures were lower during than during the late stage (i.e., the 6 hours just prior to transitioning to continuous seizure activity). Data are presented as mean ± SEM (n = 7/group). *p < 0.05; ***p < 0.001.

Fig. 2. Pharmacologic interventions.

On day 10 post-removal of Dox from the drinking water, XBP1s-TG mice (n = 4–7/group) received Dox in drinking water, valproic acid (VPA), or carbamazepine (CBZ) treatment. XBP1s-TG mice without any treatment served as control. a Survival rates (****p < 0.0001; Log-rank [Mantel-Cox] test). b Seizure scores (mean ± SEM).

Fig. 3. Integration site identification of the transgene in TRE-XBP1s mouse line.

a Genome-wide TLA coverage. A detailed view of TLA sequence coverage surrounding the integration site (indicated by a red rectangle at chromosome 14) is shown in the inset. Peak at chromosome 11 shows endogenous Xbp1 (indicated by a black circle). b Schematic of the insertion site. The transgene (red) was into the intron region between Gpc6 exon 2 and exon 3. The blue arrows denote the primes for confirmation of the left and right junctions of the transgene. c PCR confirmation of the integration site. 1 and 2: TRE-XBP1s mouse DNA, and 3: wild-type mouse DNA. TLA targeted locus amplification.

Fig. 4. Behavioral phenotype in XBP1s-TG mice.

a Body weight loss. Body weight of adult male (n = 6/group; RMANOVA: genotype effect [F(1,9) = 12.67, p = 0.006]) and female (n = 11–13/group; RMANOVA: genotype effect [F(1,12) = 12.56, p = 0.004]) mice were monitored after Dox removal. b Grip strength test. Mice were tested on the indicated days post-removal of Dox. Data are presented as mean ± SEM (n = 5/group). c Social dominance tube test (n = 5/group). Control and XBP1s-TG mice were kept on drinking water without Dox for 8 days. The latency time (in seconds) of XBP1s-TG mice are shown in the table. If XBP1s-TG mouse won, the latency time is positive (red). If XBP1s-TG mouse lost, the latency time is negative (blue-grey). *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 5. RNA-Seq analysis of hippocampal samples from control and XBP1s-TG mice.

a Verification of RNA-Seq data by quantitative RT-PCR analysis. All data were normalized to β-actin. To calculate fold change, the mean values of control samples were set to 1.0. Data are presented as means ± SEM (n = 3/group). b–d Effects of XBP1s on UPR signaling. Western blotting analysis of key components of the UPR was performed using hippocampal samples. Brain samples were collected on the indicated days (b) or day 8 (c) after Dox removal from the drinking water. d Luciferase reporter analysis. Various promoter reporter vectors with empty vector or the Xbp1s expression vector were co-transfected into HEK293T cells. Luciferase activity was measured and compared. e IPA pathway analysis. f The predicted network. XBP1 is shown in red indicating high overexpression; brown indicates upregulated genes, while green signifies down-regulated genes. g RNA-Seq data of all subunits of the GABA receptors. The expression levels for the Gabra4 and Gabrd subunits are marked by red dotted rectangles. Data are presented as mean ± SEM. *p < 0.01; **p < 0.01; ***p < 0.001.

Fig. 6. GABA_A receptor-mediated IPSCs and tonic currents.

a Attenuated GABAergic inhibitory synaptic activity in neurons expressing Xbp1s. Representative bicuculline-sensitive GABA_A-mediated sIPSCs traces recorded from control and XBP1s-TG neurons are shown in the top panels. Bar graphs for mean IPSCs frequency, mean IPSCs amplitude, and mean area (charger) under the sIPSCs, are shown in the bottom panels (Control: n = 10 neurons from 3 mice; XBP1s-TG: n = 13 neurons from 3 mice). b Reduction in extrasynaptic GABA_A receptor-mediated tonic currents in XBP1s-TG mice. Representative tonic current traces are shown in the 2 left panels. Bar graphs for tonic current amplitude and density between control (n = 13 neurons from 5 mice) and XBP1s-TG mice (n = 14 neurons from 5 mice) are shown in the 2 right panels. Data are presented as mean ± SEM. *p < 0.05. sIPSCs spontaneous inhibitory post-synaptic currents. BMI bicuculline methiodide.