Deletion of Long Isoform of Eukaryotic Elongation Factor 1Bδ Leads to Audiogenic Seizures and Aversive Stimulus-Induced Long-Lasting Activity Suppression in Mice

Abstract

Alternative splicing enables a gene to give rise to diverse protein products. The Eef1d gene produces two isoforms: a short isoform that encodes translation elongation factor 1Bδ (eEF1Bδ1), and a long isoform that encodes the heat shock-responsive transcription factor eEF1BδL. Previously, we found that eEF1BδL was a splice variant that was specific to the brain and testis, and the protein encoded is thought to have a function in the central nervous system. In this study, we generated knockout (KO) mice of C57BL/6J background that selectively lacked a specific exon in Eef1d for the long isoform. These KO mice lacked eEF1BδL, but not eEF1Bδ1, in the brain. Although the KO mice showed normal anxiety-related and learning behavior in behavioral tests, some showed severe seizures in response to loud sounds (90 dBA), an audiogenic seizures (AGS) response. Furthermore, after the KO mice had been subjected to the fear conditioning test, they showed remarkably decreased locomotor activity in their home cage and in the open-field and elevated plus-maze tests. After the fear conditioning test, a significant decrease in brain weight, atrophy of the hippocampus and midbrain, and reduced cortical layer thickness were observed in the KO mice. We also found a compensatory increase in the eEF1Bδ1 level and elevated protein synthesis with the induction of endoplasmic reticulum stress markers in these mice. Our results suggest that eEF1BδL has an important role in normal brain function especially when exposed to external stimuli.

Keywords: alternative splicing; audiogenic seizure; elongation factor; fear conditioning; protein synthesis.

Figure 1

Generation of eEF1BδL-KO mice. (A) Schematic diagram of the endogenous Eef1d locus and the targeting vector used to engineer embryonic stem (ES) cells by homologous recombination. The lengths of the genomic fragments indicated are the lengths when digested by EcoRV. The 3′ probe was used for Southern blotting to confirm homologous recombination. DT-A, diphtheria toxin-A; neo, neomycin-resistant gene. (B) Southern blot analysis of genomic DNA from F1 mice using the 3′ probe after digestion with EcoRV. (C) PCR genotyping of genomic DNA from wild-type (WT), heterozygous (HZ) and knockout (KO) mice using primers crossing loxP. (D) Western blot analysis of brain lysates from WT and KO mice. Actin was used as a loading control. (E) Body weight of WT and KO male mice at the age of 6, 12, and 18 weeks. n = 6 for each. (F) Brain weight of WT and KO male mice at the age of 5–13 weeks. n = 6 (WT mice) or 8 (KO mice). (G) Hematoxylin and eosin (H&E) staining of brain sections from WT and KO male mice sacrificed at the age of 16 weeks. Left panel: coronal section. Right panel: cortical layer. Scale bar: 200 μm.

Figure 2

qPCR analysis of the gene expression of heat shock proteins in the brain of mice exposed to heat. WT and KO mice were exposed to a hot environment (38°C) for 4 h. The brain was then dissected, and the total RNA was extracted and subjected to qPCR analysis. ^*p < 0.05, ^**p < 0.01, n = 3 or 4, Student's t-test. (+) denotes significant differences after adjustment by false discovery rate (FDR) analysis.

Figure 3

Behavioral analysis of eEF1BδL-KO mice before the fear conditioning test. (A) A timeline of the behavioral tests which were performed on WT and KO mice. 0 week corresponds to the beginning of open-field test. OF, open-field; EP, elevated plus; YM, Y-maze; FC, fear conditioning; TM, treadmill; HC, home cage. (B) Open-field test on WT and KO male mice at the age of 8–30 weeks. The total distance traveled and time spent in the center area were measured for 60 min, and the data collected at every 5 min are displayed. n = 14 for each. (C) Elevated plus-maze test on WT and KO male mice at the age of 8–30 weeks. The total distance traveled, number of total entries, percent of entries into the open arms, and time spent in the open arms were measured. n = 14 for each. (D) Y-maze test on WT and KO male mice at the age of 17–38 weeks. The number of total entries and percent of alternations were measured. n = 14 for each. (E) Fear conditioning test on WT and KO male mice at the age of 18–44 weeks. The percent of time spent freezing was measured at conditioning, context and cued testing. ^*p < 0.05, n = 14 for each, two-way repeated measures ANOVA.

Figure 4

The levels of elongation factor family proteins in eEF1BδL-KO mice. (A) Western blot analysis of the long and short isoforms of eEF1Bδ in several brain regions from WT and KO male mice sacrificed at the age of 5 weeks. Actin was used as a loading control. (B) Western blot analysis of eEF1A, eEF1Bα, eEF1Bδ, and eEF1BδL in the hippocampal tissues from WT and KO male mice sacrificed at the age of 4–8 weeks. Actin was used as a loading control. Left panel: representative blot image. Actin was used as a loading control. Right panel: graph showing the quantitative data. ^*p < 0.05, ^**p < 0.01, n = 4 for each, Student's t-test. (+) denotes significant differences after adjustment by FDR analysis.

Figure 5

Behavioral analysis of eEF1BδL-KO mice after the fear conditioning test. (A) Spontaneous locomotor activity in the home cage 6 weeks after the fear conditioning test. Locomotor activity was measured during the dark and light phases. ^*p < 0.05, n = 6 for each, Student's t-test. (+) denotes significant differences after adjustment by FDR analysis. (B) Open-field test on WT and KO male mice 11 weeks after fear conditioning. The total distance traveled and time spent in the center area were measured for 60 min, and the data obtained at 5-min intervals are displayed. ^*p < 0.05, n = 13 (WT mice) or 11 (KO mice), two-way repeated measures ANOVA. (+) denotes significant differences after adjustment by FDR analysis. (C) Elevated plus-maze test on WT and KO male mice 3 weeks after fear conditioning. The total distance traveled, number of total entries, percent of entries into the open arms, and time spent in the open arms were measured. ^**p < 0.01, n = 14 for each, Student's t-test. (+) denotes significant differences after adjustment by FDR analysis. (D) Treadmill test on WT and KO male mice 2 weeks after fear conditioning. The number of electric shock that the mice received was counted. ^*p < 0.05, n = 14 for each, Student's t-test. (+) denotes significant differences after adjustment by FDR analysis.

Figure 6

Brain atrophy in eEF1BδL-KO mice. (A) The brain weight and body weight of WT and KO male mice sacrificed at the age of 8–13 weeks. Mice were subjected to the fear conditioning test 2 weeks before they were sacrificed. Unexperienced, mice with no experience of fear conditioning test; Experienced, mice with experience of fear conditioning test. ^*p < 0.05, ^**p < 0.01, n = 9 (WT mice) or 9 (KO mice), Student's t-test. (+) denotes significant differences after adjustment by FDR analysis. (B) Hematoxylin and eosin (H&E) staining of the brain sections from WT and KO male mice sacrificed at the age of 12 weeks. Mice were subjected to the fear conditioning test 2 weeks before they were sacrificed. Left panel: coronal section. Right panel: cortical layer. Scale bar: 200 μm.

Figure 7

Protein synthesis and ER stress markers in eEF1BδL-KO mice. (A) The rate of protein translation was measured in cultured hippocampal neurons from WT and KO mice. Left panel: representative autoradiography of ³⁵S-methionine/cysteine labeling and a graph showing the quantitative data. ^**p < 0.01, n = 3 for each, Student's t-test. Right panel: representative image of a blot with puromycin labeling and a graph showing the quantitative data. ^**p < 0.01, n = 3 for each, Student's t-test. Actin was used as a loading control. (+) denotes significant differences after adjustment by FDR analysis. (B) qPCR was performed for ER stress markers using total RNA isolated from the hippocampal tissues of WT and KO mice. ^**p < 0.01, n = 4 (WT mice) or 5 (KO mice), Student's t-test. (+) denotes significant differences after adjustment by FDR analysis. (C) Western blot analysis for ATF4 in the hippocampal tissues from WT and KO mice. Actin was used as a loading control. (D) HEK293 cells were transfected with empty vector or a plasmid expressing eEF1BδL. Twenty-four hours after transfection, the rate of protein translation was measured by ³⁵S-methionine/cysteine labeling, followed by liquid scintillation counting. ^*p < 0.05, n = 5 for each, Student's t-test. (+) denotes significant differences after adjustment by FDR analysis. (E) Western blot analysis for eEF1BδL in HEK293 cells transfected with empty vector or a plasmid expressing eEF1BδL. Actin was used as a loading control.