Changes in Calcium Homeostasis and Gene Expression Implicated in Epilepsy in Hippocampi of Mice Overexpressing ORAI1

Abstract

Previously, we showed that the overexpression of ORAI1 calcium channel in neurons of murine brain led to spontaneous occurrence of seizure-like events in aged animals of transgenic line FVB/NJ-Tg(ORAI1)Ibd (Nencki Institute of Experimental Biology). We aimed to identify the mechanism that is responsible for this phenomenon. Using a modified Ca²⁺-addback assay in the CA1 region of acute hippocampal slices and FURA-2 acetomethyl ester (AM) Ca²⁺ indicator, we found that overexpression of ORAI1 in neurons led to altered Ca²⁺ response. Next, by RNA sequencing (RNAseq) we identified a set of genes, whose expression was changed in our transgenic animals. These data were validated using customized real-time PCR assays and digital droplet PCR (ddPCR) ddPCR. Using real-time PCR, up-regulation of hairy and enhancer of split-5 (Hes-5) gene and down-regulation of aristaless related homeobox (Arx), doublecortin-like kinase 1 (Dclk1), and cyclin-dependent kinase-like 5 (Cdkl5, also known as serine/threonine kinase 9 (Stk9)) genes were found. Digital droplet PCR (ddPCR) analysis revealed down-regulation of Arx. In humans, ARX, DCLK1, and CDLK5 were shown to be mutated in some rare epilepsy-associated disorders. We conclude that the occurrence of seizure-like events in aged mice overexpressing ORAI1 might be due to the down-regulation of Arx, and possibly of Cdkl5 and Dclk1 genes.

Keywords: Arx; Cdkl5; Dclk1; ORAI1; RNAseq; RT-PCR; STIM1; STIM2; epileptic seizures; nSOCE.

Figure 1

Changes in Ca²⁺ responses in the hippocampal neurons overexpressing ORAI1 calcium channel (A–D) or stromal interaction molecule 2 (STIM2) (E–G) compared with wild-type neurons. Ca²⁺ measurements were performed using Fura-2 acetomethyl ester (Fura-2 AM) indicator that was loaded into the CA1 pyramidal neurons of the hippocampal acute brain slices (H). Typically, ~20 pyramidal neurons per one slice (n) were analyzed; the slices were isolated from at least 5 animals per genetic variant. The total number of slices analyzed per each genetic variant was 9. (A,E) Averaged time-course of background-subtracted fluorescence signal (expressed as F340/F380) from slices overexpressing ORAI1 and STIM2, respectively. (B,F) Quantification of signal amplitudes observed following glutamate, cyclopiazonic acid (CPA), and Ca²⁺ (re)addition in ORAI1 and STIM2 overexpressing neurons, respectively, arbitrary units (a.u.). (C,G) Quantification of F340/F380 values at baseline (average of 0–5th min of the measurement) and following application of ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA) (at 13th and 25th min of the measurement) in ORAI1 and STIM2 overexpressing neurons, respectively, a.u. (D) Time-course of signal decay following glutamate application in ORAI1 and wild-type neurons that was fit by a logarithmic function. Student’s t-test was used to check statistical significance of the observed differences; p-values are displayed above the respective charts.

Figure 2

RNA sequencing (RNAseq) analysis using Noiseq tools based on data gathered from hippocampi of wild-type and FVB/NJ-Tg(ORAI1)Ibd females. (A) Dot plot of Gene Ontology Biological Process groups (GO BP), which are down-regulated and up-regulated in transgenic line (on the left and right, respectively). (B) Volcano plot represents differential gene expression between the tested variants. (C) Principal component analysis (PCA) of principal component 1 (PC1) and 2 (PC2). Transgenic and wild-type mice are color-coded in green and red, respectively. PCA analysis was carried out from FPKM (Fragments Per Kilobase Million) normalized RNAseq data using prcomp function in R software.

Figure 3

Expression profile of selected genes in hippocampi of the tested mouse variants. (A) Rea-time PCR analysis of the selected genes that was based on RNAseq data. The results are presented as scatter plot, each circle corresponds to one animal. Black circle corresponds to each tested wild-type probe, whereas green corresponds to transgenic ones. The obtained results suggest a down-regulation of expression of Arx, Cdkl5, and Dclk1. The data are presented as a fold change (2^−∆∆Ct) normalized to the three reference genes (Uba-2, Gapdh, Actin). The statistics was estimated by CFX software (Bio-Rad) using the post hoc method (Tukey’s test). (B) The ddPCR analysis of Arx, Cdkl5, Dclk1 using customized assays (Bio-Rad). The number of transcript copies of the selected genes are presented in relation to 1000 copies of the reference gene Uba-2. The results are presented as dot plot, where each sign corresponds to an individual animal (3 wild-type, 6 transgenic). The statistics were calculated by GraphPad Prism software using unpaired t-test. The number of Arx transcripts was about 30 times lower than that of Cdkl5 and Dclk1, therefore, it is presented on a separate chart.

Figure 4

Expression profile of the selected genes in the hippocampi of the tested mouse variants that was obtained by real-time PCR analysis. The results are presented as scatter plot; each circle corresponds to one animal. The obtained results suggest an up-regulation of the expression of Hes-5 and Strap. The data are presented as a fold change (2^−∆∆Ct) that was normalized to the reference gene (Uba-2), the statistics was estimated by CFX software (Bio-Rad) using post hoc method (Tukey’s test).