Whole-body vibration elicits 40 Hz cortical gamma oscillations and ameliorates age-related cognitive impairment through hippocampal astrocyte synapses in male rats

Abstract

Age-related cognitive impairment is a prevalent issue in developed societies. Gamma oscil2lations at 40 Hz have been identified as a potential therapeutic approach for age-related cognitive decline and can be induced through various modalities, including auditory, visual, electrical, and magnetic stimulation. In this study, we investigated a novel modality of stimulation: whole-body vibration at 40 Hz. We examined the effects of 40 Hz vibration on cognitive performance and associated neuronal activity in the brains of aged male rats. Our findings revealed that only vibration at 40 Hz, rather than 20 Hz or 80 Hz, elicited cortical gamma oscillations in aged male rats. Additionally, following 8 weeks of prolonged treatment, the implementation of 40 Hz whole-body vibration significantly augmented the cognitive function of aged male rats as evidenced by behavioral assessments. Mechanistic studies demonstrated that these beneficial effects were attributed to the reduction of neuronal apoptosis in hippocampal CA1 through regulation of synaptic connections between astrocytes and neurons via 40 Hz gamma oscillations. Collectively, this suggests a promising intervention for age-related cognitive decline and identifies neuron-astrocyte synapses as potential therapeutic targets.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10522-024-10154-2.

Introduction

With the significant rise in global life expectancy (Flatt and Partridge 2018), age-related cognitive impairment (ARCI) has emerged as a prominent societal challenge (Langa and Levine 2014; Keshavarz et al. 2023). ARCI manifests in various forms of dementia, ranging from mild cognitive impairment (MCI) to Alzheimer’s disease (AD), depending on its severity (Bowen et al. 1997). Previous research indicates that aging is the primary risk factor for ARCI, accelerating the natural decline of cognitive function and increasing susceptibility to neurodegenerative diseases (Gonzales et al. 2022). However, there is a dearth of effective treatments for ARCI; most medications merely provide temporary relief for certain symptoms without preventing or reversing cognitive decline (Anderson 2019). Consequently, it becomes imperative and pressing to focus on prevention and delaying the onset and progression of ARCI.

Gamma (γ) oscillations, being the most rapid neural oscillations in the human brain (Başar et al. 2013), are generated by a large network of synchronized neurons firing electrical impulses (Hughes 2008). These oscillations have been closely linked to learning and memory processes (Miller et al. 2018; Griffiths and Jensen 2023). Iaccarino et al. (2016) demonstrated that stimulating fast-spiking parvalbumin-positive (FS-PV) interneurons at a frequency of 40 Hz in the hippocampal CA1 region induced gamma oscillations and effectively ameliorated cognitive impairment in mice with Alzheimer’s disease (Sohal et al. 2009; Iaccarino et al. 2016). Interestingly, this improvement was not observed when other cell types were stimulated at 40 Hz or when PV interneurons were driven at frequencies different from 40 Hz (Iaccarino et al. 2016). Since then, various protocols involving 40 Hz stimulation such as light stimulation(Adaikkan et al. 2019), sound stimulation (Martorell et al. 2019), electrical stimulation (Benussi et al. 2021), and magnetic stimulation (Guo et al. 2021) have been widely employed in non-invasive studies targeting the treatment for cognitive impairment. In this study, we investigated a novel form of intervention: whole-body vibration (WBV) at 40 Hz. Whole-body vibration, as an emerging rehabilitation modality, has not only demonstrated motor rehabilitation benefits but also shown potential in alleviating cognitive decline in recent studies(Oroszi et al. 2021; Asahina et al. 2023). Therefore, 40 Hz WBV could be a more suitable intervention with fewer adverse effects for elderly individuals with impaired vision, hearing and mobility. Considering the unique relationship between WBV as an exercise paradigm and the primary motor cortex (M1), along with recent studies demonstrating a reduction in frontal gamma activity among patients with Alzheimer's Disease (Casula et al. 2022), we have chosen to focus on detecting gamma oscillations at the M1 site of aged male rats.

Astrocytes, a widely distributed type of glial cell in the brain, exert influence on both short- and long-term synaptic plasticity by releasing glutamate and D-serine, which induce structural and functional changes at neighboring synapses (Perea and Araque 2007; Henneberger et al. 2010). These findings indicate that astrocytes are actively involved in the transfer and storage of synaptic information by releasing glutamate and D-serine. The impact of these changes on behaviorally relevant responses is gradually being monitored (Martin-Fernandez et al. 2017; Adamsky et al. 2018). Research has demonstrated that, astrocytes play a pivotal role in mediating information transmission between PV interneurons and neurons through receiving gabaergic neurotransmission, which directly affect animals' cognitive function (Mederos et al. 2021). The PV interneurons, responsible for generating gamma oscillations, also play a crucial role as inhibitory neurons (Sohal et al. 2009). Disruption of inhibitory input in astrocytes not only affects their morphology (Cheng et al. 2023), but also results in the loss of morphological complexity and circuit dysfunction across various brain regions (Mederos et al. 2021; Cheng et al. 2023). Therefore, understanding the morphological changes in astrocytes in our study is essential for comprehending how 40 Hz gamma oscillations can alleviate age-related cognitive impairment. However, the precise mechanisms underlying how astrocytes interact with neurons to modulate cognitive function in aged rats remain unclear. In this study, we employed 40 Hz WBV to investigate its effects on ARCI, observe alterations in astrocyte activity during this process, and analyze the specific contributions of astrocytes to ARCI.

Overall, our study aimed to assess the therapeutic efficacy of 40 Hz WBV through behavioral and cellular analyses conducted on aged male rats while also examining the distinct involvement of astrocytes throughout this process.

Materials and methods

Animals

Male Sprague–Dawley (SD) rats were utilized in this study. It is widely acknowledged that the behavioral performance of female animals may be influenced by physiological cycles, leading to increased variability and introducing additional complexity in research studies. Therefore, for the purpose of seeking more stable and reliable results, we have restricted our subjects to male animals in this study. The aged rats were obtained from Jinan Pengyue Experimental Animal Breeding Co., Ltd. They underwent a minimum acclimation period of 6 months in our animal facility prior to the experiment. At the start of the experiment, the aged rats were between 18 and 20 months old, while the young rats were 2 months old. The aged rats were randomly divided into two treatment groups (40 Hz WBV and 80 Hz WBV) and one control group (sham WBV). The sham WBV group was exposed to identical environmental conditions, including replacement of the vibration plate and simulated movement sounds, but without actual vibration. Simultaneously, electrocorticogram (ECoG) recordings were conducted on these male SD rats. To prevent discomfort caused by social isolation, animals were housed in pairs or trios within their own home cages. Standard laboratory conditions included a 12/12 dark/light cycle with lights on at 9:00 a.m., temperature maintained at 22 ± 2 °C, and humidity levels set at 50 ± 10%. Food and water access was provided ad libitum throughout the study duration. All experimental procedures received approval from Guilin Medical College's Animal Ethics Committee (approval number: GLMC202105081) and adhered to guidelines outlined by European Communities Council Directive 2010/63/EU. Researchers monitored daily health status of animals as well as weekly body weight changes during the course of experimentation.

Surgery and electrocorticogram (ECoG) recordings

The aged male rats were deeply anesthetized with sodium pentobarbital (35 mg/kg, intraperitoneal injection), immobilized on a stereotaxic apparatus, and the dorsal surface of the skull was exposed and cleansed. A stainless steel screw with a diameter of 0.5 mm was implanted into the skull to cover the right frontal cortical area (AP + 2.00 mm, ML + 2.25 mm) for recording ECoG signals from the M1 cortex (Fig. 1A) (Yang et al. 2015). The reference electrode and ground electrode were positioned above the cerebellum within the skull, while dental cement was used to securely fixate all implants onto the skull. Following surgery, antibiotics were administered to promote recovery in separate cages for a two-week period before commencing recordings. The ECoG signals of these aged rats were recorded under completely isolated electromagnetic conditions at vibration frequencies of 20, 40, 80 Hz as well as during resting state. Recording equipment utilized was model number 8401-HS (Pinnacle Technology, Kansas, US), employing a band-pass filter ranging from 1 Hz to 1 kHz and sampling rate set at 2 kHz. To minimize circadian variations in arousal levels during data collection sessions scheduled between18:00 and 21:00 h.

Fig. 1.

Whole-body vibration selectively and persistently enhances gamma oscillations in the M1 region of aged rats. A Experimental protocol and target regions for analyses are indicated. B Representative raw M1 ECoG recordings of a correct trial (top traces) and bandpass-filtered ECoG for 40 Hz gamma oscillations (bottom traces) from aged rats during resting state and vibrations at 40, 20, and 80 Hz. C-F Representative M1 ECoG power values from aged rats during resting state during resting state and vibrations at various frequencies. G-J The spectra of M1 ECoG recorded from aged rats during resting state and vibrations at various frequencies

WBV therapy

The animals assigned to the 40 Hz WBV and 80 Hz WBV treatment groups underwent twice-daily sessions of WBV therapy, lasting for 15 min each time (once between 10:00 and 12:00, and once between 15:00 and 17:00). The aged rats were placed in empty rat cages on a frequency-modulated electromagnetic vibration platform (Shuozhou Technology, Suzhou, CN), with weights added to the cages. The vibration device was set to maintain a fixed frequency of either 40 Hz or 80 Hz, with a vertical amplitude of 0.8 mm. Our preliminary pre-experiment indicated that this amplitude effectively induced gamma oscillations (not shown in the paper). After completing training with the previous batch of rats, the cages were cleaned with alcohol before introducing subsequent rats to avoid any influence from odor or excrement. This treatment regimen was administered five days a week for two months. The sham WBV group did not undergo any vibration exposure but followed all other procedures as described above.

Morris water maze (MWM)

The MWM task was conducted in a black circular pool measuring 160 cm in diameter and 50 cm in height (Xinruan, Shanghai, CN). The pool area was divided into four virtual quadrants (NW, NE, SW, SE). VisuTrack software (Xinruan, Shanghai, CN) was utilized to record the experiment. The experimental protocol commenced with four consecutive days of daily training consisting of one trial performed four times per day. During each trial, a platform was submerged 1 cm below the water surface at the center of a designated quadrant while visual cues were placed around the maze to assist rats in learning its location. In the initial MWM experiment, the platform was positioned in the northwest quadrant. Rats were placed successively in different starting quadrants clockwise (N, E, S, W) for each trial and given 60 s to search for the hidden platform. Escape latency—defined as the time taken by animals to find the platform—was measured during each trial using VisuTrack software. Additionally, swimming speed during training days was analyzed using this software as well. If rats failed to locate the platform within an experiment's duration, they were manually placed on it for 30 s and an escape latency of 60 s was recorded accordingly. Animals' performance throughout each training day served as an indicator of their memory level while changes observed in their average daily performance represented their learning progress over time. On day five, a single probe trial without any hidden platforms assessed long-term memory recall; rats were allowed to explore freely within the pool for one minute after removal of all platforms. Number of crossings over previous platform locations during this probe trial served as an indicator for evaluating long-term memory recall.

Immunohistochemistry

The rats were intraperitoneally anesthetized with pentobarbital, followed by perfusion through the apex of the heart with saline and 4% paraformaldehyde. Subsequently, the brains were removed and fixed overnight in 4% paraformaldehyde. After fixation, dehydration was carried out using an automatic dehydrator (P.S.J medical, Changzhou, CN). The brains were then embedded in paraffin and coronal sections of 4 μm thickness were obtained from the hippocampal region using a rotary microtome (Leica, Germany). Following section drying, they underwent deparaffinization and rehydration with ethanol and xylene. High-pressure repair with PBST solution (containing 0.01% Trition100) was performed for antibody treatment. Nonspecific binding was blocked by incubating the sections with 5% BSA for 30 min. Subsequently, the sections were incubated overnight at 4 °C with corresponding primary antibodies: anti-NeuN (rabbit; Abcam; 1:1,000), anti-GFAP (mouse; Proteintech; 1:500), anti-iba1 (mouse; Proteintech; 1:200), and anti-PSD95 (rabbit; Proteintech; 1:500). After washing three times with PBS solution, corresponding secondary antibodies including enhanced enzyme-labeled goat anti-mouse/rabbit IgG polymer (OriGene), CoraLite-488 (anti-mouse; Proteintech; 1:500), CoraLite-594 (anti-rabbit; Proteintech; 1:250) were added and incubated at room temperature for one hour. Following another round of PBS washing, the immunofluorescence sections were mounted using an antifading agent containing DAPI while immunohistochemical sections were stained with DAB and counterstained with hematoxylin. The images captured using a fluorescence upright microscope (BX53 Olympus) and quantification analysis conducted utilizing Fiji software (ImageJ version NIH). All antibodies used in this study have been satisfactorily validated by commercial suppliers.

Protein isolation and ELISA test

The rats were intraperitoneally anesthetized with pentobarbital, followed by perfusion through the apex of the heart with saline solution. Subsequently, the brains were rapidly dissected to extract the bilateral hippocampus. The extracted hippocampal samples were immediately frozen in liquid nitrogen and stored at − 80 °C until further biochemical analysis. Total protein from the hippocampus was extracted using RIPA buffer supplemented with protease and phosphatase inhibitors (Beyotime Biotechnology). The concentration of total protein in each sample was determined using a BCA protein assay kit (Solarbio). Electrophoresis (FuturePAGE, ACE) was performed to separate 20 μg of protein from each sample, which was then transferred onto a PVDF membrane. Following blocking, the membrane was incubated overnight at 4 °C with primary antibodies specific for GFAP (mouse, Proteintech; dilution: 1:20,000), PSD95 (rabbit, Proteintech; dilution: 1:4000), and GAPDH (mouse, Abcam; dilution: 1:10,000). After washing steps, HRP-conjugated secondary antibodies (Abcam; dilution: 1:5000) were applied to the membrane and incubated at room temperature for one hour. Chemiluminescence detection according to manufacturer's instructions (UVP ChemStudio, Analytik Jena) allowed visualization of signals on Fiji software (ImageJ 1.54f, NIH), enabling scanning of bands for gray value quantification.

The hippocampus tissue was mixed with PBS containing 1% PMSF at five times its weight before thorough grinding and subsequent extraction of supernatant. ELISA kits from Elabscience with catalog numbers E-EL-R0012c for IL-1β measurement, E-EL-R0015c for IL-6 measurement, and E-EL-M3063 for TNF-α measurement were used to quantify their respective levels in accordance with manufacturer's protocols. The absorbance after final color reaction was promptly measured at a wavelength of 450 nm using a microplate reader (iMark, Bio-Rad).

Data analysis and statistics

The data analysis and statistical tests were conducted using Microsoft Excel and GraphPad Prism 9. In the MWM, the performance on training days was assessed by measuring the escape latency, i.e., the time taken to locate the platform. Performance in probe trials, including extinction training, was evaluated by measuring the time spent in the target quadrant where the platform had been previously located. These datasets were analyzed using independent samples t-tests or one-way ANOVAs followed by post-hoc LSD tests. The comparison of nerve cell count and protein expression in rat brains utilized independent samples t-tests. After conducting omnibus tests (i.e., ANOVA and linear mixed-effects model), pairwise comparisons were made between levels when there was at least a tendency (P < 0.1) towards a significant effect. Statistically significant differences were determined at *P < 0.05, **P < 0.01, and ***P < 0.001 with unpaired two-sided testing; NS indicates non-significant.

Results

40 Hz WBV induces and sustains gamma oscillations in the primary motor cortex of aged rats

Although various stimulation modalities at a frequency of 40 Hz have been demonstrated to induce gamma oscillations in the brain (Martorell et al. 2019; Adaikkan et al. 2019), the impact of whole-body vibration on such oscillations remains uncertain. To address this inquiry, we recorded ECoG signals from the primary motor cortex (M1) in aged rats during both rest and vibrations at frequencies of 40, 20, and 80 Hz (Fig. 1A, B). Our findings reveal that M1 ECoG power specifically increased at a frequency of 40 Hz during 40 Hz WBV compared to the resting state (Fig. 1C, D). Moreover, M1 ECoG exhibited sustained elevated power levels during continuous vibration at 40 Hz (Fig. 1G, H). In contrast, vibrations at frequencies of 20 and 80 Hz did not elicit any significant changes in M1 ECoG power corresponding to their respective frequencies (Fig. 1E, F), although they marginally enhanced overall gamma oscillations within the range of 30–100 Hz (Fig. 1I, j). These results suggest that whole-body vibration at 40 Hz selectively and persistently enhances gamma oscillations within the M1 among aged rats.

40 Hz WBV modifies behavioral performance of aged rats

To assess the impact of 40 Hz WBV on learning and memory abilities, we conducted a separate cohort study using the MWM, which evaluates hippocampus-dependent spatial learning and memory in rats (Fig. 2A). Initially, we observed impaired learning and memory abilities in aged rats, as evidenced by their increased time to locate the platform during the training phase (Fig. 2B). However, the group exposed to 40 Hz vibration demonstrated significant improvement in learning performance among aged rats, as indicated by reduced latency to find the platform during training (Fig. 2C). In the probe trial conducted 24 h after the final training session, both the young group and the 40 Hz vibration-aged group exhibited significantly better spatial memory compared to the aged group. This was evident from their increased visits to the platform location and longer duration spent in the target quadrant (Fig. 2E, F). Conversely, no significant differences were observed between either phase or probe trial results of the 80 Hz vibration-aged group when compared with those of aged rats (Fig. 2C, E, F). Additionally, swimming speed remained unchanged across all four groups (Fig. 2G).

Fig. 2.

Effect of 40 Hz WBV on spatial learning and memory in rats. A Experimental design. The 18-month-old rats underwent 8 weeks of whole-body vibration intervention. After 8 weeks, MWM was performed to assess memory functions. B Latency for finding a platform during training in young (n = 9) and aged rats (n = 11). C Latency of plateau finding during training in aged rats exposed to 40 Hz vibration (n = 11), 80 Hz vibration (n = 9) and no intervention (n = 11). D Representative trails from MWM probe trial. E Time spent in the target quadrant during probe test. F Number of platform crossings in the probe test. G Swimming velocity during MWM. # *P < 0.05, ## **P < 0.01, ### ***P < 0.001, ns indicates not significant, data are shown as the mean ± SEM

40 Hz WBV attenuates neuronal apoptosis in the hippocampal CA1

To further validate the impact of 40 Hz WBV on cognitive ability in aged rats objectively, we employed HE staining and immunohistochemical staining to examine neuronal alterations in the CA1 of the rat hippocampus, which is crucial for learning and memory processes. The results from HE staining revealed a substantial increase in apoptotic neurons within the CA1 region of aged rats (Fig. 3A), characterized by irregularly shaped nuclei with intense blue staining (Fig. 3B), accompanied by nuclear shrinkage, dissolution, fragmentation, and other related phenomena. This also indicated that apoptosis directly influenced neuron numbers within the hippocampal CA1. We quantified NeuN-immunostained cells within the CA1 region as a measure of neuron count (Fig. 3C). Compared to young rats, there was a significant reduction in neuron count within the CA1 region of aged rats (Fig. 3D). Notably, exposure to 40 Hz vibration significantly inhibited this apoptotic trend; however, no significant reduction in apoptosis was observed with an 80 Hz vibration (Fig. 3A, D).

Fig. 3.

Immunohistochemical staining was performed on the CA1 region of the rat hippocampus. A HE staining images of the hippocampal CA1 region from each rat group, picture above scale bar: 200 μm, picture below scale bar: 40 μm. B Left: HE staining of normal neuronal nuclei, right: HE staining of apoptotic neuronal nuclei. C NeuN immunohistochemical staining images of the hippocampal CA1 region were obtained for each group of rats., picture above scale bar: 200 μm, picture below scale bar: 40 μm. D Neuronal counts within the visual field of the CA1 region in the hippocampus were assessed for each group of rats, utilizing a 40X magnification. E GFAP immunohistochemical staining images of the hippocampal CA1 region were obtained for each group of rats., picture above scale bar: 200 μm, picture below scale bar: 40 μm. F Astrocyte counts within the visual field of the CA1 region in the hippocampus were assessed for each group of rats, utilizing a 40× magnification. n = 4 rat per group # *P < 0.05, ## **P < 0.01, ### ***P < 0.001, ns indicates not significant, data are shown as the mean ± SD

Hz WBV reduces the synaptic connection between astrocytes and neurons

To investigate the underlying mechanism behind the neuroprotective effects of 40 Hz WBV or 40 Hz gamma oscillation on neuronal apoptosis, we initially conducted GFAP immunostaining on the hippocampal CA1 (Fig. 3E). The findings revealed a significant decrease in astrocyte population in aged rats compared to young rats; however, neither 40 Hz nor 80 Hz vibration ameliorated this reduction (Fig. 3F). Interestingly, we observed that astrocytic process were more abundant and evenly distributed in the 40 Hz vibration-aged group when compared to both the aged group and the 80 Hz vibration-aged group (Fig. 3E). This observation may be crucial for understanding how 40 Hz gamma oscillation rescuing cognitive impairment (Cheng et al. 2023).

To further elucidate the specific causes of astrocytic morphological changes, we performed immunofluorescence staining for GFAP and PSD-95 on hippocampal slices from various groups of rats (Fig. 4A). The results revealed significant colocalization of GFAP and PSD-95 fluorescence around apoptotic neurons in aged rats (Fig. 4B), indicating a possible increase in synaptic connections between astrocytes and apoptotic neurons. Interestingly, 40 Hz gamma oscillations significantly reduced this colocalization area without altering the average fluorescence intensity (Fig. 4C, D), providing a potential explanation for the mitigation of neuronal apoptosis in aged rats by 40 Hz gamma oscillations. How do 40 Hz gamma oscillations reduce these synaptic connections? Further analysis of Fig. 4A showed that 40 Hz gamma oscillations did not change the positive area or average fluorescence intensity of GFAP expression in the hippocampus of aged rats (Fig. 4E, F). In contrast, 40 Hz gamma oscillations significantly increased the positive area of PSD-95 expression while reducing its average fluorescence intensity in the hippocampus (Fig. 4G, H). These findings suggest that 40 Hz gamma oscillations might reduce synaptic expression around apoptotic neurons and could promote more extensive synaptic connections in other regions of the hippocampus in aged rats.

Fig. 4.

Immunofluorescence staining and western blot analysis were performed on the rat hippocampus. A Immunofluorescence staining images of the CA1 region were obtained from each group of rats. Scale bar: 40 μm. B Extensive co-staining of PSD95 and GFAP was observed surrounding apoptotic neurons in the CA1 region of aged rats. C The positive area of PSD95 and GFAP co-staining in the CA1 region of rats was normalized to that in young rats. D The mean fluorescence intensity of PSD95 and GFAP co-staining in the CA1 region was measured for each group of rats. E The positive area of GFAP expression in the CA1 region was normalized to that in young rats. F The mean fluorescence intensity of GFAP expression in the CA1 region was quantified. G The positive area of PSD95 expression in the CA1 region was normalized to that in young rats. H The mean fluorescence intensity of PSD95 expression in the CA1 region was determined. I A representative western blot showed levels of GFAP and GAPDH proteins in the rat hippocampus. J The relative fold change of GFAP, normalized to the young group. K A representative western blot demonstrated levels of PSD95 and GAPDH proteins in the rat hippocampus. L The relative fold change of GFAP, normalized to the young group. n = 4 rat per group. # *P < 0.05, ## **P < 0.01, ### ***P < 0.001, ns indicates not significant, data are shown as the mean ± SD

To validate the above conclusions, Western blotting was employed to detect the levels of GFAP and PSD-95 in the hippocampus of various groups of rats (Fig. 4I, K). The results similarly indicated that 40 Hz gamma oscillations did not alter GFAP expression in the hippocampus of aged rats (Fig. 4J), while significantly increasing PSD-95 expression (Fig. 4L). This further supports the idea that 40 Hz gamma oscillations may influence astrocytic synaptic connections in the hippocampus by regulating synaptic expression.

40 Hz and 80 Hz WBV effectively enhances the hippocampal inflammatory response

Not only does hippocampal synaptic expression impact cognitive function in aged rats, but it is widely believed that inflammatory responses are closely associated with cognitive impairment. Microglia, in particular, are closely linked to neuroinflammation (Leng and Edison 2021). We quantified the number of Iba1 immunostained cells in the CA1 region as a measure of microglial count (Fig. 5A, red arrow). Compared to young rats, there was a significant increase in the number of microglia in the CA1 region of aged rats. Interestingly, 40 Hz whole-body vibration not only reduced the number of microglia in the CA1 region of the hippocampus in aged rats but also did 80 Hz whole-body vibration training (Fig. 5B).

Fig. 5.

The detection of inflammatory markers in the hippocampus of rats A Iba1 immunohistochemical staining images of the hippocampal CA1 region were obtained for each group of rats. Red arrow: microglia. Picture above, scale bar: 200 μm, picture below scale bar: 40 μm. B Microglia counts within the visual field of the CA1 region in the hippocampus were assessed for each group of rats, utilizing a 40× magnification. C–E Relevant levels of pro-inflammatory factors in the hippocampus of rats were measured

Moreover, the levels of inflammatory factors were significantly higher in aged rats compared to young rats. Both 40 and 80 Hz whole-body vibration training reduced the levels of inflammatory factors in aged rats (Fig. 5C–E). As 80 Hz whole-body vibration does not lead to a significant increase in gamma oscillation power in aged rats, we posit that the improvement of age-related cognitive impairment by 40 Hz gamma oscillations is not specifically mediated through an inflammatory response mechanism. Instead, it is more likely that whole-body vibration training itself can moderately alleviate the inflammatory response in the hippocampus of aged rats.

Discussion

In this study, we have demonstrated, for the first time, that 40 Hz WBV can influence cognitive function by inducing gamma oscillations in the primary motor cortex. However, the relationship between specific frequencies of WBV and gamma oscillations in primary motor cortex remains unclear. We propose that this may be related to the mechanism of Short Latency Afferent Inhibition (SAI), a neurophysiological phenomenon where a transcranial magnetic stimulation (TMS) pulse delivered over the primary motor cortex is preceded by peripheral electrical nerve stimulation at a short inter-stimulus interval (around 20–28 ms). This results in a reduction of the motor evoked potential (MEP) amplitude (Tokimura et al. 2000). SAI is often used in research to study conditions affecting sensorimotor pathways, but more important is the potential as a result biomarker for cognitive impairments. Because SAI is not only crucial for the memory process (Bonnì et al. 2017), but also exhibits impairment in elderly individuals (Young-Bernier et al. 2012), patients with Alzheimer's disease (Di Lorenzo et al. 2016), and patients with mild cognitive impairment (Martorana et al. 2014). This impairment is directly associated with cognitive decline (Young-Bernier et al. 2012). Interestingly, the 40 Hz WBV corresponds to a stimulation interval of 25 ms, precisely matching the required interval for SAI to occur. Furthermore, the GABA secreted by PV interneurons, which generate gamma oscillations, is closely linked to SAI. Despite these intriguing coincidences, further investigation is needed to understand the relationship between 40 Hz WBV, SAI, and gamma oscillations.

On the other hand, we have demonstrated that whole-body vibration at 40 Hz effectively induces gamma oscillations in the primary motor cortex of aged rats. Furthermore, long-term training with 40 Hz WBV has been shown to mitigate neuronal loss in the hippocampal CA1 region and improve behavioral performance in aged rats (Yang et al. 2016; Bidonde et al. 2017). As an emerging rehabilitation method, WBV exhibits significant potential across various scenarios due to its economic feasibility, convenience, and minimal side effects (Wollersheim et al. 2017; Tan et al. 2023). Consequently, we have applied WBV for the first time using a 40 Hz scheme as a treatment for ARCI. To ensure accurate ECoG recordings without interference from frequency-modulated electromagnetic vibrations emitted by the platform, we employed silver-containing cloth to completely isolate the platform from our recording system. This allowed only mechanical waves to be transmitted to the test animals. However, considering practical application scenarios during subsequent long-term training sessions involving both 40 and 80 Hz WBV interventions, we did not isolate electromagnetic signal transmission between the vibration platform and animals.

We observed a significant number of apoptotic neurons in the hippocampal CA1 region of cognitively impaired aged rats through immunohistochemical staining, confirming the strong association between CA1 and cognitive impairment (Mizutani et al. 1990). However, the underlying cause of neuronal apoptosis remains unclear. Additionally, we discovered that 40 Hz WBV induced 40 Hz gamma oscillation in the cortex and mitigated neuronal apoptosis in aged rats. Furthermore, significant morphological changes were observed in astrocytes surrounding the hippocampal CA1 region, with a concentration around apoptotic neurons. The findings established a connection between astrocytes and neuronal apoptosis (Jung et al. 2023). Considering the profound impact of synaptic connections between astrocytes and neurons on their functionality (Takano et al. 2020; Hösli et al. 2022), we investigated synaptic changes in astrocytes. The results revealed an abundance of synaptic connections between these astrocytes and apoptotic neurons within the hippocampal CA1 region of aged rats. Moreover, 40 Hz gamma oscillation significantly reduced these specific synaptic connections without impairing other more extensive ones.

Interestingly, we observed a decrease in the number of astrocytes in the CA1 region of the hippocampus in older rats, which contradicts the prevailing belief that GFAP expression is upregulated and astrocyte hypertrophy occurs with age. In fact, early stages of various neurodegenerative diseases such as Alzheimer's disease, amyotrophic lateral sclerosis, and Wernicke encephalopathy have been associated with the detection of astroglial atrophy (Rossi et al. 2008; Rossi and Volterra 2009; Heneka et al. 2010; Verkhratsky et al. 2019). And this early atrophy of astrocytes may have significant functional implications, a as atrophic astrocytes provide less synaptic coverage, leading to detrimental effects on synaptic transmission associated with compromised ion and neurotransmitter homeostasis or reduced local metabolic support. Additionally, astroglial asthenia results in decreased neuroprotection. These changes are likely to weaken synaptic transmission and impact synaptic plasticity, thus contributing to the initial cognitive deficiency observed during the early stages of ARCI (Verkhratsky et al. 2019).

In the assessment of inflammatory markers, as both frequencies significantly reduced hippocampal inflammation without significant differences between them regarding inflammatory markers assessment, we speculate that attenuation of hippocampal inflammatory response mainly originates from WBV itself rather than specifically from 40 Hz gamma oscillation (Oroszi et al. 2022). The precise relationship between 40 Hz gamma oscillation and inflammation still requires further investigation.

Conclusion

In summary, our study findings suggest that 40 Hz WBV significantly enhances gamma oscillation at 40 Hz in the motor cortex and improves cognitive function in aged rats. These results indicate that this novel intervention holds potential as a therapeutic method for preventing and treating ARCI. Additionally, we have identified the neuron-astrocyte synapses play a crucial role in 40 Hz gamma oscillation therapy for ARCI, providing valuable insights for future mechanistic investigations into this therapeutic approach (Francesco and Koch 2021). While the results are encouraging, several limitations must be addressed. Firstly, the study's sample size was relatively small, which might affect the generalizability of the findings. Secondly, the duration and frequency of WBV were consistent; varying these parameters might yield different outcomes. It is noteworthy that our experiments demonstrate the impact of 40 Hz cortical gamma oscillations on age-related cognitive impairment at the level of cognitive function in aging. This discovery may serve as a crucial determinant in comprehending the heterogeneity of cognitive dysfunction observed in biological aging, thereby bridging important gaps outlined in “Seven Knowledge Gaps in Modern Biogerontology” (Rattan 2024).

In conclusion, WBV presents a novel, non-invasive intervention with potential to enhance cognitive function in older adults. Future studies should aim to validate these findings in clinical settings and explore the optimal WBV protocols for maximizing cognitive benefits.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

XW.X. and MS.L. planned the study, while MS.L. and L.L. conducted surgery on rats. RZ.C. and QL.W. carried out the immunohistochemical experiment, and LM.S. and TF.Z. performed western blotting. JH.H., J.X., and CZ.Y. conducted the ELISA tests. MS.L. analyzed the data and drafted the manuscript. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China Grant (81860449). The funders had no role in the design of the study, in the collection, analysis, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.

Data availability

The data that supports the fundings of this study are available from the corresponding author upon reasonable request. No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

All experimental procedures received approval from Guilin Medical College's Animal Ethics Committee (Approval Number: GLMC202105081) and adhered to guidelines outlined by European Union Council Directive (86/609/European Economic Community).