Decreased expression of Rev-Erbα in the epileptic foci of temporal lobe epilepsy and activation of Rev-Erbα have anti-inflammatory and neuroprotective effects in the pilocarpine model

Adult male C57 BL/6 mice (21–24 g) were purchased from the Daping Hospital Animal Center of the Army Medical University. All mice were housed under climate-controlled conditions on a 12-h light/dark cycle, which was divided into 24 h Zeitgeber time (ZT) units, with the light being turned on at ZT 0 (7 a.m.) and the light being turned off at ZT 12 (7 p.m.). Food and water were available ad libitum. All animal experimental procedures were reviewed and approved by the Internal Animal Care and Use Committee of the Army Medical University. Appropriate measures were taken during experimental procedures to minimize animal suffering and reduce the number of animals used.

Mice were injected with pilocarpine (280–300 mg/kg, i.p., Sigma-Aldrich, USA) to induce SE [1]. Scopolamine methyl nitrate (5 mg/kg, i.p., Sigma-Aldrich, USA) was injected 30 min prior to pilocarpine administration in order to limit the peripheral muscarinic effects [21]. Seizure severity was rated using the Racine scale [22]. Categories 1 and 2 include facial automatisms, tail stiffening, and wet-dog shakes; category 3, the low-intensity tonic-clonic seizure marked by unilateral forelimb myoclonus in addition to the symptoms above; category 4, the addition of bilateral forelimb myoclonus and rearing; and category 5, bilateral fore- and hindlimb myoclonus and transient loss of postural control. Only those mice that attained categories 4–5 were used. Diazepam (10 mg/kg, i.p.) was administered 2 h after the onset of SE to suppress convulsions. The post-SE supportive care of the animals included body temperature maintenance and rehydration therapy to increase the number of survivors. Injections (i.p.) of sterile 0.9% saline were administered for replenish fluids. Sliced peeled apples were placed in all mouse cages in addition to rodent chow. Spontaneous recurrent seizures (SRSs) were found in mice by monitoring after SE, which confirmed the successful establishment of chronic epilepsy model. Age- and weight-matched control mice received the same treatment with scopolamine methyl nitrate and diazepam, and we used saline instead of pilocarpine.

In experiment 2, we explored the effects of SR9009 on inflammatory reactions, glial proliferation, neuronal apoptosis, and neuronal loss in the hippocampus 7 days after pilocarpine induced-SE (Fig. 1c). In total, 44 mice were distributed into the following four groups: the vehicle-treated control group, the SR9009-treated control group, the vehicle-treated pilocarpine group, and the SR9009-treated pilocarpine group. SR9009 is a specific brain-permeable small molecule agonist of Rev-Erbα [23–25]. SR9009 (MedChem Express, USA) was dissolved in DMSO at a final concentration of 0.1% DMSO and administered intraperitoneally at a dose of 100 mg/kg twice daily (7 a.m. and 7 p.m.) for 7 consecutive days [15, 16]. The first injection was 1 h after SE. It should be noted that SR9009 had no effect on the expression levels of Rev-Erbα in the control and SE model animals (data not shown).

Tissues from patients were immediately collected at the time of surgery. One set of tissue was immediately frozen in liquid nitrogen and stored at − 80 °C for western blotting. The other samples were fixed in 4% phosphate-buffered paraformaldehyde for 48 h, and the tissues were then embedded in paraffin, sectioned at a thickness of 6 μm for immunohistochemistry and 10 μm for immunofluorescence staining. The bilateral temporal neocortex and hippocampus were dissected from mice for protein extraction and western blotting. For immunofluorescence and TUNEL staining, the mouse brains were obtained after systemic saline and 4% phosphate-buffered paraformaldehyde infusion and then further processed as described above for human tissues.

A whole protein extraction kit (Beyotime, Jiangsu, China) was used to extract total protein from tissues. Total tissue lysates were centrifuged at 14,000 rpm for 15 min at 4 °C, and the protein concentration in the supernatant was determined using the bicinchoninic acid protein assay (Bio-Rad, Hercules, CA, USA). Equal amounts of protein (60 μg per lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (5% spacer gel, 80 V, 20–30 min; 8–10% separating gel, 120 V, 50–60 min) and electrophoretically transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Temecula, CA, USA) using a semidry electroblotting system (Transblot SD; Bio-Rad) (300 mA, 80 min). Then, blotting membranes were incubated at room temperature for 4 h in 5% nonfat dry milk to block nonspecific binding. Next, the membranes were incubated overnight at 4 °C with primary antibodies: rabbit anti-Rev-Erbα (1:1000, Abcam, UK), rabbit anti-NLRP3 (1:1000, Abcam, UK), rabbit anti-ASC (1:1000, Immunoway, USA), rabbit anti-Caspase1 (1:200, Boster, Wuhan, China), rabbit anti-IL-1β (1:1000, Abcam, UK), rabbit anti-IL-18 (1:1500, Invitrogen, USA), rabbit anti-IL-6 (1:1000, Abcam, UK), rabbit anti-TNF-α (1:1000, Abcam, UK), mouse anti-GFAP (1:500, Sigma, USA), rabbit anti-Iba1 (1:800, Wako), rabbit anti-cleaved Caspase3 (1:500, Cell Signaling Technology, Boston, MA), and rabbit anti-GAPDH (1:12000, Abcam, UK). After four rinses in TBST (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.5% Tween-20), the samples were treated with a horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse secondary antibody (1:1000, Zhongshan Golden Bridge Biotechnology, China) for 1 h in a 37 °C incubator. The immunoreactive bands were visualized using enhanced chemiluminescence and were scanned and analyzed with Quantity One software (Bio-Rad Laboratories, Hercules, CA, USA). The optical densities (ODs) of each protein band were calculated relative to the OD of the reference protein, GAPDH.

Formalin-fixed, paraffin-embedded sections (6 μm thick) were mounted on polylysine-coated slides. Paraffin sections were dewaxed in xylene and rehydrated in a series (100%, 90%, 75%, 50%) of ethanol. Sections were incubated for 30 min in 0.3% hydrogen peroxide in methyl alcohol to inactivate endogenous peroxidase. Antigen retrieval was performed by boiling in a microwave for 10 min in citrate buffer (pH 6.0). The sections were then blocked in 5% bovine serum albumin and 0.3% Triton X-100 for 1 h at room temperature. After the removal of excess serum, the sections were incubated with primary antibody (rabbit anti-Rev-Erbα, 1:100, Abcam, UK) overnight at 4 °C. After washing, the sections were incubated with a HRP-conjugated goat anti-rabbit IgG secondary antibody (Boster, China) for 1 h at 37 °C. Then, the color was developed using 3,3′-diaminobenzidine (DAB; Boster, China) substrate. Finally, the sections were dehydrated, diaphanized, and covered with coverslips. No immunoreactive cells were observed in the negative control experiments, which involved the omission of the primary antibody, preabsorption with a tenfold excess of specific blocking antigen, or incubation with an isotype-matched rabbit polyclonal antibody. Images of Rev-Erbα immunostaining for each clinical surgical specimen section were captured with a × 40 objective (Olympus microscope digital camera system, DP80, Tokyo, Japan). The camera aperture and power and the exposure time of all images were constant. Five nonoverlapping gray matter visual fields and five nonoverlapping white matter visual fields from each section were randomly chosen for further semiquantitative analysis of Rev-Erbα expression (Image-Pro Plus 6.0 software, Media Cybermetrics Inc., USA). For each field, areas of interest were selected to measure the integrated optical density (OD) and area. Moreover, each integrated OD/area value was calculated by subtracting the background integrated OD/area value from the directly measured integrated OD/area value.

For double immunofluorescence staining, paraffin sections from patients were incubated at 4 °C overnight with a rabbit anti-Rev-Erbα primary antibody (1:100, Abcam, UK) combined with mouse anti-NeuN (1:200, Millipore, USA), mouse anti-GFAP (1:500, Sigma, USA), and mouse anti-Iba1 (1:300, Wako) primary antibodies. After rinsing in Tris-buffered saline (TBS) (4 times × 10 min), the sections were incubated with a mixture of secondary antibodies (Cy3-conjugated donkey anti-mouse, 1:400; 488-conjugated donkey anti-rabbit, 1:400; Jackson ImmunoResearch, West Grove, PA, USA) for 1 h at 37 °C. For fluorescence label staining for NeuN, GFAP, and Iba1 in mouse model sections, the three abovementioned primary antibodies and a Cy3-conjugated donkey anti-mouse secondary antibody (1:400, Jackson ImmunoResearch, West Grove, PA, USA) were used. The nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Boster, China). The fluorescence signals were acquired using a confocal fluorescence microscope (TCS-SP5, Leica, Nussloch, Germany).

Based on previous reports [26, 27], we evaluated the density of neurons, astrocytes, and microglia in the CA1, CA3, and dentate gyrus (DG) regions of the hippocampus using a modified method. Coronal sections of the brain from similar anatomical positions (Fig. 1a; three coordinate positions: 1.46 mm, 1.70 mm, and 1.94 mm posterior to bregma; three consecutive sections at each position) were collected from each mouse. Three areas from the CA1, CA3, or DG region of each section were randomly selected, three sections from each coordinate position form each mouse, and three mice from each group were measured, that is, 27 areas of the CA1, CA3, or DG region were analyzed for each group. The number of NeuN⁺ neurons, GFAP⁺ astrocyte cell soma, and Iba1⁺ microglia cell soma were counted using ImageJ software (version 1.42 V, NIH, USA). These analyses were conducted in a blinded fashion. The results are expressed as the mean ± SD number of cells per square millimeter.

Pretreated of 6-μm thick paraffin-embedded sections were processed as described for the immunohistochemistry procedure. Apoptosis was determined with the TUNEL kit (Boster, China) according to the manufacturer’s protocol. Section positions were selected as described above (Fig. 1a). Images of TUNEL staining were captured using an Olympus microscope (DP80, Olympus, Tokyo, Japan) with a × 40 objective. Image-Pro Plus 6.0 software was used to count TUNEL-positive cells in each visual field. Three microscopic fields of the CA1 or CA3 region from each section, three sections from each mouse, and three mice from each group were used. All of the mouse sections used for the analysis were viewed in a blinded manner. We averaged the results for each group.

GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA) was used for statistical analyses. The analysis between the TLE patients and the controls was carried out using Student’s t test. Two-way ANOVA followed by post hoc Bonferroni’s test was used to compare the groups (control and experimental) throughout the 24-h period by western blotting analysis. One-way ANOVA analysis followed by post hoc Bonferroni’s test was used to determine the differences among multiple groups. The data are expressed as the mean ± SD. Significance was accepted at p < 0.05.

An examination of sex and age revealed no difference between the subjects in the control and TLE groups. The control group contained 5 female and 5 male individuals with a mean age of 24.00 ± 1.680 years (range, 17–33 years). The mean age of the intractable TLE patients was 27.09 ± 1.475 years (range, 15–41 years), with 10 males and 12 females.

Next, we analyzed the expression patterns of Rev-Erbα using IHC and immunofluorescence labeling staining. In controls (n = 10), Rev-Erbα displayed moderate to strong IR in neurons within the gray matter (GM; Fig. 2c) and glial cells within the white matter (WM; Fig. 2e). However, in the TLE patient group, Rev-Erbα showed weak IR in corresponding regions (Fig. 2d, f; n = 22). The optical density (OD) analysis revealed that the mean OD of Rev-Erbα in the GM and WM was significantly reduced in the TLE group compared with the control group. (Fig. 2g, h; p < 0.001). Subsequently, double-labeled immunofluorescence experiments (Fig. 2i) demonstrated that Rev-Erbα was colocalized with the neuronal marker NeuN and the astrocyte marker GFAP. In addition, Rev-Erbα IR was detectable in Iba1-positive microglia.

AEDs may have an effect on human Rev-Erbα expression. Additionally, Rev-Erbα is a circadian regulator that may present 24-h cyclic patterns in some brain regions [28, 29], and although each operation was performed between 10 a.m. and 1 p.m., the timing of the surgical removal of the temporal neocortex from epileptic patients and control patients may not have been entirely consistent. Thus, we assessed the temporal changes in Rev-Erbα protein levels in the hippocampus and temporal neocortex of control mice and mice subjected to SE (early post-SE and the chronic phases), which were sacrificed every 2 h over a 24 h timescale.

Then, we detected the expression of cleaved-caspase3 (17 kDa) in the hippocampus to determine the apoptotic status of the four groups (Fig. 8d). The level of cleaved caspase3 was increased in the hippocampus of the SR9009-treated pilocarpine group and the vehicle-treated pilocarpine group (p < 0.001). Interestingly, a significant decrease in cleaved caspase3 levels was observed in the SR9009-treated pilocarpine group compared to the vehicle-treated pilocarpine group (p < 0.001).

To determine the effect of SR9009 on the density of neurons in the hippocampal CA1 and CA3 regions and dentate gyrus (DG), immunofluorescence labeling staining for NeuN was performed on day 7 post-SE (Fig. 8e–h). No differences in the number of NeuN-positive cells were detected in the hippocampal CA1 region, hippocampal CA3 region or DG between the vehicle-treated control and SR9009-treated control groups (p > 0.05). We observed that the number of NeuN-positive cells was decreased in the CA1 and CA3 regions of the hippocampus in the vehicle-treated pilocarpine group and the SR9009-treated pilocarpine group compared to that in the two control groups (p < 0.05 or p < 0.001). Notably, the number of NeuN-positive cells in the SR9009-treated pilocarpine group was significantly higher than the vehicle-treated pilocarpine group (CA1, p < 0.05; CA3, p < 0.001). SR9009 did not affect the number of NeuN-positive cells in the DG after SE (p > 0.05). Collectively, these data demonstrate that SR9009 exerts neuroprotection in the hippocampus in the pilocarpine-induced SE model.

Clinical observations and data from animal studies have proposed that neuroinflammation is a common factor contributing to, or predisposing one to, the occurrence of seizures in various kinds of epilepsy, especially TLE [34, 35]. For instance, some inflammatory cytokines (IL-1β, IL-6, TNF-α, etc.) synthesized and released by astrocytes and microglia have been reported to significantly contribute to the mechanisms of seizure generation and epileptogenesis [32, 36]. Recently, the NLRP3 inflammasome, a key constituent of the inflammasome complex, has been implicated in the sterile inflammatory response via the processing of caspase-1, IL-18, and IL-1β in the setting of epilepsy and some other neurological diseases [37–39].

Rev-Erbα has been implicated in the regulation of immune activation and cytokine release in various tissues and organs [40, 41]. In studies of NR1D1^−/− mice and human macrophages involving the pharmacological activation of Rev-Erbα, researchers have found that Rev-Erbα regulates the expression of NLRP3 and the secretion of cytokines by macrophages [24]. Indeed, recent in vivo and in vitro experiments have suggested Rev-Erbα is an important negative regulator of neuroinflammation [16]. On the one hand, the loss of Rev-Erbα promotes proinflammatory microglial activation, microgliosis, and astrocyte activation in the hippocampus [16]. On the other hand, treatment with Rev-Erbα agonists suppresses inflammatory gene expression in vivo and in cultured glial cells [16–18]. However, little is known about the role of Rev-Erbα in TLE. In the current study, immunohistochemical assays clarified the characteristically low level of expression of Rev-Erbα in glial cells in the epileptic foci of TLE patients. Double labeling experiments delineated that Rev-Erbα colocalized with GFAP and Iba1 in reactive astrocytes and microglia, respectively. Furthermore, SR9009 application for 7 days was effective for suppressing SE-induced inflammation in the hippocampus. This was evinced mainly through reduced protein concentrations of NLRP3, ASC, Caspase1, IL-1β, IL-18, IL-6, TNF-α, GFAP, and Iba1 and diminished numbers of activated astrocytes and Iba1-positive cells in the hippocampus. Hence, it is tempting to propose that the downregulation of Rev-Erbα in the TLE epileptic zone may be involved in the pathophysiology of TLE through the modulation of neuroinflammation. The application of SR9009 after SE may partially alleviate neuroinflammation. It should be noted that both activated resident microglia and infiltrated monocytes contribute to the expansion of Iba1-positive population following pilocarpine-induced SE [42]. Therefore, further investigation is needed to clarify the inhibitory effect of SR9009 on activated resident microglia and infiltrated monocytes.

Several studies have demonstrated that Rev-Erbα and neuronal apoptosis are inextricably linked. For example, some investigators have found a significant increase in apoptosis of internal granule cell layer neurons in the cerebellum of NR1D1 mutant mice [19]. Additionally, dopaminergic neuronal loss in the substantia nigra pars compacta (SNpc) induced by 6-hydroxydopamine is accelerated and intensified in Rev-Erbα knockout mice [20]. These studies suggest that Rev-Erbα deficiency may directly or mediately cause neuronal damage in some areas of the brain. In our research, Rev-Erbα was downregulated in surgical samples from intractable TLE patients. We also observed a decrease in Rev-Erbα in the hippocampus and temporal neocortex of mice in both the early and chronic phases after SE. The results of IHC demonstrated that Rev-Erbα was downregulated in the neurons of TLE patients. Importantly, SR9009 application for 7 consecutive days after SE produced a neuroprotective effect in the CA1 and CA3 regions of the hippocampus. It is well recognized that the hippocampus is the region that is most prone to neuronal apoptosis and loss caused by recurrent epileptic seizures, especially initial SE [33, 43, 44]. Accordingly, it is reasonable to presume that the low concentration of Rev-Erbα in epileptic tissue in TLE may aggravate neuronal apoptosis and loss after epileptic seizures. However, it is worth pointing out that a persistent neuroinflammatory response in the epileptic brain may result in neuronal dysfunction, neuronal cell damage, and death [45, 46]. Interestingly, in a primary neuron and glia coculture model, Rev-Erbα-deficient primary glial cultures exacerbate oxidative damage in cultured neurons, which may be related to the increased inflammatory response of Rev-Erbα-deficient glial cells [16]. Therefore, the next step is to elaborate on the specific mechanisms by which Rev-Erbα reduction leads to neuronal apoptosis and loss in TLE.

Our study also has some limitations. Hippocampal lesions are a characteristic pathological change in TLE patients. Unfortunately, there were no normal hippocampal samples obtained from autopsy or by other means during our study period because of ethical principles. The sample size of each group, which is marginally acceptable, may limit the generalizability of our results. In addition, to avoid the influence of traumatic factors on Rev-Erbα expression, we also examined the expression of Rev-Erbα in direct traumatic tissues and distant tissues within 4 h after traumatic brain injury and found no difference between them, which further verified the appropriateness of our control samples (data not shown).