Hippocampal glucocorticoid receptors modulate status epilepticus severity

Abstract

Status epilepticus (SE) is a life-threatening medical emergency with significant morbidity and mortality. SE is associated with a robust and sustained increase in serum glucocorticoids, reaching concentrations sufficient to activate the dense population of glucocorticoid receptors (GRs) expressed among hippocampal excitatory neurons. Glucocorticoid exposure can increase hippocampal neuron excitability; however, whether activation of hippocampal GRs during SE exacerbates seizure severity remains unknown. To test this, a viral strategy was used to delete GRs from a subset of hippocampal excitatory neurons in adult male and female mice, producing hippocampal GR knockdown mice. Two weeks after GR knockdown, mice were challenged with the convulsant drug pilocarpine to induce SE. GR knockdown had opposing effects on early vs late seizure behaviors, with sex influencing responses. For both male and female mice, the onset of mild behavioral seizures was accelerated by GR knockdown. In contrast, GR knockdown delayed the onset of more severe convulsive seizures and death in male mice. Concordantly, GR knockdown also blunted the SE-induced rise in serum corticosterone in male mice. GR knockdown did not alter survival times or serum corticosterone in females. To assess whether loss of GR affected susceptibility to SE-induced cell death, within-animal analyses were conducted comparing local GR knockdown rates to local cell loss. GR knockdown did not affect the degree of localized neuronal loss, suggesting cell-intrinsic GR signaling neither protects nor sensitizes neurons to acute SE-induced death. Overall, the findings reveal that hippocampal GRs exert an anti-convulsant role in both males and females in the early stages of SE, followed by a switch to a pro-convulsive role for males only. Findings reveal an unexpected complexity in the interaction between hippocampal GR activation and the progression of SE.

Keywords: Dentate gyrus; Epilepsy; Fluoro-Jade B; Hypothalamic-pituitary adrenal axis; Pilocarpine; Seizure threshold; Sex differences.

Fig. 1.

Representative micrographs of mCherry labeling (red) and GR immunostaining (green) among CA1 pyramidal cells (A) and dentate granule cells (B) in wild-type and hGRKD mice. The merged images include a nuclear blue fluorescent counterstain. Outlines in the middle (GR) panels show mCherry+ regions in wild-type mice with intact GRs (white lines), and mCherry+ regions in hGRKDs devoid of GR immunofluorescence (red lines). Scale bar = 25 μm. C, D: Mean fluorescence intensity in arbitrary units (a.u.) within the GR channel of the somas of mCherry+ cells from wild-type and hGRKD mice. Fluorescence intensity was significantly higher among mCherry+ CA1 pyramidal cells (C) and dentate granule cells (D) from wild-type mice relative to cells from hGRKD mice. E: The percentage of mCherry+ cells meeting criteria as GR-immunopositive (2× background) was significantly reduced in hGRKD mice relative to wild-type mice. Dots = individual cells in C and D, and individual mice in E. ****p < 0.0001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2.

A: Adult male and female mice received hippocampal viral injections to yield hGRKD and wild-type (WT) groups. One week after viral injection, mice began daily acclimation to the researcher and testing conditions. The day before pilocarpine injection, blood was collected for baseline serum CORT measures (T₀). Two weeks after viral injection, mice underwent pilocarpine-induced SE. Blood was sampled either at the time of death T_d) or 75 min (T₇₅) after pilocarpine treatment. B: Baseline morning CORT concentrations (T₀) were equivalent between hGRKD and wild-type mice, with no effect of sex. C: Graph depicts paired baseline (B) and T₀/T₇₅ (SE) CORT measurements for each animal. Only animals that entered SE and had both baseline and SE measurements are shown. ^###p < 0.001 for main effect of baseline vs SE. D: GR knockdown blunted the T_d/T₇₅ SE-induced increase in CORT in male, but not female mice. **p < 0.01, ***p < 0.001. Portions of this figure were generated using Biorender.

Fig. 3.

A-C: Progression of behavioral seizures through each seizure class (A) is shown for wild-type (B) and hGRKD mice (C). In the heat maps, each row gives the data for a single animal. D-K: Analysis of the data with males and females combined revealed that GR knockdown accelerated the onset of class 3 seizures (D), but delayed the onset of class 4 seizures (F), class 5 seizures (H) and death (J). To visualize how animal sex drove these findings, data are also presented with males and females separated (E,G,I,K). Both males and females contributed to the accelerated onset of class 3 seizures in hGRKDs (E). Increased latency to class 4 seizures, class 5 seizures and death, however, was only evident in hGRKD males. Restated, control males developed severe outcomes more rapidly, and GR knockdown in males shifted their curves towards the more resistant female pattern (G,I,K). *, p < 0.05. **p < 0.01, ***p < 0.001. ****p < 0.0001. Asterisks in B,D,F and H denote comparisons between hGRKD and sex-matched wild-type groups.

Fig. 4.

A: Representative images of viral spread along the dorso-ventral axis of the hippocampus, as revealed by mCherry+ expression among dentate granule cells (DGCs), mossy cells (MCs), pyramidal cells of CA1, CA2, CA3 and ventral subiculum. Among animals, labeling was most expansive and consistent in CA1 and dentate gyrus. Scale bar = 1 mm. B, C: The percentage of mCherry+ CA1 pyramidal cells and dentate granule cells was quantified for each wild-type (black) and hGRKD (red) animal in the study. Ventral regions (3–4 mm posterior to bregma) exhibited the highest viral load for both CA1 (B) and dentate gyrus (C). The pattern of viral expression was similar between wild-type and hGRKD groups. D: Stereological data on mCherry expression in the dentate and CA1 at each bregma level was averaged to generate a single viral load score for each mouse. Each dot represents the score for a single animal. Portions of this figure were generated using Biorender. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5.

A: Cell loss was assessed by FluoroJade B (FJB) staining in the CA1 pyramidal cell layer and dentate hilus of wild-type (n = 4) and hGRKD (n = 5) mice that survived for 48 h after SE. Scale bar = 25 μm. B: Stereological cell counts produced similar estimates for the number of FJB+ dying cells in the CA1 pyramidal cell layer of wild-type and hGRKD mice. C: Greater variability in the degree of cell loss was evident in the dentate hilus of hGRKDs, however, similar to CA1, cell loss was statistically similar between hGRKDs and wild-type groups. D,E,F,G: Graphs show combined data for mCherry-expression in CA1 (orange, left y-axis) and dentate (red, left y-axis) relative to FJB+ cell densities (blue, right y-axis) along the dorso-ventral extent of the hippocampus (x-axis). Each point is the group average for each measure at that bregma level ± SEM. The smaller cohort of survivors parallels the larger group, with greater mCherry expression in ventral hippocampus. No clear relationship between hGRKD rates and cell loss, however, was evident. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)