Dynamic Foot Stimulations During Short-Term Hindlimb Unloading Prevent Dysregulation of the Neurotransmission in the Hippocampus of Rats

Abstract

Spaceflight and simulated microgravity both affect learning and memory, which are mostly controlled by the hippocampus. However, data about molecular alterations in the hippocampus in real or simulated microgravity conditions are limited. Adult Wistar rats were recruited in the experiments. Here we analyzed whether short-term simulated microgravity caused by 3-day hindlimb unloading (HU) will affect the glutamatergic and GABAergic systems of the hippocampus and how dynamic foot stimulation (DFS) to the plantar surface applied during HU can contribute in the regulation of hippocampus functioning. The results demonstrated a decreased expression of vesicular glutamate transporters 1 and 2 (VGLUT1/2) in the hippocampus after 3 days of HU, while glutamate decarboxylase 67 (GAD67) expression was not affected. HU also significantly induced Akt signaling and transcriptional factor CREB that are supposed to activate the neuroprotective mechanisms. On the other hand, DFS led to normalization of VGLUT1/2 expression and activity of Akt and CREB. Analysis of exocytosis proteins revealed the inhibition of SNAP-25, VAMP-2, and syntaxin 1 expression in DFS group proposing attenuation of excitatory neurotransmission. Thus, we revealed that short-term HU causes dysregulation of glutamatergic system of the hippocampus, but, at the same time, stimulates neuroprotective Akt-dependent mechanism. In addition, most importantly, we demonstrated positive effect of DFS on the hippocampus functioning that probably depends on the regulation of neurotransmitter exocytosis.

Electronic supplementary material

The online version of this article (10.1007/s10571-020-00922-2) contains supplementary material, which is available to authorized users.

Introduction

Despite the extensive study of effects of spaceflight and simulated microgravity on the human body, our knowledge about functional changes in the brain is still limited. MRI studies revealed various morphological changes in the astronauts’ brain after long-term space flight such as changed gray matter density and volume, increased ventricular volume, upward shift of the brain, and decreased functional connectivity (Demertzi et al. 2016; Roberts et al. 2017; Van Ombergen et al. 2017a, b). The same alterations were also observed in volunteers after head-down bed rest (HDBR), which is the most ground-based space analog (Van Ombergen et al. 2017a, b; Roberts et al. 2015, 2017; Demertzi et al. 2016; Koppelmans et al. 2017). Moreover, long-term spaceflight and HDBR can cause some cognitive alterations, like impairments in short-term memory, choice reaction time, attention switching, orientation, etc. (Casler and Cook 1999). The reported psychology problems can be related to altered hippocampal functioning caused by real or simulated microgravity. The hippocampus plays important role in general cognitive performance as well as in spatial navigation. Environmental changes and some pathology like epilepsy and neurodegenerative diseases can significantly affect the hippocampal microanatomy and functions (Dekeyzer et al. 2017; Anand and Dhikav 2012; Vecchio et al. 2018). However, the data about the effects of microgravity on the hippocampus are still limited. In humans, there is only one published work demonstrating the detected by MRI decreasing the gray matter volume in the hippocampus after HDBR lasted for 30 days (Li et al. 2015).

Space mission with the rodents aboard and simulated microgravity by hindlimb unloading (HU) have told us a bit more about the hippocampal neurochemistry, but these data are mostly systemless (Table 1). Analysis of these data revealed some similarity in neurochemical changes in the hippocampus both after space flight and HU. After 7 days in space and 14 days of HU, the activity of 5-HT was increased (Miller et al. 1989; Kokhan et al. 2017). However, 28 days of HU led to a decreasing of 5-HT in the hippocampus (Wang et al. 2017). Also, it was demonstrated that expression of Bcl-XL mRNA increased both after 30th day in space and 30th day of HU in the hippocampus (Kulikova et al. 2017; Naumenko et al. 2015). Also long-term space flight and HU led to attenuation of BDNF expression (Santucci et al. 2012; Yasuhara et al. 2007).

Table 1.

Molecular changes in the hippocampus of rodents caused by space flight or simulated microgravity

Days	Space flight	Simulated microgravity
7	↑5-HT1 receptors (Miller et al. 1989)	↑ Lipid peroxidation, ↑ ERK1/2 activity, ↑ NF-kB activity (Wise et al. 2005)
14	↓GFAP mRNA (Day et al. 1998)	↓ NR2A, ↓ NR2B, ↓ glutamate (Shang et al. 2017) ↓ BDNF (Yasuhara et al. 2007) ↑ 5-HT activity (Kokhan et al. 2017)
21		↓↑ Expression of presynaptic proteins, ↓ GABA, ↑ glutamate (Wang et al. 2015)
28		↑ GluR1, ↑ GluR4, ↑ glutamate, ↓ GABA, ↓ DA, ↓ 5-HT (Wang et al. 2017)
30	↑ Bcl-XL mRNA, ↓ BAX mRNA (Naumenko et al. 2015)	↑ Bcl-XL mRNA (Kulikova et al. 2017)
91	↓ BDNF, ↓ NGF (Santucci et al. 2012)

5-HT 5-hydroxytryptamine, GFAP glial fibrillary acidic protein, NR2A NMDA subtype 2A glutamate receptor, NR2B NMDA subtype 2B glutamate receptor, GluR1 AMPA glutamate receptor 1, GluR4 AMPA glutamate receptor 4, DA dopamine, BDNF brain-derived neurotrophic factor, NGF nerve growth factor, ↑ upregulation, ↓ downregulation, ↓↑ variable changes

It is known that excitatory glutamate neurotransmission in the hippocampus is critical for the main cognitive functions such as memory formation, learning, etc. (Robbins and Murphy 2006; Tamminga et al. 2012). And if some alteration in cognitive performance in astronauts was already documented (Casler and Cook 1999), there is not any study of glutamatergic system of the hippocampus of animals after space flight. Ground-based experiments demonstrated a decrease of the expression of NR2A and NR2B subunits of NMDA receptors and glutamate in the hippocampus after 14th day of hindlimb unloading (HU) (Shang et al. 2017). While after 21 and 28 days of HU, the expression of GluR1 and GluR4 subunits of AMPA receptors and level of glutamate were increased (Wang et al. 2015, 2017). These data supposed variable changing of glutamatergic system of the hippocampus that may depend on the duration of weightless. On the other hand, GABAergic system of the hippocampus is also tightly involved in the control of memory formation by regulation of excitation balance (Mann and Paulsen 2007; Schmitz et al. 2017). However, data about GABAergic system during or after weightless conditions are limited. There are only two published papers that demonstrated a decreasing GABA content in the hippocampus after long-term HU (Wang et al. 2015, 2017). A decrease in GABA and upregulation of the activity of the glutamatergic system indicate increased excitation, which finally can lead to excitotoxicity and the subsequent development of neurodegenerative processes.

One of the factors for the brain function impairment during real or simulated microgravity is a sensory deprivation and, additionally, withdrawal of the support afferentation may trigger the brain remodeling (Kozlovskaya et al. 1988). It was shown that dynamic foot stimulation (DFS) of the plantar surface during HU not only simulates the support afferentation but also prevents soleus muscle atrophy (Kyparos et al. 2005; Layne et al. 1998). However, the effects of DFS on the functional states of the hippocampus or other parts of the brain have not been studied yet. Here we explored how short-term hindlimb unloading, which simulates microgravity, and dynamic foot stimulation during hindlimb unloading could change the GABA or glutamatergic neurotransmitter systems of the hippocampus.

Methods

Animals

Thirty-six adult male Wistar rats (130–160 g) were recruited in the experiments, and no randomization was performed. The rats were obtained from the certified Nursery for laboratory animals of the Institute of Bioorganic Chemistry of the Russian Academy of Sciences (Pushchino, Moscow region) and housed in individual cages at 12/12 light–dark cycle with free water and food access. All procedures using animals were conducted in accordance with recommendations of the Federation for Laboratory Animal Science Associations and approved by the Biomedical Ethics Committee of the Institute of Biomedical Problems.

The rats were divided for 3 groups: C group (n = 12)—vivarium control, HU group (n = 12)—3-day hindlimb unloading (Morey-Holton and Globus 2002), and DFS group (n = 12)—4-h dynamic foot stimulation of the plantar surface each day during hindlimb unloading. DFS was provided using a model previously described by Kyparos et al. with modifications (Kyparos et al. 2005). The experimenter was not blinded to the groups. For HU group, a strip of elastic medical patch was applied to the tail of the animal. The patch was passed through a swivel that was attached to a metal rod at the top of the cage. The suspension angle was adjusted to approximately 30°, which prevents the hind limbs from touching the floor. This method allows the forelimbs to contact with the floor and the animal can freely move around the cage. For DFS group, a custom-built boot with a plastic plate was contacted at the sole of the foot, and inflation with a pressure at 104 mmHg was used to stimulate the sensory receptors in the soles of the rat’s foot in conscious animals during HU. The pressure stimulation protocol consisted of a 1-s inflation/1-s deflation of the air bladder for a total of 20 min followed by a 10-min rest interval. This cycle was repeated eight times over 4-h period during each day from 10 am to 2 pm of the 3-day HU period. We did not use any exclusion criteria and no rats died during the experiments. All suspended rats were monitored three times per day for signs of pain or suffering. The rats were anesthetized routinely with an intraperitoneal injection of tribromoethanol (240 mg/kg; # T48402, Sigma-Aldrich) and perfused with 4% paraformaldehyde for immunohistochemical study (n = 4 for each groups) or decapitated and the hippocampi were dissected for two parts and homogenized for Western blot assay (n = 8 for each groups) or collected in TRI Reagent (#T9424, Sigma) for RNA isolation. Animals were euthanized sequentially in groups starting from control, 4 rats from a group per day (Fig. 1). In total, two experiments were performed, one (n = 4 rats per group) to collect the brains for immunohistochemistry and another (n = 8 rats per group) to collect the hippocampi for Western blot assay and qRT-PCR.

Fig. 1.

Experimental design. The rats were divided for 3 groups: control, hindlimb unloading (HU), and dynamic foot stimulation during hindlimb unloading (DFS) (n = 12 rats per group). Experiments started at 8 am. DFS was applied 4 h per day from 10 am to 2 pm. After 3 days, the rat were sacrificed subsequently (4 rats per day) for immunohistochemical assay (n = 4 rats per group) and Western blot assay (n = 8 rats per group). Black arrow indicates collection of samples for endpoint assay

Quantitative Real-Time PCR

Total RNA was isolated from the hippocampi (n = 5 for each groups), using TRI Reagent (#T9424, Sigma) and first-strand cDNA synthesis was performed on 0.5 μg total RNA using Oligo (dT) primers and RevertAid First-Strand cDNA Synthesis Kit (#K1622, ThermoFisher Scientific). qPCR was performed on Applied Biosystems 7500 real-time PCR system (ThermoFisher Scientific) using a qPCRmix-HS SYBR (#PK156, Eurogen, Russia). qPCR was performed with the corresponding primers: GAD65 forward—5′-GTA CGC CAT GCT CAT TGC CC-3′, GAD65 reverse—5′-AGA GAG GAT CAA AAG CCC CG-3′; GAD67 forward—5′-GCT GGA AGG CAT GGA AGG TTT TA-3′, GAD67 reverse—5′-ACG GGT GCA ATT TCA TAT GTG AAC ATA-3′; VGLUT1 forward—5′-ACT GCA CTT GCT TTG GCC TC-3′, VGLUT1 reverse—5′-GGA TCC TCA CGA AGA TGA CA-3′; VGLUT2 forward—5′-CAG TGG GAT TCA GTG GAT TTG CT-3′, VGLUT2 reverse—5′-ACT GCC ATT CTT CAC GGG ACT TG-3′; and GAPDH forward—5′-TCC CTC AAG ATT GTC AGC AA-3′, GAPDH reverse—5′-AGA TCC ACA ACG GAT ACA TT-3′. Analysis was performed in Applied Biosystems 7500 software. Cycle thresholds were normalized against the reference gene GAPDH. Expression in control hippocampi was estimated as ‘1′. Relative fold expression of genes was calculated in Microsoft Excel by the 2-ΔΔCt method.

Western Blot Analysis

Each hippocampus was dissected from the brain and weighed, and then homogenized separately in 1.5 ml tubes by handheld homogenizer (#BT704, BT Lab Systems). Tissues were homogenized in 10 volumes of lysis buffer (20 mM Tris, pH 7.5; 1 mM EDTA; 1 mM EGTA; 150 mM NaCl; 1% Triton X-100) containing protease inhibitors (#4693116001, Sigma-Aldrich, St. Louis, MO, USA) and phosphatase inhibitor cocktail (#4906837001, Sigma-Aldrich). The total protein concentrations were determined by Bio-Rad protein assay (bovine serum albumin standard, #5000002, Bio-Rad Laboratories Inc., Hercules, CA, USA). Equal amounts of protein (15 µg per line) in sample buffer (#1610738, Bio-Rad) were denatured at 95 °C for 5 min and separated on 10% or 12% acrylamide gels using Mini-PROTEAN tetra Handcast System (Bio-Rad). The proteins from the gel were transferred to a nitrocellulose membrane (#sc-3718, Santa Cruz). The membranes were incubated in 3% non-fat milk in Tris buffer (0.1% Tween 20, 0.2 mM Tris, 137 mM NaCl) for 30 min and then incubated overnight with primary antibodies against pPTEN Ser380, pPDK1 Ser241, Akt, pAkt Thr308, and Ser473, GSK3b Ser9 (all from Phospho-Akt pathway antibody sampler kit, #9916, Cell Signaling); GAD67 (1:2000; #MAB5406, Millipore); VGLUT1 (1:500; #MAB5502, Millipore); VGLUT2 (1:500; #MAB5504, Millipore); pCREB (Ser133; 1:1000; #05–807, Millipore); p-synapsin I (Ser62/67; 1:1000; #AB9848, Millipore); synapsin I (1:1000; #AB1543P, Millipore); and p-PKA substrates (1:2000; #9624, Cell Signaling) or GAPDH (1:2000; # ab8245, Abcam). Subsequently, the membranes were incubated with secondary antibodies, anti-rabbit (1:80000; #A0545, Sigma-Aldrich), or anti-mouse (1:80000; #A9044, Sigma-Aldrich), followed by chemiluminescent detection by SuperSignal@West Dura Extended Duration Substrate (#34075, ThermoFisher Scientific). The signals were captured by ChemiDoc MP Imaging System (#12003154, Bio-Rad) and analyzed by using ImageJ software.

Immunohistochemistry

The cut sections (10 μm) containing the hippocampus were incubated overnight at room temperature with primary antibodies against pCREB (Ser133; 1:600; #05-807, Millipore); mixture of VGLUT2 (1: 200; #MAB5504, Abcam) and VGLUT1 (1:400; #MAB5502, Millipore); or GAD67 (1:600; #MAB5406, Millipore). Then the sections were incubated with biotinylated secondary anti-rabbit antibody (1:600; #BA1000, Vector Laboratories) or anti-mouse antibody (1:600; #BA-2020, Vector Laboratories) and subsequently incubated with streptavidin-peroxidase (1:800; #SA202, Sigma-Aldrich). The peroxidase reaction was revealed in the PBS buffer containing 3,3′ diaminobenzidine (0.05%; #D8001, Sigma-Aldrich) and hydrogen peroxide (0.01%; #H3410, Sigma-Aldrich). Additionally, reactions lacking primary antibodies were done to ensure the specificity of the observed staining.

Evaluation of Sections and Statistical Analysis

Calculation of sample size was not done, but 4 rats per group were taken for immunohistochemical assay and 8 rats per group for Western blot assay to achieve statistical significance using non-parametric Kruskal–Wallis test.

The sections of control and experimental groups were processed for immunostaining simultaneously that allowed semi-quantitative analysis of protein amount in the histological slices by measurement of optical density (Taylor and Levenson 2006). Relative optical density of immunopositive substances for each individual hippocampus was estimated using ImageJ software by analysis of five sections of the same zones of the hippocampus for each animal (n = 4 for each group). The images were taken with Leica AF7000 microscope.

Statistical analysis was done by a non-parametric Kruskal–Wallis test using GrafPad Prism 8.2.1. Values were considered statistically significant at p < 0.05. In all graphs, values are expressed as median with interquartile range for both immunohistochemistry and Western blot analysis.

Results

Dynamic Foot Stimulation During Hindlimb Unloading Prevents Alterations of Glutamatergic Systems of the Hippocampus

Firstly, we analyzed the hippocampal glutamatergic cells since these cells present the main type of cells in the hippocampus. As a marker for glutamatergic neurotransmission, we used vesicular glutamate transporters 1 and 2 (VGLUT1/2). It is well known that expression of VUGLTs tightly correlates with glutamate production and secretion (Hackett and Ueda 2015). We processed the hippocampus of the experimental and control rats for immunohistochemical assay of VGLUT1/2 expression (Fig. 2a–f). This method allows revealing a localization of the positive cells and a measurement of immunoreactivity density. Compared with control group, VGLUT1/2 in the hippocampus of HU rats demonstrated significant reduction of immunopositive reaction in the molecular layer of the dentate gyrus (Fig. 2c, g; p = 0.0243),in suprapyramidal blade (Fig. 2c, h; p = 0.0324) containing mossy fiber projections from granular cells to CA3,and in the stratum radiatum of CA3 (Fig. 2f, i; p = 0.0427);however, DFS reduced this effect (Fig. 2b, f, g–i). There were not any significant differences in the expression of VGLT1/2 between HU and DFS groups in all studied zones of the hippocampus.

Fig. 2.

Dynamic foot stimulation during hindlimb unloading prevents alterations of glutamatergic systems of the hippocampus. a–f VLUT1/2 immunostaining of the dentate gyrus CA3 zone of the hippocampus of control rats (a, d), the rats of DFS group (b, e) and after 3-day HU (c, f). SR stratum radiatum, DGg granular layer of the dentate gyrus, DGm molecular layer of the dentate gyrus, SPB suprapyramidal blade. Scale bar – 200 μm. g–i analysis of the relative optical density of VGLUT1/2 immunopositive signal in the molecular layer of the dentate gyrus (g), in the suprapyramidal blade (h) containing mossy fiber projections from granular cells to the stratum lucidum, and in the stratum lucidum (i). Five sections of the hippocampus from each rats were analyzed (n = 4 rats per group). Optical density presented in arbitrary units. Axis y: optical density (arbitrary units). Data are shown as median with interquartile range. *p ˂ 0.05 versus control. j, k Western blot analysis of VGLUT1/2 expression: j representative images of Western blots of VGLUT1/2 and GAPDH (n = 4 rats per group); k analysis of VGLUT1/2 expression in the hippocampus by calculation of the ration between VGLUT1/2 and GAPDH (n = 8 animals per group). Data are shown as median with interquartile range. *p ˂ 0.05 vs control. l qRT-PCR analysis demonstrated that expression of VGLUT1 mRNA was the same in all groups, while VGLUT2 mRNA levels in the hippocampus of HU rats was significantly low. n = 5 for each group. Data are shown as median with interquartile range. *p ˂ 0.05 vs control. c control, DFS the rats with dynamic foot stimulation during 3-day hindlimb unloading, HU 3-day hindlimb unloading

Western blot results also revealed significant decrease in VGLUT1/2 in the hippocampus of HU group in comparison with both control and DFS groups (Fig. 2j, k; HU vs control p = 0.0157; HU vs DFS p = 0.014).

Analysis of qRT-PCR data did not demonstrate any difference among all groups for VGLUT1 mRNA,, but expression of VGLUT2 mRNA was significantly low in HU group in comparison with control. DFS group was neither different from control nor from HU (Fig. 2l; HU vs control p = 0.0175).

Neither Hindlimb Unloading nor Dynamic Foot Stimulation Altered GABA Expression in the Hippocampus

Estimation of GABA production was done by analysis of glutamic acid decarboxylases GAD67, which regulates GABA synthesis (Erlander and Tobin 1991). We made the immunohistochemical assay of GAD67 expression in the hippocampus of rats and did not reveal any changes in the GAD67-positive cell distribution between all groups (Online Resource 1a–c). Western blot analysis and qRT-PCR also did not show alteration in GAD67 expression (Online Resource 1d–f). Thus, our data demonstrated that neither short-term HU nor DFS applied during HU affected the GAD67 expression in the hippocampus.

Dynamic Foot Stimulation Alters the Exocytosis Machinery

Here, we investigated some features of the exocytosis machinery such as phosphorylation of synapsin I at Ser62/67 and expression of the SNARE complex proteins (Fig. 3; Online Resource 2). Neurotransmitter exocytosis is mainly regulated by the proteins of SNARE protein complex in cooperation with synapsins (Song and Augustine 2015). SNARE complex is formed by proteins such as SNAP-25 (Synaptosomal Nerve-Associated Protein 25), VAMP-2 (Vesicular-Associated Membrane Protein 2), and syntaxin 1 (Sudhof 2013). Synapsins tightly participate in synaptic vesicle exocytosis being in phosphorylated, means activated, forms (Fdez and Hilfiker 2006). It was demonstrated that in the hippocampus synapsins regulate the release of glutamate and GABA (Song and Augustine 2015, 2016). Our data demonstrated significant increase in phosphorylation of synapsin I in the hippocampus of HU rats (Fig. 3a, b; p = 0.0144). As for the SNARE complex proteins, an intriguing decreasing of SNAP-25 (Fig. 3a, c; p = 0.0175) in comparison with HU, and VAMP-2 (Fig. 3a, d; p = 0.0043), and syntaxin 1 (Fig. 3a, e; p = 0.0472) in comparison with control was demonstrated in the hippocampus.

Fig. 3.

Dynamic foot stimulation alters the exocytosis machinery. a representative images of Western blots of p-Synapsin I Ser62/67, synapsin I, SNAP-25, VAMP-2, and syntaxin 1 (n = 5 rats per group). b analysis of Synapsyn I phosphorylation at Ser62/67 by calculation of the ratio of pSynapsyn I to total synapsin I, c–e analysis of the expression of SNAP-25 (c), VAMP-2 (d), and syntaxin 1 (e) was done by calculation of the ratio of optical density of SNAP-25, VAMP-2, or syntaxin 1 to the optical density of Ponceau S staining, which is presented in electronic supplementary material (Online Recourse 2) (Bliznyuk et al. 2019). c control, DFS the rats with dynamic foot stimulation during 3-day hindlimb unloading, HU 3-day hindlimb unloading. n = 8 rats per group. Data are shown as median with interquartile range. *p ˂ 0.05 vs control. #p < 0.05 DFS versus HS

Akt-Dependent Signaling is Activated After 3-Day Hindlimb Unloading

We focused on analyzing the Akt signaling pathway since it regulates cell survival, cell cycle progression, protein synthesis, and synaptic vesicle recycling (Smillie and Cousin 2012; Hemmings and Restuccia 2012). Phosphatase and tensin homolog (PTEN) phosphatase is one of the main regulators that control balance between active/inactive forms of Phosphoinositide 3-kinase (PI3K) by dephosphorylation of Phosphatidylinositol (3,4,5)-trisphosphate (PIP3). Being phosphorylated, PTEN is inactive and Akt cascade can be continuously activated (Manning and Toker 2017). We started the analysis of Akt-dependent signaling with estimation of phosphorylation of PTEN. While in DFS rats the activity of PTEN in the hippocampus was attenuated (Fig. 4a, b; p = 0.0327) in relation to both control and HU, in the hippocampus of rats after 3-day HU phosphorylation of PTEN was not altered (Fig. 4a, b; p = 0.327) in comparison to control, but was significantly increased compared to DFS (Fig. 4a, b; p < 0.0001). The canonical Akt pathway starting with PIP3 activation combines phosphorylation of phosphoinositide-dependent protein kinase 1 (PDK1) that in turns phosphorylates Akt at Ser308 and finally active Akt phosphorylates glycogen synthase kinase 3b (GSK3b) at Ser9 (Manning and Toker 2017). Indeed, after 3-day HU, our results demonstrated sequential increasing of phosphorylation of PTEN at Ser380 (Fig. 4a, b; p < 0.0001 HU vs DFS), PDK at Ser241 (Fig. 4a, c; p = 0.0027), Akt at Thr308 (Fig. 4a, d; p = 0.0341), and GSK3b at Ser9 (Fig. 4a, f; p = 0.0003). While in DFS group we did not observesuch significant changes in the activity of the members of PDK/Akt cascade in the hippocampus (Fig. 4a, c–f), but phosphorylation of GSK3b was increased (Fig. 4a, f; p = 0.0327.) We also analyzed phosphorylation of Akt at Ser473, which is mTOR dependent, but GSK3b is a target for both Akt Ser308 and Akt Ser473 (Bayascas and Alessi 2005). We showed that DFS animals demonstrated increasing phosphorylation of Akt at Ser473 in comparison with control (Fig. 4a, e; p = 0.0002), but in HU group it was not altered (Fig. 4a, e).

Fig. 4.

Akt-dependent signaling is activated after 3-day hindlimb unloading. a representative images of Western blots of pPTEN, pPDK1, pAkt Thr308, pAkt Ser473, and GSK3b Ser9 and corresponding GAPDH and total Akt and GSK3b. b–f quantification of phosphorylation of PTEN at Ser380 (b), PDK1 at Ser241 (c), Akt at Thr308, and Ser473 (d, e) and GSK3b at ser9 (f) from the blots. Phosphorylation of PTEN and PDK1 was calculated as ratio between its phosphorylated forms to GAPDH; total Akt and GSK3 were used as standards for pAkt and pGSK3b. c control, DFS the rats with dynamic foot stimulation during 3-day hindlimb unloading, HU 3-day hindlimb unloading. Data are shown as median with interquartile range. *p ˂ 0.05 vs control. **p ˂ 0.01 vs control. ***p ˂ 0.001 vs control. ####p < 0.0001 DFS vs HU. n = 8 (number of animals) for each group

PKA Activity do not Change After Hindlimb Unloading or Dynamic Foot Stimulation

The third major player in the regulation of neuronal functioning is protein kinase A (Seino and Shibasaki 2005; Waltereit and Weller 2003; Dagda and Das Banerjee 2015). We analyzed activity of PKA using phospho-PKA substrate, which detects the major PKA substrates in active (phosphorylated) forms. The results demonstrated that in both HU and DFS groups, the activity of PKA was not changed in comparison with control (Online Resource 3).

Dynamic Foot Stimulation Prevents Over Activation of CREB in the Dentate Gyrus

We also analyzed transcription factor CREB (cyclic AMP response element-binding protein), whose activity is mainly regulated by Akt and PKA (Lonze and Ginty 2002). Our result revealed unchanged activity of PKA (Suppl. 3) and increase in Akt activity after 3 day HU (Fig. 4). Indeed, pronounced activation of Akt activity in HU group was accompanied with elevated CREB phosphorylation (Fig. 5d, e), where immunohistochemical analysis revealed dramatic increase in the number of pCREB-positive nuclei in the granular cell of the Dentate Gyrus and in the pyramidal cells of CA3 and CA1 in the hippocampus of rats after 3 day HU (Fig. 5a, c). In control hippocampi and in the hippocampus of DFS animals, the pCREB immunopositive nuclei were mainly localized in the subgranular zone of the dentate gyrus (Fig. 5a, b).

Fig. 5.

Dynamic foot stimulation prevents over activation of CREB in the dentate gyrus. a–c pCREB immunostaining of the hippocampus of control rats (a), rats of DFS group (b), and HU group (c). The right panels present the selected zones (mark by rectangles) of CA1 and granular layer of the dentate gyrus (DGg). d–f Western blot analysis of CREB phosphorylation. d pCREB blot; e Ponceau S staining used as loading control; f quantification of CREB phosphorylation as ration pCREB/Ponceau S. n = 8 (number of animals) for each group. c control, DFS the rats with dynamic foot stimulation during 3-day hindlimb unloading, HU 3-day hindlimb unloading. Data are shown as median with interquartile range. **p ˂ 0.01 vs control

Discussion

Existing data on the effect of microgravity on the hippocampus are extremely limited (Table. 1). And nothing is known about the effects of dynamic foot stimulation of the plantar surface in microgravity condition on the brain in general and onto the hippocampus in particular. Our results demonstrated that after 3-day HU the expression of VGLUT1/2 was significantly decreased. Shang with co-authors also shown a decrease in glutamate in the hippocampus after 14 day HU (Shang et al. 2017), while Wang and co-authors revealed upregulation of glutamate content after 21 and 28 day HU (Wang et al. 2017, 2015). So we can conclude that the production of glutamate in the hippocampus depends on the duration of HU. VGLUTs control all steps of glutamatergic transmission, i.e., loading of glutamate into synaptic vesicular (SV), cargo SV to the presynaptic terminals, release of glutamate and recycling of SV (Santos et al. 2009; Hackett and Ueda 2015), and neurotransmitter synthesis directly correlates with vesicle loading (Wilson et al. 2005; Jin et al. 2003). Thus, our data revealed a depletion of glutamatergic system of the hippocampus caused by short-term HU. On the other hand, it is important to mention that DFS during HU completely prevented the disturbance of glutamatergic cells in the hippocampus.

It could be supposed that such attenuation of glutamatergic neurotransmission during HU may be caused by stress. It was showed a decrease of glutamate in the hippocampus after chronic stress, in opposite acute stress that induced glutamate production (Popoli et al. 2011). Human studies of the effects of space flight revealed stress reaction only on launch and landing, but did not demonstrated significant impact of actual space flight (Grigoriev et al. 1997; Borchers et al. 2002). Ground base experiments demonstrated that in human prolong HDBR caused chronical stress and elevation of stress-related hormones (Rai and Kaur 2011). However, in rodents HU caused elevation of corticosterone level in the blood only up to 2nd–3rd day and then returned to control (Feng et al. 2016; Morey-Holton and Globus 2002). Moreover, in our hands we did not observe any induction of corticosterone secretion in rats after 1 day HU (Tyganov et al. 2019). Thereby we cannot exclude the effect of stress on glutamatergic cells of the hippocampus, but in the other side, an unchanged VGLUTs expression in animals with DFS can be used as an argument of no stress at 3 day HU, or we can also suppose prevention of stress by DFS.

Here we revealed that simulated microgravity affects the glutamatergic system of the hippocampus, but, surprisingly, we did not find any alteration in GABA production related to GAD67 expression. There are two different roles for GAD65 and GAD67. GAD65 functions locally being associated with synaptic vesicles and controls transport of newly synthesized GABA into synaptic vesicles (Jin et al. 2003). GAD67 regulates cytosolic synthesis of GABA, which can then be used for packaging into synaptic vesicles and plays a role as a neurotransmitter, and also can be an energy source through metabolism via the GABA shunt pathway (Hertz and Chen 2017). The role of GAD67 was also demonstrated by Asada with co-authors in studies of GAD65 and GAD67 knockout mice (Asada et al. 1996, 1997). While GAD65 deficiency did not significantly affect mouse development and GABA production (Asada et al. 1996), the GAD67 knockout was lethal due to respiratory failure and cleft palate supposing significant role of GAD67 in the regulation of non-neuronal tissue development (Asada et al. 1997). Interesting, while the brain of GAD67 knockout mice looked normal, GABA was almost eliminated in the brain (Asada et al. 1997). So, we concluded that the unchanged GAD67 expression can demonstrate both stable energy metabolism and GABA production.

We also analyzed exocytosis machinery be checking of expression of SNARE proteins and an activity of synapsin I, as these proteins regulate release of the neurotransmitters including glutamate. Here we observed unchanged expression level of VAMP-2, syntaxin and SNAP-25 in the hippocampus of HU rats, while DFS led to inhibition of SNARE protein expression. At the same time phosphorylation of synapsin I in the hippocampus of HU rats was significantly increased. Phosphorylation of synapsin I at Ser62/67 can argue for both increased and decreased exocytosis. On one hand phosphorylation of synapsin I at Ser62/67 leads to dissociation of synapsin I from actin filaments and thus activates vesicle exocytosis dependent from activation of TrkB receptors (Jovanovic et al. 1996, 2000). On the other hand, stimulation of Ca2 + dependent exocytosis leads to calcineurin-dependent de-phosphorylation of synapsin I at Ser62/67 (Jovanovic et al. 2001). However, decreased expression of SNARE proteins associates with attenuation of exocytosis. It was demonstrated that in pancreatic islands of patients with type 2 diabetes and diabetic rats the secretion of insulin was correlated with decrease expression of SNAP-25, VAMP-2 and syntaxin 1 (Ostenson et al. 2006; Nagamatsu et al. 1999). In neurons with partial deletion of SNARE-binding protein munk18-1 secretion of BDNF was diminished as well as BDNF heterozygous mice with low secretion of BDNF were also characterized by decrease expression of SNAP-25 (Abidin et al. 2018; Lee et al. 2019). Important data were obtained by Biswal with co-authors demonstrated that inhibition of SNAP-25 expression during hypoxia has a protective effect against glutamate excitotoxicity in the hippocampus (Biswal et al. 2017). We supposed that attenuation of neurotransmitter exocytosis in the hippocampus of DFS rats can play a protective role to prevent over excitation.

Disturbance in glutamatergic system of the hippocampus during short-term HU, that presumably caused by stress at the beginning of the experiment, can be associated with alteration in the activity of intracellular signaling that play a role in the cell survival like Akt or PKA dependent pathways (Hemmings and Restuccia 2012; Hetman and Gozdz 2004; Waltereit and Weller 2003). Indeed, our data demonstrated the increasing of Akt signaling activity, while PKA activity was not changed. As a result of Akt activation we observed inhibition of GSK3b by its phosphorylation at Ser9. GSK3 regulates many intracellular evens including survival, metabolism, growth and proliferation in negative order by degradation or inhibition of the substrates (Kaidanovich-Beilin and Woodgett 2011; Manning and Toker 2017). Phosphorylation of GSK3b by Akt at Ser9 inhibits GSK3 thus stabilizing of GSK3 targeted proteins (Manning and Toker 2017). Revealing of inactivation of GSK3b allowed us supposed that activation of Akt signaling during HU were mainly connected with activation of cell survival mechanisms.

The other target of Akt is a transcriptional factor CREB (Lonze and Ginty 2002). CREB controls transcription of hundreds of different genes and is activated by acute and chronic stress in the hippocampus (Pahlevani et al. 2014; Boer et al. 2007). However, others, in opposite, demonstrated a decrease of CREB activation after chronic stress exposure (Gronli et al. 2006) or unchanged activity of CREB after acute stress in the hippocampus (Miller et al. 2007). It is interesting that newly published work demonstrated strong resilience effect to stress in mice with CREB deletion in the hippocampus (Manners et al. 2019). In our experiments we observed a dramatically increased CREB phosphorylation in the dentate gyrus as well as in CA3 and CA1 of the hippocampus after 3-day HU in comparison with control rats, but not in hippocampus of DFS group. Our data that demonstrated exhaustion of glutamatergic cell, significant upregulation of CREB and Akt activity in the hippocampus of rats after shot-term HU supposed a kind of neuronal disturbance or stress caused by loss of support afferentation, fluid shift or antiorthostatic hypokinesia. On the other hand, stimulation of plantar surface prevented these alterations in the hippocampus and demonstrated positive effect of simulation of support afferentation in the regulation of neuronal functioning.

As a result, we can conclude that short-term HU causes dysregulation of glutamatergic system of the hippocampus, but, at the same time, stimulates neuroprotective Akt-dependent mechanism. In addition, and most importantly, we demonstrated positive effect of DFS on the hippocampus that probably depends on the regulation of neurotransmitter exocytosis.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author Contributions

ASB performed experiments, analyzed the data, and revising manuscript. SAT and NSM performed animal experiments. SDN and AAN performed immunohistochemical staining and data analysis. BSS designed the experiments and revised manuscript. MVG conceived the study, designed the experiments, and wrote the manuscript. All authors discussed and commented on the manuscript.

Funding

This study was supported by grants from RFBR 20-015-00062 (MVG), 17-29-01029-ofi_m (BSS), and 17-29-01034-ofi_m (NSM) and by Russian Government program.

Data Availability

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Compliance with Ethical Standards

Conflict of interest

The authors declare no conflict of interest.