Aims: Circadian clocks in the hippocampus (HPC) align memory processing with appropriate time of day. Our study was aimed at ascertaining the specificity of glycogen synthase kinase 3-beta (GSK3β)- and glucocorticoid (GC)-dependent pathways in the entrainment of clocks in individual HPC regions, CA1-3, and dentate gyrus (DG). Methods: The role of GCs was addressed in vivo by comparing the effects of adrenalectomy (ADX) and subsequent dexamethasone (DEX) supplementation on clock gene expression profiles (Per1, Per2, Nr1d1, and Bmal1). In vitro the effects of DEX and the GSK3β inhibitor, CHIR-99021, were assessed from recordings of bioluminescence rhythms in HPC organotypic explants of mPER2Luc mice. Results: Circadian rhythms of clock gene expression in all HPC regions were abolished by ADX, and DEX injections to the rats rescued those rhythms in DG. The DEX treatment of the HPC explants significantly lengthened periods of the bioluminescence rhythms in all HPC regions with the most significant effect in DG. In contrast to DEX, CHIR-99021 significantly shortened the period of bioluminescence rhythm. Again, the effect was most significant in DG which lacks the endogenously inactivated (phosphorylated) form of GSK3β. Co-treatment of the explants with CHIR-99021 and DEX produced the CHIR-99021 response. Therefore, the GSK3β-mediated pathway had dominant effect on the clocks. Conclusion: GSK3β- and GC-dependent pathways entrain the clock in individual HPC regions by modulating their periods in an opposite manner. The results provide novel insights into the mechanisms connecting the arousal state-relevant signals with temporal control of HPC-dependent memory and cognitive functions.

The CA1-3 and dentate gyrus (DG) regions of the hippocampus (HPC) are key structures involved in cognition and memory formation, which are temporally regulated by the circadian system [1]. The system is composed of circadian clocks located throughout the body, including the HPC, which are under the general control of a light-entrainable clock in the suprachiasmatic nucleus (SCN). Cell autonomous molecular oscillations are produced by transcriptional-translational feedback loops (TTFLs) [2] which employ an interplay among several canonical clock genes (Per1,2, Cry1,2, Clock, Bmal1, and Nr1d1) and their protein products, resulting in rhythmic transcription of the clock genes. The circadian clock plays an important role in regulation of the HPC function since its genetic disruption (either globally or specifically in the forebrain) produced deficits in HPC-dependent memory [3‒5].

In the HPC, numerous studies detected daily rhythms in expression of clock genes participating in the TTFL in the HPC tissue collected from rats, mice, and hamsters [6‒16], although some studies reported an absence of in vivo rhythms [17, 18]. Perhaps, the best direct in vivo evidence for a functional circadian clock within the HPC is the demonstration of rhythmic Cry1 expression in CA1/2 detected by fluorescence reporter recording in mice maintained under constant conditions [19]. The autonomous nature of the HPC clock has been demonstrated in animals lacking a functional SCN [8]. This was later supported by evidence that the HPC tissue explanted from neonatal animals maintains rhythmicity in vitro [14]. Additionally, the circadian rhythms in PER2-driven bioluminescence persist in single HPC neurons maintained in cultures for several days [20].

In rats and mice, the in vivo phases of the clock in the HPC and other brain areas are phase delayed relative to the SCN clock [21]. This appears to reflect the nocturnal behavioural pattern, suggesting that the phase of the HPC clock may be an important determinant of the behavioural state. Signals that provide specific timing of HPC clocks are unknown. For their entrainment, the HPC clocks must be responsive not only to the SCN-derived signalling pathways but also to inputs that indicate ongoing sensory responses. Theoretically, the SCN may set the HPC phase via indirect neuronal signals [22] and/or plausible hormonal signals. Glucocorticoids (GCs) are strong candidates as synchronizing cues of the HPC clock because HPC is rich in GC receptors (GRs) [23], and GCs were shown to modulate memory formation in the HPC [24‒28]. Importantly, their levels exhibit a SCN-driven, high-amplitude circadian variation [29, 30] coordinated with the activity state, and they can synchronize circadian clocks in various cellular models, tissues, and organs [31]. In accordance, clocks in other brain extra-SCN regions, such as BNST-OV, nuclei raphe, and amygdala, have been shown to be sensitive to GCs [11, 32, 33]. Additionally, the HPC function is regulated intrinsically by activity of the enzyme glycogen synthase kinase 3-beta (GSK3β) [34‒36]. Importantly, the endogenous GSK3β activity exhibits significant circadian rhythmicity in CA1 [34, 37], and treatment with a selective GSK3β inhibitor, CHIR-99021, significantly shortens period of bioluminescence rhythms monitored in the brain explants containing the HPC tissue [34].

So far, none of the studies focused on the effects of modulation of GC- and GSK3β-dependent pathways on the circadian clocks in real time and separately in the individual HPC subregions. Also, a relative strength of entrainment of the clocks in HPC subregions via these 2 mechanisms has not been examined. Therefore, in this study, we tested the hypothesis that the 2 pathways are involved in mechanisms which keep the HPC clocks synchronized with the time of day when the HPC function is most demanding. In an in vivo experiment using rats, we first studied whether and how the absence of endogenous GC signals due to adrenalectomy (ADX) impacts the clocks in individual HPC subregions and whether application of synthetic GC analogue dexamethasone (DEX) to those animals can compensate for these effects selectively in CA1-3 and DG. Next, using in vitro-cultured organotypic explants prepared from neonatal mPER2Luc mice, we examined effects of DEX and selective inhibitor of GSK3β, CHIR-99021, on the clocks in HPC subregions in real time. Finally, we assessed the competition between external humoral signals (GCs) and signals related with the intrinsic regionally specific GSK3β activity to reset the clocks in the HPC subregions. As the CA2 region of neonatal HPC is difficult to reliably demark in the in vitro cultures (see representative photographs below and video in online suppl. Material; for all online suppl. material, see www.karger.com/doi/10.1159/000517689), the analyses in this study were mostly focused on comparison between CA1, CA3, and DG subregions.

Animals

Adult male Wistar rats (Institute of Physiology, the Czech Academy of Sciences) and mPER2Luc mouse dams with her 6-day-old pups of both sexes (strain B6.129S6-Per2tm1Jt/J, JAX, USA; a colony maintained at the Institute of Physiology, the Czech Academy of Sciences) were housed under a 12-h light/12-h dark cycle; lights on at 6:00 was assigned as Zeitgeber time 0 (ZT0) and lights off at 18:00 was assigned as ZT12. Food and water were provided ad libitum throughout the experiment.

In vivo Experiments

For detection of the acute responses to DEX, intact rats maintained in LD12:12 were treated with DEX (1 mg/kg b.w.; n = 30) or vehicle (VEH; phosphate-buffered saline as a vehicle; n = 30) 3 h after lights on. The dose of DEX was used based on the previous studies [38], and timing of its injection was chosen to match the low levels of endogenous GCs in rats [29]. Animals were sacrificed by rapid cervical dislocation at 0, 30, 60, 120, 240, and 480 min following the VEH or DEX application (5 animals per time point and group). The brains were frozen on dry ice and kept at −80°C.

For detection of the effect of DEX on the clock gene expression profiles, the rats were subjected to ADX or sham surgery (the same procedure but without removal of adrenal glands) according to the method described previously [39]. The ADX animals received saline in their drinking water and were either untreated (ADX; n = 35) or treated with DEX (1 mg/kg b.w.) at ZT12 for 8 days (ADX + DEX; n = 27). DEX was administered at time corresponding to the peak of endogenous GC rhythm in rats [29]. The sham-operated rats (sham; n = 35) were untreated and used as controls. The rats were sacrificed 8 days after surgery under deep isoflurane anaesthesia in constant darkness in 4 h intervals during the 24 h (4–5 animals per time point). The brains were frozen on dry ice and kept at −80°C. Time was expressed as circadian time with time zero corresponding to the previous lights-on. For confirmation of the surgery success, see [39]. The same group of rats was previously used for analysis of clock gene expression in the choroid plexus in an earlier publication [40].

In situ Hybridization

Frozen rat brain sections were processed for in situ hybridization, as previously described [41], to determine the expression of Per1, Per2, Nr1d1, and Bmal1 in the HPC regions (CA1, CA3, and DG). In brief, the cDNA fragments of rat rPer1 (980 bp; 581–1,561; GenBank AB_002108), rPer2 (1,512 bp; 369–1,881; GenBank NM_031678), rBmal1 (841 bp; 257–1,098; GenBank AB012600), and Nr1d1 (1,109 bp, BC_062047) were used as templates for the in vitro transcription of cRNA probes. The probes were labelled using 35S-UTP, and the in situ hybridizations were performed, as previously described. The brain sections were hybridized for 20 h at 60°C. Following a post-hybridization wash, the sections were dehydrated in ethanol and dried. Finally, the slides were exposed to a BIOMAX MR film (Kodak, Rochester, NY, USA) for 10–14 days and developed with the ADEFO-MIX-S developer and ADEFOFIX fixer (Adefo-Chemie, Dietzenbach, Germany) in an automatic film processing machine (Protec, Oberstenfeld, Germany). Brain sections of all experimental groups were processed simultaneously under identical conditions. Thereafter, the sections were processed for cresyl violet staining to identify the position of the individual HPC regions. Autoradiographs of the sections were analysed with an image analysis system (Image Pro, Olympus, New York, NY, USA) to detect the relative optical density of the specific hybridization signal in the areas of the HPC. For comparisons between the experimental groups, the data were normalized to the highest value of each of the daily profiles.

Real-Time qPCR

Frozen rat brains were sectioned on cryostat into 30-µm coronal sections. CA1, CA3, and DG were separated bilaterally using a laser microdissector (LMD6000, Leica, Wetzlar, Germany), total RNA was isolated, and reverse-transcribed into cDNA, as described previously [42]. The cDNA samples were analysed by RT-qPCR using 5x HOT FIREPol Probe qPCR mix Plus (Solis Biodyne, Tartu, Estonia) and TaqMan Gene Expression Assays (Thermo Fisher Scientific, Waltham, MA, USA), spanning exon junctions specific for rat genes Gilz (Rn00580222_m1), Per1 (Rn01325256_m1), Per2 (Rn01427704_m1), Nr1d1 (Rn01460662_m1), and Bmal1 (Rn00577590_m1). The mRNA concentrations were normalized relative to the beta-2-microglobulin (Rn00560865_m1) housekeeping gene expression, measured in a duplex reaction. Relative cDNA concentrations were quantified using the Pfaffl ΔΔCt method.

Immunohistochemistry

Adult Wistar rats were sacrificed under deep isoflurane anaesthesia at ZT12-14. Their brains were frozen on dry ice, sectioned into 15-μm-thick sections containing dorsal HPC and processed for standard immunohistochemistry, as described elsewhere [43]. For detection of immunostaining on the entire HPC-containing sections, the antigens were visualized by diaminobenzidine (DAB; CAT# D5905, Sigma Aldrich, St. Louis, MO, USA). The double-labelling technique was used to detect the protein co-localizations using fluorescence-labelled goat polyclonal secondary antibodies in final concentration 1:600 (anti-rabbit antibody conjugated with Alexa Fluor® 594, Cat# A-11037, RRID:AB_2534095, Thermo Fisher Scientific; anti-mouse antibody conjugated with FITC, Cat# F-2761, RRID:AB_2536524, Thermo Fisher Scientific; anti-chicken antibody conjugated with Cy3®, Cat# ab97145, RRID:AB_10679516, Abcam, Cambridge, UK). The primary antibodies were used as follows: rabbit GR polyclonal antibody with epitope mapping at the N-terminus of GR α, final concentration 1:100 (M-20, CAT#:sc-1004, RRID:AB_2155786, Santa Cruz Biotechnology, Dallas, TX, USA); rabbit PER2 polyclonal antibody with epitope mapping at the N-terminus of PER2, final concentration 1:400 (CAT#:AB2202, RRID:AB_1587380, Millipore, Burlington, MA, USA); rabbit phospho-GSK-3β (Ser9) monoclonal antibody with epitope mapping at residues surrounding phosphorylated Ser9 of GSK-3β, final concentration 1:400 (D85E12, CAT#:5558, Cell Signalling, Danvers, MA, USA); chicken GFAP polyclonal primary antibody with epitope mapping full protein of GFAP final concentration 1:1,000 (CAT#:ab4674, RRID:AB_304558, Abcam); and mouse HuC/HuD monoclonal antibody with epitope mapping KLH-conjugated linear peptide corresponding to HuC/HuD, final concentration 1:800 (15A7.1, CAT#:MABN153, Millipore). The immunofluorescence was detected using Leica SP8 WLL MP laser-scanning confocal microscope; the images with the DAB staining were taken using Olympus BX53 microscope.

Organotypic Explants, Bioluminescence Recordings, and Treatment Protocol

Six-day-old mPER2Luc mice were sacrificed by rapid cervical dislocation (between 12:00 and 15:00 h), and brains were sectioned into 300-µm coronal sections using a vibratome (Leica, Wetzlar, Germany) in cold Earle’s Balanced Salt Solution ( Sigma). HPC was dissected and immediately placed onto Millicell Culture Inserts (Merck) inside 35-mm petri dishes containing 1 mL of recording media. HPC explants were kept in CO2-buffered recording media (MEM supplemented with 25% heat-inactivated horse serum, 25% Earle’s Balanced Salt Solution, 100 U/mL penicillin, 100 µg/mL streptomycin, and 2 mM GlutaMAX [all Thermo Fisher Scientific]). The explants were placed in a motorized Luminoview LV200 luminescence microscope (Olympus, Tokyo, Japan) with a LUCPLFLN20X or UPLSAPO10X2 objective (Olympus) and ImageEM X2 EMCCD camera (Hamamatsu, Hamamatsu City, Japan) water cooled by Minichiller 280 (Huber, Offenburg, Germany) with exposure time of 3–6 min. The 6-day-old mice were used for preparation of the explants based on the pilot study which proved that explants prepared from adult mice did not exhibit bioluminescence rhythms within the HPC regions (for more details, see online suppl. material).

For the treatment procedure, the organotypic HPC explants were cultured in fresh recording media for 3 days, then they were exposed to a treatment with 1 µL of either DEX (Sigma-Aldrich) (100 µM), CHIR-99021 (Sigma-Aldrich, CAT#:SML1046) (20 mM), or with 1 µL of the corresponding vehicles (0.001% ethanol in ddH2O for DEX and DMSO for CHIR-99021), per 1 mL of media. A separate set of explants was co-treated with 1 μL of both CHIR-99021 (or corresponding vehicle – DMSO) and DEX.

Data Analysis

Daily profiles of normalized clock gene expression were analysed for rhythmic expression by fitting 2 alternative regression models: horizontal straight line (null hypothesis) or cosine curve defined by the equation Y = mesor + {amplitude × cos (2 × π × [X-acrophase]/wavelength)} with a constant wavelength of 24 h (alternative hypothesis). p values, R2 (goodness of fit), amplitudes, and acrophases were determined. Fold changes in gene expression detected after acute DEX injection in vivo based on VEH controls were analysed using 2-way ANOVA. All statistics were performed using GraphPad Prism 7 software (GraphPad, San Diego, CA, USA).

The microscopic images obtained by the Luminoview LV200 were analysed using ImageJ (NIH, Bethesda, MD, USA). The data were fitted with a damped sine wave to calculate the period, amplitude, mesor, and R2 in Prism 7 software (GraphPad, La Jolla, CA, USA). Cosmic rays registered by the LV200 system were removed before analysis by pixel-wise subtraction of consecutive images.

For assessment of the effects of various treatments of explants on the parameters of bioluminescence rhythms (amplitude, mesor, R2), ratios of their values before and after each treatment were calculated (value 1 means no change and values above and below 1 mean increase and decrease, respectively) and compared by unpaired t test (2 groups) or one-way ANOVA (3 groups) with the post hoc analysis of Sidak’s multiple comparison method. Period changes (values before and after the treatment for each explant) were statistically compared with the paired t test.

Localization of Immunopositive Cells for GR, PER2, and P-GSK3β in the HPC Subregions

We first used immunohistochemistry to localize cells expressing GR, PER2, and P-GSK3β in the individual HPC subregions (Fig. 1). Images of brain sections containing the whole HPC are presented in the left panels and individual HPC subregions, in which the proteins were co-localized with neuronal or glial markers, as shown in the right panels. The results provided gross comparison of the protein expression levels in the HPC subregions; for purpose of this study, no detailed quantification was performed. Cells expressing GRs (Fig. 1a) were detected in all parts of HPC, and the immunostaining was comparable in CA1 and DG and lower level in CA3, in accordance with previously published results [44]. In all 3 subregions, the GR-immunopositive cells were identified mostly as neurons and sparsely also as glia. Cells expressing PER2 (Fig. 1b) spanned across the HPC with the highest density in CA1 and, similar to GR, immunopositive cells were mostly neurons and sparsely also glia. The higher PER2 levels in CA1 compared to other parts are reflected in the highest level of PER2-bioluminescence, as shown below. In contrast to GR and PER2, cells expressing inhibited form of GSK3β (P-GSK3β) (Fig. 1c) were restricted solely to the CA1 neurons. Altogether, the results confirm that CA1 and DG regions express comparable levels of GRs but they differ in absolute levels of 2 clock components; the CA1 neurons exhibit higher levels of the PER2 than DG and express P-GSK3β, and the cells in DG (as well as CA3) have relatively lower levels of PER2 than CA1 and lack the inhibited form of GSK3β. The differences in the spatial localization of these 2 clock components in the individual HPC regions provided rationale for testing a hypothesis that the clocks in CA1 and DG differ in their sensitivity to signalling pathways related to the levels of GCs and GSK3β activity.

Fig. 1.

Localization of immunopositive cells for GR, PER2, and P-GSK3β in the HPC subregions. a Left panel: representative coronal brain section overviewing distribution of DAB-labelled GR-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of GR (red) with Huc/Hud neuronal marker (green), GFAP glial marker (yellow), and DAPI (blue) in CA1, CA3, and DG subregions. Examples of GR immunopositivity present in neuronal cells (white arrows) and glial cells (yellow arrows) are depicted. b Left panel: representative coronal brain section overviewing distribution of DAB-labelled PER2-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of PER2 (red) with Huc/Hud neuronal marker (green), GFAP glial marker (yellow), and DAPI (blue) in CA1, CA3, and DG subregions. Examples of GR-immunopositivity present in neuronal cells (white arrows) and glial cells (yellow arrows) are depicted. c Left panel: representative coronal brain section overviewing distribution of DAB-labelled GSK3β-Ser9P-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of GSK3β-Ser9P (red) with Huc/Hud neuronal marker (green) and DAPI (blue) in CA1. HPC, hippocampus; GR, glucocorticoid receptor; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Fig. 1.

Localization of immunopositive cells for GR, PER2, and P-GSK3β in the HPC subregions. a Left panel: representative coronal brain section overviewing distribution of DAB-labelled GR-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of GR (red) with Huc/Hud neuronal marker (green), GFAP glial marker (yellow), and DAPI (blue) in CA1, CA3, and DG subregions. Examples of GR immunopositivity present in neuronal cells (white arrows) and glial cells (yellow arrows) are depicted. b Left panel: representative coronal brain section overviewing distribution of DAB-labelled PER2-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of PER2 (red) with Huc/Hud neuronal marker (green), GFAP glial marker (yellow), and DAPI (blue) in CA1, CA3, and DG subregions. Examples of GR-immunopositivity present in neuronal cells (white arrows) and glial cells (yellow arrows) are depicted. c Left panel: representative coronal brain section overviewing distribution of DAB-labelled GSK3β-Ser9P-immunopositive cells in the rat HPC. Position of CA1, CA3, and DG regions is depicted. Right panel: co-localization of GSK3β-Ser9P (red) with Huc/Hud neuronal marker (green) and DAPI (blue) in CA1. HPC, hippocampus; GR, glucocorticoid receptor; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Close modal

GCs Are Required for HPC Rhythm Expression with Region-Specific Recovery Response to DEX Treatment in vivo

HPC subregions responded acutely to DEX administration in vivo (Fig. 2a). The Wistar rats (n = 5 animals per time point and the treatment group) were injected with a single dose of DEX (1 mg per kg) or vehicle (PBS), and the acute effect of DEX on gene expression was detected by RT qPCR in the laser-dissected CA1 and DG regions. The DEX injection acutely induced expression of the GC-immediate responsive gene Gilz, which was significantly elevated already 120 min after the intraperitoneal (i.p.) injection (CA1, p = 0.0164 and DG, p = 0.0045) and in CA1 the expression remained significantly elevated also 240 min after the injection (p < 0.0001). Therefore, DEX is able to reach HPC neurons and initiate transcriptional responses soon after the i.p. injection. The proof of concept was needed for the next experiment in which we examined effects of systemic DEX injection to rats on daily profiles of clock gene expression in the HPC (see below). However, unlike Gilz, DEX did not acutely affect expression of any of the tested clock genes (Per1, Per2, and Nr1d1) within the 480 min-interval after the injection.

Fig. 2.

Effect of GCs on the HPC clock in vivo. a Acute changes in Gilz, Per1, Per2, and Nr1d1 mRNA levels detected by RT qPCR in CA1 (light blue) and DG (dark blue) following a single injection of rats with DEX or VEH. The transcript fold changes detected 30, 60, 120, and 480 min after the injection were compared with the baseline value (time 0) (n = 5 animals per time point and treatment group). ****p < 0.0001, ***p = 0.0002, **p = 0.0045, and *p = 0.0164. b Daily profiles of Per1, Per2, Nr1d1, and Bmal1 mRNA levels in CA1 (full circles, full line), CA3 (open circles, dashed line), and DG (full triangles, full line) of control (sham) (left) and ADX (middle) rats or ADX rats which were repeatedly injected with DEX (ADX + DEX) (right), detected by in situ hybridization (n = 4–5 animals per time point). The data were fitted with cosine curves; the strait lines mean non-significant cosine fit (results are summarized in Table 1). Time is expressed as CT (h). Representative film autoradiographs of unilateral HPC formation (far right) refer to the expression profiles and show troughs and peaks of expression of each of the studied clock gene within CA1, CA3, and DG regions (areas denoted by yellow dotted lines in 1 of the autoradiographs). DEX, dexamethasone; VEH, vehicle; ADX, adrenalectomy; HPC, hippocampus; DG, dentate gyrus; CT, circadian time.

Fig. 2.

Effect of GCs on the HPC clock in vivo. a Acute changes in Gilz, Per1, Per2, and Nr1d1 mRNA levels detected by RT qPCR in CA1 (light blue) and DG (dark blue) following a single injection of rats with DEX or VEH. The transcript fold changes detected 30, 60, 120, and 480 min after the injection were compared with the baseline value (time 0) (n = 5 animals per time point and treatment group). ****p < 0.0001, ***p = 0.0002, **p = 0.0045, and *p = 0.0164. b Daily profiles of Per1, Per2, Nr1d1, and Bmal1 mRNA levels in CA1 (full circles, full line), CA3 (open circles, dashed line), and DG (full triangles, full line) of control (sham) (left) and ADX (middle) rats or ADX rats which were repeatedly injected with DEX (ADX + DEX) (right), detected by in situ hybridization (n = 4–5 animals per time point). The data were fitted with cosine curves; the strait lines mean non-significant cosine fit (results are summarized in Table 1). Time is expressed as CT (h). Representative film autoradiographs of unilateral HPC formation (far right) refer to the expression profiles and show troughs and peaks of expression of each of the studied clock gene within CA1, CA3, and DG regions (areas denoted by yellow dotted lines in 1 of the autoradiographs). DEX, dexamethasone; VEH, vehicle; ADX, adrenalectomy; HPC, hippocampus; DG, dentate gyrus; CT, circadian time.

Close modal

Using in situ hybridization, we next detected impact of ADX on daily profiles of Per1, Per2, Nr1d1, and Bmal1 clock gene expressions in the individual HPC subregions. The gene expression profiles and representative autoradiographs showing the maximal and the minimal expression of each gene during the 24 h profiles are depicted in Figure 2b. In control animals (sham group), we found that all 3 HPC parts (CA1, CA3, and DG) exhibited significant circadian rhythms in expression of clock genes (see Table 1 – for cosinor analysis results). Within each of the 3 regions, expression profiles of individual clock genes were roughly in synchrony. The rhythms were rather shallow with the exception of Nr1d1 which exhibited the largest amplitude of all studied genes in all 3 HPC regions (Table 1). The ADX almost completely abolished circadian rhythms in expression of all studied clock genes in CA1, CA3, and DG, leaving their expression constitutive (ADX; Table 1). Finally, we tested the effect of repeated injections of ADX animals with DEX (1 mg per kg, i.p. injection before lights-off) into ADX animals for 8 days preceding collection of samples (ADX + DEX group). DEX administration produced circadian rhythmic expression of the 4 clock genes in DG. However, in CA1, only Per2 and Nr1d1 and in CA3, only Per2 rhythms were significant (ADX + DEX; Table 1). Thus, of all studied clock genes, only Per2 rhythmicity was reinitiated by DEX injections in all parts of HPC, and the rhythm in DG was in opposite phase to CA1 and CA3 (Table 1, acrophases). Importantly, the DG was the region where DEX reinitiated rhythms with the largest amplitudes of all HPC parts (Table 1, amplitudes). The results demonstrated that DEX injections are not only able to reinitiate but also to reset the HPC clock in vivo with the most significant effect on Per2 in the DG region.

Table 1.

Results of cosinor analysis of clock gene expression profiles in the hippocampal CA1, CA3, and DG of control (sham), ADX and ADX + DEX rats

 Results of cosinor analysis of clock gene expression profiles in the hippocampal CA1, CA3, and DG of control (sham), ADX and ADX + DEX rats
 Results of cosinor analysis of clock gene expression profiles in the hippocampal CA1, CA3, and DG of control (sham), ADX and ADX + DEX rats

Circadian Clocks in the Individual HPC Subregions Respond Differently to DEX and CHIR-99021 in vitro

To examine resetting of the clocks in the individual HPC subregions in vitro, we prepared explants from 6-day-old mPER2Luc mice and recorded bioluminescence using Luminoview LV200 luminescence microscope. Representative microscopic images of an explant with denoted areas of the individual HPC subregions where the bioluminescence was measured are depicted in Figure 3a (for representative video, see online suppl. Material). Bioluminescence was monitored for 3 cycles before and after the treatments with DEX (100 nM) or selective GSK3β inhibitor CHIR-99021 (20 µM) or corresponding VEHs, that is, 0.001% ethanol in ddH2O for DEX (VEHDEX) and DMSO for CHIR-99021 (VEHCHIR). Representative bioluminescence traces in which the circadian parameters (amplitude, period, and mesor) were calculated (for more details, see Material and Methods) are shown in Figure 3b. We confirmed that treatment procedure itself (application of both VEHs) had no or only negligible effect on the PER2-driven bioluminescence rhythms; the rhythms continued with a minor gradual decay in course of the culturing, as apparent from the values of amplitude and mesor ratios (ratios of values before and after the treatment) being approximately equal or slightly below 1 (Fig. 3c, d). The periods of clocks in CA1, CA3, and DG were not affected by application of VEHDEX (CA1: n = 7, p = 0.1893; CA3: n = 7, p = 0.613; DG: n = 11, p = 0.1721) and were only marginally shortened after the treatment with VEHCHIR (CA1: n = 9, p = 0.0107; CA3: n = 9, p = 0.0148; and DG: n = 9, p = 0.0349) (Fig. 3e).

Fig. 3.

Effect of DEX and the GSK3β inhibitor CHIR-99021 on clocks in the HPC subregions in vitro. a Representative photographs from LV200 luminescence microscope showing organotypic HPC explant from 6-day-old mPER2Luc mouse. CA1, CA3, and DG subregions are highlighted (left). The same explant with bioluminescence signal at its trough (centre) and peak (right) is shown. b Representative traces of bioluminescence measured in individual subregions of HPC organotypic explants from mPer2Luc mice treated with VEHDEX, DEX, VEHCHIR, or CHIR-99021 (CHIR) (treatment is depicted by dotted line). Each graph represents traces from 1 explant. c Amplitude ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 10), DEX (blue, nCA1 = 9, nCA3 = 7, nDG = 10), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9) or CHIR (red, nCA1 = 8, nCA3 = 5, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM. CA1: *PVEH/DEX = 0.0028 CA3: **PVEH/DEX = 0.0037, *PVEH/CHIR = 0.0192. d Mesor ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 11), DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 10), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9), or CHIR (red, nCA1 = 8, nCA3 = 5, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM ****p < 0.0001. e Periods of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions detected before and after the treatment with VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 11), DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 9), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9), or CHIR (red, nCA1 = 8, nCA3 = 4, nDG = 10). Changes in periods of the individual explants are depicted CA1: *PVEH = 0.0107, PCHIR = 0.0127, CA3: **PDEX = 0.0048, *PVEH = 0.0148, DG: ****PCHIR < 0.0001, ***PDEX = 0.0002, and *PVEH = 0.0349. fR2 ratios (ratio of the value before and after the treatment) from cosinor analysis of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 10) or CHIR (red, nCA1 = 8, nCA3 = 6, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM ***p = 0.0010, *p = 0.0277. DEX, dexamethasone; VEH, vehicle; HPC, hippocampus; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Fig. 3.

Effect of DEX and the GSK3β inhibitor CHIR-99021 on clocks in the HPC subregions in vitro. a Representative photographs from LV200 luminescence microscope showing organotypic HPC explant from 6-day-old mPER2Luc mouse. CA1, CA3, and DG subregions are highlighted (left). The same explant with bioluminescence signal at its trough (centre) and peak (right) is shown. b Representative traces of bioluminescence measured in individual subregions of HPC organotypic explants from mPer2Luc mice treated with VEHDEX, DEX, VEHCHIR, or CHIR-99021 (CHIR) (treatment is depicted by dotted line). Each graph represents traces from 1 explant. c Amplitude ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 10), DEX (blue, nCA1 = 9, nCA3 = 7, nDG = 10), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9) or CHIR (red, nCA1 = 8, nCA3 = 5, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM. CA1: *PVEH/DEX = 0.0028 CA3: **PVEH/DEX = 0.0037, *PVEH/CHIR = 0.0192. d Mesor ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 11), DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 10), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9), or CHIR (red, nCA1 = 8, nCA3 = 5, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM ****p < 0.0001. e Periods of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions detected before and after the treatment with VEHDEX (grey, nCA1 = 7, nCA3 = 7, nDG = 11), DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 9), VEHCHIR (grey, nCA1 = 9, nCA3 = 9, nDG = 9), or CHIR (red, nCA1 = 8, nCA3 = 4, nDG = 10). Changes in periods of the individual explants are depicted CA1: *PVEH = 0.0107, PCHIR = 0.0127, CA3: **PDEX = 0.0048, *PVEH = 0.0148, DG: ****PCHIR < 0.0001, ***PDEX = 0.0002, and *PVEH = 0.0349. fR2 ratios (ratio of the value before and after the treatment) from cosinor analysis of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after DEX (blue, nCA1 = 8, nCA3 = 7, nDG = 10) or CHIR (red, nCA1 = 8, nCA3 = 6, nDG = 10) treatments. Data are expressed as individual ratios and the mean ± SEM ***p = 0.0010, *p = 0.0277. DEX, dexamethasone; VEH, vehicle; HPC, hippocampus; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Close modal

Compared to VEHDEX administration, the DEX treatment significantly increased the amplitudes of the rhythms in CA1 (nDEX = 9, nVEH = 7, p = 0.0131) and CA3 (nDEX = 7, nVEH = 7, p = 0.0225) but not in DG (nDEX = 10, nVEH = 10, p = 0.3041) (Fig. 3c). However, the mesors, that is, absolute PER2 protein levels, were not affected by the DEX application in CA1 (nDEX = 8, nVEH = 7, p > 0.9999), CA3 (nDEX = 7, nVEH = 7, p = 0.7215), and DG (nDEX = 10, nVEH = 11, p = 0.8572) (Fig. 3d). The treatment with CHIR-99021 did not affect amplitude of bioluminescence rhythms in CA1 (nCHIR = 8, nVEH = 9, p = 0.2698) and DG (nCHIR = 10, nVEH = 9, p = 0.1373) and only increased the amplitude relative to its decrease due to VEHCHIR in CA3 (nCHIR = 5, nVEH = 9, p = 0.0192) (Fig. 3c). However, unlike DEX, CHIR-99021 significantly increased mesors of the rhythms (Fig. 3d) in CA1 (nCHIR = 8, nVEH = 9, p < 0.0001), CA3 (nCHIR = 5, nVEH = 9, p < 0.0001), and DG (nCHIR = 10, nVEH = 9, p < 0.0001).

DEX and CHIR-99021 affected the periods in opposite manners, that is, DEX prolonged and CHIR-99021 shortened the periods (Fig. 3e). The effect of DEX was significant in CA3 (n = 7, p = 0.0048) and DG (n = 9, p = 0.0002) but was only suggested in CA1 (n = 8, p = 0.2090), and the effect of CHIR-99021 was significant in CA1 (n = 8, p = 0.0127) and DG (n = 10, p < 0.0001) but not in CA3 (n = 4, p = 0.2566). Additionally, we compared how DEX and CHIR affected quality of the circadian rhythms via assessment of ratios of goodness of the cosinor fits (R2) before and after the treatments in the individual HPC regions (Fig. 3f). The results showed that treatment with DEX did not affect R2 of the rhythms in CA1 but it significantly improved the rhythms in DG (CA1 vs. DG; nCA1 = 8, nDG = 10, p = 0.0010). CHIR-99021 had the opposite effect; it did not affect R2 of the rhythms in DG (n = 10), but it significantly worsened the rhythms in CA1 (DG vs. CA1; nCA1 = 8, nDG = 10, p = 0.0277). The impact on rhythms in CA3 region was intermediate, that is, a mild improvement of R2 after DEX compared to DG and mild worsening of R2 after CHIR-9921 compared to CA1.

The results support the conclusion that DEX had the largest effect on the clock in DG where it prolonged the period and improved quality of the rhythms. The inhibition of GSK3β had also most significant effect in DG where it shortened the period. In contrast, in CA1, the GSK3β inhibition had less significant effect on the period and impaired the rhythmicity. The results revealed that clocks in CA1, CA3, and DG regions differ in their sensitivity to DEX and CHIR-99021. Their application causes opposite changes in periods. Remarkably, the effects of both DEX and CHIR-99021 on the period were most robustly pronounced in the DG.

Inhibition of GSK3β Overrides Effects of DEX on the Period of Clocks in Individual HPC Subregions

Finally, we tested the relative strength of DEX- and GSK3β-dependent entraining pathways (Fig. 4). For this, we recorded bioluminescence in the individual HPC subregions for 3 cycles before and after they were co-treated with CHIR-99021 + DEX or VEH + DEX, and compared the rhythm parameters (amplitude, mesor, and period) between both groups separately in CA1, CA3, and DG. Representative bioluminescence traces are shown in Figure 4a.

Fig. 4.

Effect of co-treatment with DEX and GSK3β inhibitor CHIR-99021 on clocks in the HPC subregions in vitro. a Representative traces of bioluminescence measured in individual HPC subregions of organotypic explants from mPer2Luc mouse treated with VEH + DEX or CHIR + DEX (treatment is depicted by dotted line). Each graph shows traces of 1 explant. b Amplitude ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEH + DEX (grey, nCA1 = 10, nCA3 = 4, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Data are expressed as individual ratios and the mean ± SEM. c Mesor ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEH + DEX (grey, nCA1 = 10, nCA3 = 4, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Data are expressed as individual ratios and mean ± SEM. ****p < 0.0001 ***p = 0.0001. d Periods of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions detected before and after the treatment VEH + DEX (grey, nCA1 = 7, nCA3 = 5, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Periods of the individual explants are depicted; CA1: **p = 0.0082, CA3: **p = 0.0074, DG: ***p = 0.0009, *p = 0.0417. DEX, dexamethasone; VEH, vehicle; HPC, hippocampus; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Fig. 4.

Effect of co-treatment with DEX and GSK3β inhibitor CHIR-99021 on clocks in the HPC subregions in vitro. a Representative traces of bioluminescence measured in individual HPC subregions of organotypic explants from mPer2Luc mouse treated with VEH + DEX or CHIR + DEX (treatment is depicted by dotted line). Each graph shows traces of 1 explant. b Amplitude ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEH + DEX (grey, nCA1 = 10, nCA3 = 4, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Data are expressed as individual ratios and the mean ± SEM. c Mesor ratios (ratio of the value before and after the treatment) of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions after VEH + DEX (grey, nCA1 = 10, nCA3 = 4, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Data are expressed as individual ratios and mean ± SEM. ****p < 0.0001 ***p = 0.0001. d Periods of the bioluminescence rhythms of the CA1, CA3, and DG HPC subregions detected before and after the treatment VEH + DEX (grey, nCA1 = 7, nCA3 = 5, nDG = 6) or CHIR + DEX (purple, nCA1 = 7, nCA3 = 5, nDG = 6) treatments. Periods of the individual explants are depicted; CA1: **p = 0.0082, CA3: **p = 0.0074, DG: ***p = 0.0009, *p = 0.0417. DEX, dexamethasone; VEH, vehicle; HPC, hippocampus; DG, dentate gyrus; GSK3β, glycogen synthase kinase 3-beta.

Close modal

The VEH + DEX co-treatment on the HPC clocks had no significant effect on the amplitude (Fig. 4b) nor the mesor (Fig. 4c) because ratios of their values before and after the treatment were close to 1. However, it lengthened the periods (Fig. 4d), and the effect was significant in CA1 (n = 10, p = 0.0082) and DG (n = 8, p = 0.0417) but not in CA3 (n = 5, p = 0.0776). The co-treatment of CHIR-99021 + DEX had effects which were similar to the treatment with CHIR-99021 alone (see Fig. 3); compared to VEH + DEX, it did not affect the amplitudes (Fig. 4b; CA1: nVEH+DEX = 10, nCHIR + DEX = 6, p = 0.3975; CA3: nVEH+DEX = 4, nCHIR + DEX = 5, p = 0.4381; and DG: nVEH + DEX = 6, nCHIR + DEX = 5, p = 0.1148) but significantly increased the mesors (Fig. 4c; CA1: nVEH + DEX = 10, nCHIR + DEX = 7, p < 0.0001; CA3: nVEH + DEX = 4, nCHIR + DEX = 5, p = 0.0001; and DG: nVEH + DEX = 7, nCHIR + DEX = 6, p < 0.0001). Importantly, the periods after CHIR-99021 + DEX were shortened (Fig. 4d), which was significant in CA3 (n = 5, p = 0.0074) and DG (n = 6, p = 0.0009) but not in CA1 (n = 7, p = 0.0540). Therefore, the results show that co-treatment of the HPC explants with CHIR-99021 + DEX produced CHIR-99021-like response of clocks in the HPC subregions.

In this study, we provide a novel insight into complexity of mechanisms involved in entrainment of the circadian clocks within the individual HPC subregions, namely those related with systemic, adrenal gland-derived signals (GCs) and local, HPC neuron-derived signals (GSK3β activity). We revealed that intact adrenal glands play a crucial role in maintenance of the HPC clock as their absence leads to severe dampening of rhythmicity detected in situ in all parts of HPC. Repeated daily DEX injections can reinitiate those rhythms with the most pronounced effect in the DG. Similarly, in the HPC explants cultured in vitro, DEX lengthened periods of clocks in the individual HPC subregions with the most significant effect in DG. The inhibition of GSK3β activity by CHIR-99021 shortened the period and thus had opposite effect on the periods than DEX. Surprisingly, co-treatment experiments revealed that pharmacological inhibition of GSK3β has dominant effect over DEX application on period and amplitude of the HPC clock. Altogether, the data demonstrate that rhythms in endogenous GC levels and GSK3β activity represent 2 important pathways with opposing regulatory effects on clocks in the HPC and that their relative impact on period and entrainment is region-specific.

Existence of the circadian clocks in the HPC has been previously confirmed by many studies in which rhythmic expression of clock genes in the HPC tissue sampled from animals around the clock was detected [6‒15, 20, 21, 34, 45‒58]. In agreement with these reports, in our control animals, we also detected in situ shallow but significant daily rhythmic profiles of clock gene expression within the CA1, CA3, and DG. The antiphase Bmal1 and Nr1d1 profiles comply with the current TTFL model and are shifted relative to the clock in the SCN of the same rat strain [59]. The relatively low amplitude of the HPC clock detected in situ was not due to a poor mutual synchrony among the individuals because in brains of same animals, clock gene expression profiles in another extra-SCN region, choroid plexus, exhibited high-amplitude oscillation [40]. The autonomous nature of the HPC rhythmicity has been previously elegantly demonstrated by in vivo recording of bioluminescence rhythms in real time within the HPC from living mice [19]. Additionally, several previously published studies employedex vivo organotypic brain explants containing HPC tissues from adult mPER2Luc mice in which bioluminescence rhythms were monitored by photomultiplier tubes. However, over the course of this study, we questioned whether such approach might lead to misinterpretation of the bioluminescence source within these brain explants. This is due mainly to the fact that apart from HPC, the cultured brain sections contain also closely adjacent structures and some of those have been shown to possess robust oscillators and exhibit long-lasting persistence even in tissue explanted from adult animals (for more details, see online suppl. Material). Therefore, in this study, we prepared HPC explants from neonatal mPER2Luc mice which retain viability in long-term cultures and localized the bioluminescence signal with the microscopic resolution in the individual subregions of HPC. To our knowledge, there are no studies as yet that have used this approach to analyse properties and/or sensitivity to entraining signals of clocks in the individual HPC subregions.

Our results demonstrate that GCs are important for maintaining the rhythmicity of the circadian clocks in the individual HPC subregions. As previously demonstrated in other rat models [60‒62], we confirmed in our Wistar rats that GR distribution spans across the entire HPC with high density in CA1 and DG and lower density in CA3. In all parts, we detected GRs dominantly in neurons and sparsely also in glia. Removal of endogenous source of GCs via ADX almost completely abolished in situ-detected expression rhythms of the selected clock genes within CA1, CA3, and DG regions. The GC supplementation by daily repeated injections of rats with synthetic analogue DEX restored those rhythms completely in DG (although with different phases) but only partially in CA1 and CA3. Using DEX as a selective GR agonist confirmed that GCs are the most likely candidates of the adrenal gland-derived hormones involved in the effect. Studies focussing on the effects of GCs on clocks in brain regions are rather sparse and usually address effect of GCs, a single clock gene/protein. Moderate effects of ADX on clock in prefrontal cortex [63] and complete abolishment of Per2 rhythm in BNST-OV [32] have been reported. Corticosterone administration has been reported to reverse these ADX effects [63]. The role of GCs in regulating DG clock was studied but with remarkably controversial results. Inactivation of GR gene in neural tissues did not abolish PER2 rhythm in this structure [12]; however, the presence/absence of the rhythm in that study was assessed only based on 1 daytime (ZT1) versus 1 night-time (ZT13) value. On the other hand, ADX had moderate effect of on PER2-immunopositivity rhythm in DG [8, 11] and elimination of the GC rhythmicity via clamping GCs to a stably high level abolished Per1-luciferase rhythms in the explanted rat HPC [9]. Intriguingly, the peak of long-term potentiation in DG shifted from the night-time to daytime in ADX rats [64], suggesting that in DG hippocampal long-term potentiation is regulated by adrenal glands hormones. In our study, ADX caused complete abolishment of rhythms in expression of genes forming both the positive (Bmal1) and the negative (Per2 and Nr1d1) arms of the TTFL [2] in all HPC subregions, and DEX injections recovered the DG rhythms so that they attained appropriate mutual phases according to the model. Our in vivo results pointed at a significant role of Per2 in DEX resetting of the HPC clock because it was the only one of the studied clock genes whose rhythmicity was induced by DEX injections in all parts of the HPC in ADX animals, and again, with the highest amplitude in the DG. In remarkable accordance, DEX applied to HPC explants in vitro increased quality of the PER2-bioluminescence rhythm (R2 ratio) and affected its period most significantly in the DG. We noticed that thein vivoDEX-reinitiated Per2, Nr1d1, and Bmal1 (but not Per1) rhythms in DG were shifted relative to sham controls, in spite of the DEX injection to ADX rats was performed at the time of expected GCs peak (just before beginning of their activity phase). Interestingly, clock in the choroid plexus of the same animals was shifted in the same way [40]. It is possible that either endogenous pulsatile pattern of rhythmic GC release [29], the local GC metabolism [65], and/or other adrenal-derived signals, participate in fine-tuning the HPC clock phase in vivo. Also, it cannot be excluded that different stability of the synthetic analogue compared to native hormone played a role. Importantly, the in vitro data confirmed that DG was the subregion most sensitive to DEX treatment.

The subregion differences in sensitivity of clocks in the HPC to DEX, and the presence of endogenously inhibited form of GSK3β (P-GSK3β) exclusively in CA1 led us to examine the role of this enzyme activity in entrainment of circadian clocks within the individual HPC subregions. Previously, inhibition of GSK3β activity was reported to enhance the amplitude and shorten the period of luminometry-measured bioluminescence rhythms in organotypic SCN cultures and in brain explants of adult mPER2Luc mice [34, 66]. Nevertheless, this approach could not distinguish between the signal in the HPC and surrounding tissue. In our study, we detected the impact of GSK3β inhibition on the clocks located within the individual HPC subregions. We found that CHIR-99021 shortened period most significantly in DG. In CA1, a less significant effect was accompanied by worsening of the rhythmicity likely due to the high endogenous level of inhibited GSK3β in this region. In DG, the effectiveness of pharmacological inhibition of GSK3β in modulation of the circadian period demonstrates that constitutive activity of the enzyme endogenously present in this region is needed for maintaining proper pace of the clock. Furthermore, as the DG clock was the most sensitive of the clocks in the HPC subregion to the effects of both DEX and CHIR-99021 and responded to each of them with opposite changes in period, we tested a previously unaddressed question which of these 2 signals represents a stronger entraining stimulus for the DG clock. Based on the well-recognized unique position of GCs as dominant entraining stimuli for various peripheral circadian clocks across the circadian system [32, 39, 67‒69], and their significant effects on the HPC clock in vivo (this study), we expected that effects of DEX will dominate the effects of CHIR-99021 on the HPC clocks. Surprisingly, we found the opposite was true. Co-treatment of both DEX and CHIR-99021 clearly produced a pure CHIR-99021-like response which demonstrates that clocks in the individual HPC subregions, and especially in DG, are highly dependent on endogenous tonic activity of this enzyme. It means that at least under the specificinvitro conditions of constant DEX presence in the media, blockade of endogenous GSK3β activity was more impactful. It might be related with the finding that DEX injected to rats did not acutely affect expression of GC response elements (GREs) containing clock genes Per1, Per2, and Nr1d1 [70, 71], although it induced expression of Gilz. The result suggests that employment of the most common mechanism underlying the DEX effects on the circadian clocks may not be operating in the HPC. Therefore, the mechanism of how DEX entrains the HPC clock remains to be elucidated. It may either involve GRE-mediated induction of other genes, which may feedback on the clock mechanism, or multiple non-genomic GRE-independent pathways [72, 73]. Additionally, DEX- and GSK3β-related pathways may potentially compete. We confirmed absence of the DEX effect on acute induction of clock gene expression in samples of all 3 laser-dissected HPC regions. Previous studies using whole HPC homogenates reported conflicting results; corticosterone injections either caused approximately a 1.5-fold elevation of Per1 mRNA after 1 h [74] or, similar to our results, did not increase Per1 expression [75]. Interestingly, in contrast to HPC, Per1 mRNA in the choroid plexus adjacent to the HPC formation in the same animals responded to DEX injections acutely [40]. The dominance of the GSK3β-mediated over the GC-mediated signalling that we found in vitro confirms the effectiveness of the entrainment pathway for the HPC clock. It is intriguing because the enzyme has previously been demonstrated to play a crucial role in connecting the HPC clock and memory processing [34, 35, 76]. As already mentioned, unlike in conditions in vitro, GC levels in the body greatly vary over days and nights which favour a scenario of existence of 2 complementary mechanisms which may align the HPC clocks according to actual activity/arousal state and memory processes demands.

GC levels are under control of the central SCN clock, which uses them as a powerful signal to orchestrate most clocks within the body. The SCN-driven rhythm in GC levels facilitates rhythmicity of clocks in the HPC. GCs modulate the pace of the HPC clocks, aligning them with the activity/arousal state when the highest need for cognitive performance is expected. Additionally, the HPC clocks are tightly intertwined with neuronal processes (GSK3β activity) which are connected with memory formation and can contribute to entrainment of the HPC clock. Altogether, we show that the SCN-driven humoral signals and the local HPC-neuronal signals adjust the pace of the HPC clocks in order to align them with the time of actual need for higher performance of the brain structure. The mechanisms also suggest a missing link between disrupted daily activity rhythms, memory impairment, and mood disorders.

The authors thank Mrs. Eva Suchanová for her technical assistance, Prof. Jiří Pácha, and his colleagues for performing the surgery and Prof. Jihwan Myung for valuable discussion on the methodology.

All experiments were approved by the Animal Care and Use Committee of the Institute of Physiology and were in agreement with the Animal Protection Law of the Czech Republic (reference number 33/2019), as well as the European Community Council directives 86/609/EEC. All efforts were made to lessen the suffering of animals.

The authors have no competing financial interests to declare.

The study was supported by the Grant from Charles University 396218 (to K.L.), the OPPK BrainView CZ.2.16/3.1.00/21544, MEYS (LM2015062 Czech-BioImaging), and the Research Project RV0: 67985823. One part of the study was supported by the Czech Science Foundation grant 21-09745S (to A.S.)

Conceptualization: A.S.; study design: K.L., A.S., and M.S.; drafting the manuscript: K.L. and A.S.; experimental work: K.L., P.H., V.L., and N.S.; methodological expertize: M.S. and M.R.R.; final editing of the manuscript: K.L., A.S., M.S., and M.R.R.; approval of final version: all the authors.

All data generated or analysed during this study are included in this article and its online suppl. material files. Further enquiries can be directed to the corresponding author.

1.
Chaudhury
D
,
Colwell
CS
.
Circadian modulation of learning and memory in fear-conditioned mice
.
Behav Brain Res
.
2002 Jun
;
133
(
1
):
95
108
.
2.
Takahashi
JS
,
Hong
HK
,
Ko
CH
,
McDearmon
EL
.
The genetics of mammalian circadian order and disorder: implications for physiology and disease
.
Nat Rev Genet
.
2008
;
9
(
10
):
764
75
.
3.
Rawashdeh
O
,
Jilg
A
,
Maronde
E
,
Fahrenkrug
J
,
Stehle
JH
.
Period1 gates the circadian modulation of memory-relevant signaling in mouse hippocampus by regulating the nuclear shuttling of the CREB kinase pP90RSK
.
J Neurochem
.
2016 Sep
;
138
(
5
):
731
45
.
4.
Wardlaw
SM
,
Phan
TX
,
Saraf
A
,
Chen
X
,
Storm
DR
.
Genetic disruption of the core circadian clock impairs hippocampus-dependent memory
.
Learn Mem
.
2014
;
21
(
8
):
417
23
.
5.
Snider
KH
,
Dziema
H
,
Aten
S
,
Loeser
J
,
Norona
FE
,
Hoyt
K
,
.
Modulation of learning and memory by the targeted deletion of the circadian clock gene Bmal1 in forebrain circuits
.
Behav Brain Res
.
2016 Jul
;
308
:
222
35
.
6.
Amir
S
,
Harbour
VL
,
Robinson
B
.
Pinealectomy does not affect diurnal PER2 expression in the rat limbic forebrain
.
Neurosci Lett
.
2006
;
399
(
1–2
):
147
50
.
7.
Duncan
MJ
,
Prochot
JR
,
Cook
DH
,
Tyler Smith
J
,
Franklin
KM
.
Influence of aging on Bmal1 and Per2 expression in extra-SCN oscillators in hamster brain
.
Brain Res
.
2013
;
1491
:
44
53
.
8.
Lamont
EW
,
Robinson
B
,
Stewart
J
,
Amir
S
.
The central and basolateral nuclei of the amygdala exhibit opposite diurnal rhythms of expression of the clock protein Period2
.
Proc Natl Acad Sci U S A
.
2005
;
102
(
11
):
4180
4
.
9.
Gilhooley
MJ
,
Pinnock
SB
,
Herbert
J
.
Rhythmic expression of per1 in the dentate gyrus is suppressed by corticosterone: implications for neurogenesis
.
Neurosci Lett
.
2011
;
489
(
3
):
177
81
.
10.
Ikeno
T
,
Weil
ZM
,
Nelson
RJ
.
Photoperiod affects the diurnal rhythm of hippocampal neuronal morphology of siberian hamsters
.
Chronobiol Int
.
2013
;
30
(
9
):
1089
100
.
11.
Segall
LA
,
Perrin
JS
,
Walker
CD
,
Stewart
J
,
Amir
S
.
Glucocorticoid rhythms control the rhythm of expression of the clock protein, Period2, in oval nucleus of the bed nucleus of the stria terminalis and central nucleus of the amygdala in rats
.
Neuroscience
.
2006
;
140
(
3
):
753
7
.
12.
Segall
LA
,
Milet
A
,
Tronche
F
,
Amir
S
.
Brain glucocorticoid receptors are necessary for the rhythmic expression of the clock protein, PERIOD2, in the central extended amygdala in mice
.
Neurosci Lett
.
2009
;
457
(
1
):
58
60
.
13.
Valnegri
P
,
Khelfaoui
M
,
Dorseuil
O
,
Bassani
S
,
Lagneaux
C
,
Gianfelice
A
,
.
A circadian clock in hippocampus is regulated by interaction between oligophrenin-1 and Rev-erbα
.
Nat Neurosci
.
2011
;
14
(
10
):
1293
301
.
14.
Wang
LM
,
Dragich
JM
,
Kudo
T
,
Odom
IH
,
Welsh
DK
,
O'Dell
TJ
,
.
Expression of the circadian clock gene Period2 in the hippocampus: possible implications for synaptic plasticity and learned behaviour
.
ASN Neuro
.
2009
;
1
(
3
):
AN20090020
.
15.
Wyse
CA
,
Coogan
AN
.
Impact of aging on diurnal expression patterns of CLOCK and BMAL1 in the mouse brain
.
Brain Res
.
2010 Jun
;
1337
:
21
31
.
16.
Chaudhury
D
,
Loh
DH
,
Dragich
JM
,
Hagopian
A
,
Colwell
CS
.
Select cognitive deficits in vasoactive intestinal peptide deficient mice
.
BMC Neurosci
.
2008
;
9
:
1
12
.
17.
Albrecht
U
,
Sun
ZS
,
Eichele
G
,
Lee
CC
.
A differential response of two putative mammalian circadian regulators, mper1 and mper2, to light
.
Cell
.
1997
;
91
(
7
):
1055
64
.
18.
Borgs
L
,
Beukelaers
P
,
Vandenbosch
R
,
Nguyen
L
,
Moonen
G
,
Maquet
P
,
.
Period 2 regulates neural stem/progenitor cell proliferation in the adult hippocampus
.
BMC Neurosci
.
2009 Jan
;
10
(
1
):
30
.
19.
Mei
L
,
Fan
Y
,
Lv
X
,
Welsh
DK
,
Zhan
C
,
Zhang
EE
.
Long-term in vivo recording of circadian rhythms in brains of freely moving mice
.
Proc Natl Acad Sci U S A
.
2018 Apr
;
115
(
16
):
4276
81
.
20.
Shimizu
K
,
Kobayashi
Y
,
Nakatsuji
E
,
Yamazaki
M
,
Shimba
S
,
Sakimura
K
,
.
SCOP/PHLPP1β mediates circadian regulation of long-term recognition memory
.
Nat Commun
.
2016
;
7
:
1
12
.
21.
Jilg
A
,
Lesny
S
,
Peruzki
N
,
Schwegler
H
,
Selbach
O
,
Dehghani
F
,
.
Temporal dynamics of mouse hippocampal clock gene expression support memory processing
.
Hippocampus
.
2010
;
20
(
3
):
377
88
.
22.
Wyss
JM
,
Swanson
LW
,
Cowan
WM
.
A study of subcortical afferents to the hippocampal formation in the rat
.
Neuroscience
.
1979
;
4
(
4
):
463
76
.
23.
McEwen
BS
,
Weiss
JM
,
Schwartz
LS
.
Selective retention of corticosterone by limbic structures in rat brain
.
Nature
.
1968 Nov
;
220
(
5170
):
911
2
.
24.
Maggio
N
,
Segal
M
.
Striking variations in corticosteroid modulation of long-term potentiation along the septotemporal axis of the hippocampus
.
J Neurosci
.
2007
;
27
(
21
):
5757
65
.
25.
Schoenfeld
TJ
,
Gould
E
.
Differential effects of stress and glucocorticoids on adult neurogenesis
.
Curr Top Behav Neurosci
.
2013
;
15
:
139
64
.
26.
Cain
SW
,
Karatsoreos
I
,
Gautam
N
,
Konar
Y
,
Funk
D
,
McDonald
RJ
,
.
Blunted cortisol rhythm is associated with learning impairment in aged hamsters
.
Physiol Behav
.
2004
;
82
(
2–3
):
339
44
.
27.
McGaugh
JL
,
Roozendaal
B
.
Role of adrenal stress hormones in forming lasting memories in the brain
.
Curr Opin Neurobiol
.
2002
;
12
(
2
):
205
10
.
28.
De Quervain
D
,
Schwabe
L
,
Roozendaal
B
.
Stress, glucocorticoids and memory: Implications for treating fear-related disorders
.
Nat Rev Neurosci
.
2017
;
18
(
1
):
7
19
.
29.
Cheifetz
PN
.
The daily rhythm of the secretion of corticotrophin and corticosterone in rats and mice
.
J Endocrinol
.
1971 Mar
;
49
(
3
):
xi
xii
.
30.
Moore
RY
,
Eichler
VB
.
Loss of a circadian adrenal corticosterone rhythm following suprachiasmatic lesions in the rat
.
Brain Res
.
1972
;
42
(
1
):
201
6
.
31.
Spencer
RL
,
Chun
LE
,
Hartsock
MJ
,
Woodruff
ER
.
Glucocorticoid hormones are both a major circadian signal and major stress signal: how this shared signal contributes to a dynamic relationship between the circadian and stress systems
.
Front Neuroendocrinol
.
2018
;
49
(
October 2017
):
52
71
.
32.
Amir
S
,
Lamont
EW
,
Robinson
B
,
Stewart
J
.
A circadian rhythm in the expression of PERIOD2 protein reveals a novel SCN-controlled oscillator in the oval nucleus of the bed nucleus of the stria terminalis
.
J Neurosci
.
2004
;
24
(
4
):
781
90
.
33.
Malek
ZS
,
Sage
D
,
Pévet
P
,
Raison
S
.
Daily rhythm of tryptophan hydroxylase-2 messenger ribonucleic acid within raphe neurons is induced by corticoid daily surge and modulated by enhanced locomotor activity
.
Endocrinology
.
2007
;
148
(
11
):
5165
72
.
34.
Besing
RC
,
Rogers
CO
,
Paul
JR
,
Hablitz
LM
,
Johnson
RL
,
McMahon
LL
,
.
GSK3 activity regulates rhythms in hippocampal clock gene expression and synaptic plasticity
.
Hippocampus
.
2017 Aug
;
27
(
8
):
890
8
.
35.
Peineau
S
,
Bradley
C
,
Taghibiglou
C
,
Doherty
A
,
Bortolotto
ZA
,
Wang
YT
,
.
The role of GSK-3 in synaptic plasticity
.
Br J Pharmacol
.
2008
;
153
(
Suppl 1
):
428
37
.
36.
Peineau
S
,
Taghibiglou
C
,
Bradley
C
,
Wong
TP
,
Liu
L
,
Lu
J
,
.
LTP inhibits LTD in the hippocampus via regulation of GSK3beta
.
Neuron
.
2007
;
53
(
5
):
703
17
.
37.
Kinoshita
C
,
Miyazaki
K
,
Ishida
N
.
Chronic stress affects PERIOD2 expression through glycogen synthase kinase-3β phosphorylation in the central clock
.
Neuroreport
.
2012
;
23
(
2
):
98
102
.
38.
Dwivedi
Y
,
Pandey
GN
.
Administration of dexamethasone up-regulates protein kinase C activity and the expression of gamma and epsilon protein kinase C isozymes in the rat brain
.
J Neurochem
.
1999 Jan
;
72
(
1
):
380
7
.
39.
Soták
M
,
Bryndová
J
,
Ergang
P
,
Vagnerová
K
,
Kvapilová
P
,
Vodička
M
,
.
Peripheral circadian clocks are diversely affected by adrenalectomy
.
Chronobiol Int
.
2016
;
0528
(
April
):
1
10
.
40.
Liška
K
,
Sládek
M
,
Čečmanová
V
,
Sumová
A
.
Glucocorticoids reset circadian clock in choroid plexus via period genes
.
J Endocrinol
.
2020 Dec
.
41.
Sumová
A
,
Jáč
M
,
Sládek
M
,
Šauman
I
,
Illnerová
H
.
Clock gene daily profiles and their phase relationship in the rat suprachiasmatic nucleus are affected by photoperiod
.
J Biol Rhythms
.
2003
;
18
(
2
):
134
44
.
42.
Houdek
P
,
Sumová
A
.
In vivo initiation of clock gene expression rhythmicity in fetal rat suprachiasmatic nuclei
.
PLoS One
.
2014 Sep
;
9
(
9
):
e107360
.
43.
Sumová
A
,
Sládek
M
,
Jáč
M
,
Illnerová
H
.
The circadian rhythm of Per1 gene product in the rat suprachiasmatic nucleus and its modulation by seasonal changes in daylength
.
Brain Res
.
2002 Aug
;
947
(
2
):
260
70
.
44.
Herman
JP
,
Dolgas
CM
,
Marcinek
R
,
Langub
MC
.
Expression and glucocorticoid regulation of natriuretic peptide clearance receptor (NPR-C) mRNA in rat brain and choroid plexus
.
J Chem Neuroanat
.
1996
;
11
(
4
):
257
65
.
45.
Wakamatsu
H
,
Yoshinobu
Y
,
Aida
R
,
Moriya
T
,
Akiyama
M
,
Shibata
S
.
Restricted-feeding-induced anticipatory activity rhythm is associated with a phase-shift of the expression of mPer1 and mPer2 mRNA in the cerebral cortex and hippocampus but not in the suprachiasmatic nucleus of mice
.
Eur J Neurosci
.
2001
;
13
(
6
):
1190
6
.
46.
Debski
K
,
Ceglia
N
,
Ghestem
A
,
Ivanov
A
,
Brancati
GE
,
Broer
S
,
.
The circadian hippocampus and its reprogramming in epilepsy: impact for chronotherapeutics
.
DoiOrg
.
2017
:
199372
.
47.
Woodie
LN
,
Johnson
RM
,
Ahmed
B
,
Fowler
S
,
Haynes
W
,
Carmona
B
,
.
Western diet-induced obesity disrupts the diurnal rhythmicity of hippocampal core clock gene expression in a mouse model
.
Brain Behav Immun
.
2020
;
88
:
815
25
.
48.
Tahara
Y
,
Shiraishi
T
,
Kikuchi
Y
,
Haraguchi
A
,
Kuriki
D
,
Sasaki
H
,
.
Entrainment of the mouse circadian clock by sub-acute physical and psychological stress
.
Sci Rep
.
2015
;
5
:
11417
.
49.
Chun
LE
,
Woodruff
ER
,
Morton
S
,
Hinds
LR
,
Spencer
RL
.
Variations in phase and amplitude of rhythmic clock gene expression across prefrontal cortex, hippocampus, amygdala, and hypothalamic paraventricular and suprachiasmatic nuclei of male and female rats
.
J Biol Rhythms
.
2015
;
30
(
5
):
417
36
.
50.
Bering
T
,
Carstensen
MB
,
Wörtwein
G
,
Weikop
P
,
Rath
MF
.
The circadian oscillator of the cerebral cortex: molecular, biochemical and behavioral effects of deleting the Arntl clock gene in cortical neurons
.
Cereb Cortex
.
2017
:
1
14
.
51.
Ikeno
T
,
Yan
L
.
Chronic light exposure in the middle of the night disturbs the circadian system and emotional regulation
.
J Biol Rhythms
.
2016
;
31
(
4
):
352
64
.
52.
Verwey
M
,
Khoja
Z
,
Stewart
J
,
Amir
S
.
Differential regulation of the expression of Period2 protein in the limbic forebrain and dorsomedial hypothalamus by daily limited access to highly palatable food in food-deprived and free-fed rats
.
Neuroscience
.
2007
;
147
(
2
):
277
85
.
53.
Schnell
A
,
Chappuis
S
,
Schmutz
I
,
Brai
E
,
Ripperger
JA
,
Schaad
O
,
.
The nuclear receptor REV-ERBα regulates Fabp7 and modulates adult hippocampal neurogenesis
.
PLoS One
.
2014
;
9
(
6
):
e99883
14
.
54.
Bellanti
F
,
Iannelli
G
,
Blonda
M
,
Tamborra
R
,
Villani
R
,
Romano
A
,
.
Alterations of clock gene RNA expression in brain regions of a triple transgenic model of Alzheimer’s disease
.
J Alzheimers Dis
.
2017
;
59
(
2
):
615
31
.
55.
Ángeles-Castellanos
M
,
Mendoza
J
,
Escobar
C
.
Restricted feeding schedules phase shift daily rhythms of c-Fos and protein Per1 immunoreactivity in corticolimbic regions in rats
.
Neuroscience
.
2007
;
144
(
1
):
344
55
.
56.
Pantazopoulos
H
,
Dolatshad
H
,
Davis
FC
.
A fear-inducing odor alters PER2 and c-Fos expression in brain regions involved in fear memory
.
PLoS One
.
2011
;
6
(
5
):
e20658
.
57.
Navigatore-Fonzo
LS
,
Delgado
SM
,
Golini
RS
,
Anzulovich
AC
.
Circadian rhythms of locomotor activity and hippocampal clock genes expression are dampened in vitamin A-deficient rats
.
Nutr Res
.
2014
;
34
(
4
):
326
35
.
58.
Renaud
J
,
Dumont
F
,
Khelfaoui
M
,
Foisset
SR
,
Letourneur
F
,
Bienvenu
T
,
.
Identification of intellectual disability genes showing circadian clock-dependent expression in the mouse hippocampus
.
Neuroscience
.
2015
;
308
:
11
50
.
59.
Houdek
P
,
Nováková
M
,
Polidarová
L
,
Sládek
M
,
Sumová
A
.
Melatonin is a redundant entraining signal in the rat circadian system
.
Horm Behav
.
2016
;
83
(
May
):
1
5
.
60.
Murphy
EK
,
Spencer
RL
,
Sipe
KJ
,
Herman
JP
.
Decrements in nuclear glucocorticoid receptor (GR) protein levels and dna binding in aged rat hippocampus
.
Endocrinology
.
2002
;
143
(
4
):
1362
70
.
61.
Morimoto
M
,
Morita
N
,
Ozawa
H
,
Yokoyama
K
,
Kawata
M
.
Distribution of glucocorticoid receptor immunoreactivity and mRNA in the rat brain: an immunohistochemical and in situ hybridization study
.
Neurosci Res
.
1996
;
26
(
3
):
235
69
.
62.
Murphy
VE
,
Zakar
T
,
Smith
R
,
Giles
WB
,
Gibson
PG
,
Clifton
VL
.
Reduced 11beta-hydroxysteroid dehydrogenase type 2 activity is associated with decreased birth weight centile in pregnancies complicated by asthma
.
J Clin Endocrinol Metab
.
2002
;
87
(
4
):
1660
8
.
63.
Woodruff
ER
,
Chun
LE
,
Hinds
LR
,
Spencer
RL
.
Diurnal corticosterone presence and phase modulate clock gene expression in the male rat prefrontal cortex
.
Endocrinology
.
2016
;
157
(
4
):
1522
34
.
64.
Dana
RC
,
Martinez
JL
.
Effect of adrenalectomy on the circadian rhythm of LTP
.
Brain Res
.
1984 Aug
;
308
(
2
):
392
5
.
65.
Wyrwoll
CS
,
Holmes
MC
,
Seckl
JR
.
11β-hydroxysteroid dehydrogenases and the brain: from zero to hero, a decade of progress
.
Front Neuroendocrinol
.
2011
;
32
(
3
):
265
86
.
66.
Besing
RC
,
Paul
JR
,
Hablitz
LM
,
Rogers
CO
,
Johnson
RL
,
Young
ME
,
.
Circadian rhythmicity of active GSK3 isoforms modulates molecular clock gene rhythms in the suprachiasmatic nucleus
.
J Biol Rhythms
.
2015
;
30
(
2
):
155
60
.
67.
Spencer
RL
,
Chun
LE
,
Hartsock
MJ
,
Elizabeth
R
,
Robertspencercoloradoedu
RS
,
Chun
L
,
.
Glucocorticoid hormones are both a major circadian signal and major stress signal: how this shared signal contributes to a dynamic relationship between the circadian and stress systems
.
Front Neuroendocrinol
.
2017
.
68.
Balsalobre
A
,
Brown
SA
,
Marcacci
L
,
Tronche
F
,
Kellendonk
C
,
Reichardt
HM
,
.
Resetting of circadian time in peripheral tissues by glucocorticoid signaling
.
Science
.
2000
;
289
(
5488
):
2344
7
.
69.
Kiessling
S
,
Eichele
G
,
Oster
H
.
Adrenal glucocorticoids have a key role in circadian resynchronization in a mouse model of jet lag
.
J Clin Invest
.
2010
;
120
(
7
):
2600
9
.
70.
Torra
IP
,
Tsibulsky
V
,
Delaunay
F
,
Saladin
R
,
Laudet
V
,
Fruchart
JC
,
.
Circadian and glucocorticoid regulation of Rev-erbalpha expression in liver
.
Endocrinology
.
2000
;
141
(
10
):
3799
806
.
71.
So
AY-L
,
Bernal
TU
,
Pillsbury
ML
,
Yamamoto
KR
,
Feldman
BJ
.
Glucocorticoid regulation of the circadian clock modulates glucose homeostasis
.
Proc Natl Acad Sci
.
2009
;
106
(
41
):
2
7
.
72.
Groeneweg
FL
,
Karst
H
,
de Kloet
ER
,
Joëls
M
.
Mineralocorticoid and glucocorticoid receptors at the neuronal membrane, regulators of nongenomic corticosteroid signalling
.
Mol Cell Endocrinol
.
2012 Mar
;
350
(
2
):
299
309
.
73.
Kellendonk
C
,
Gass
P
,
Kretz
O
,
Schütz
G
,
Tronche
F
.
Corticosteroid receptors in the brain: gene targeting studies
.
Brain Res Bull
.
2002 Jan
;
57
(
1
):
73
83
.
74.
Conway-Campbell
BL
,
Sarabdjitsingh
RA
,
McKenna
MA
,
Pooley
JR
,
Kershaw
YM
,
Meijer
OC
,
.
Glucocorticoid ultradian rhythmicity directs cyclical gene pulsing of the clock gene period 1 in rat hippocampus
.
J Neuroendocrinol
.
2010
;
22
(
10
):
1093
100
.
75.
Bohacek
J
,
Manuella
F
,
Roszkowski
M
,
Mansuy
IM
.
Hippocampal gene expression induced by cold swim stress depends on sex and handling
.
Psychoneuroendocrinology
.
2015
;
52
:
1
12
.
76.
Hooper
C
,
Markevich
V
,
Plattner
F
,
Killick
R
,
Schofield
E
,
Engel
T
,
.
Glycogen synthase kinase-3 inhibition is integral to long-term potentiation
.
Eur J Neurosci
.
2007
;
25
(
1
):
81
6
.