Abstract
As the most common form of dementia in the world, Alzheimer's disease (AD) is a progressive neurological disorder marked by cognitive and behavioral impairment. According to previous researches, abundant social connections shield against dementia. However, it is still unclear how exactly social interactions benefit cognitive abilities in people with AD and how this process is used to increase their general cognitive performance. In this study, we found that single novel social (SNS) stimulation promoted c-Fos expression and increased the protein levels of mature ADAM10/17 and sAPPα in the ventral hippocampus (vHPC) of wild-type (WT) mice, which are hippocampal dorsal CA2 (dCA2) neuron activity and vHPC NMDAR dependent. Additionally, we discovered that SNS caused similar changes in an AD model, FAD4T mice, and these alterations could be reversed by α-secretase inhibitor. Furthermore, we also found that multiple novel social (MNS) stimulation improved synaptic plasticity and memory impairments in both male and female FAD4T mice, accompanied by α-secretase activation and Aβ reduction. These findings provide insight into the process underpinning how social interaction helps AD patients who are experiencing cognitive decline, and we also imply that novel social interaction and activation of the α-secretase may be preventative and therapeutic in the early stages of AD.
Significance Statement
Alzheimer's disease is a neurodegenerative disease that endangers the health of humans all over the world, yet no effective treatment is available. Here, we propose that novel social communication is able to effectively alleviate synaptic plasticity and cognitive deficits in early AD model mice. The mechanism is related to the activation of vHPC α-secretase, which alters amyloid precursor protein (APP) cleavage pathways, leading to a decrease in Aβ generation. Our findings shed light on the underlying mechanisms by which social communication improves cognition in AD models or patients and emphasize the preventive and therapeutic potential of novel social communication and α-secretase activation in the early stages of AD.
Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disease with insidious onset. It is also the fifth leading cause of death for the elderly people globally (Alzheimer’s Association, 2021; Knopman et al., 2021; Alzheimer’s Association, 2023). To date, the pathogenesis of AD is still controversial, and there is a lack of effective treatment measures.
Regarding the pathogenesis of AD, there are numerous hypotheses, including the Aβ hypothesis, tau protein hypothesis, cholinergic theory, etc. (Du et al., 2018; Long and Holtzman, 2019). The Aβ hypothesis, which contends that the buildup of Aβ protein in the brain is the initial and main cause of AD, although there has always been controversy, is still the one that has received the most research attention (Hardy and Higgins, 1992; Selkoe and Hardy, 2016; Hampel and Hardy, 2021). Aβ is a hydrophobic peptide with 38–42 aa long and is generated from amyloid precursor protein (APP; LaFerla et al., 2007). After being produced as soluble monomers, Aβ is seen in a variety of intermediate aggregation states, such as dimers and trimers, soluble oligomers, and protofibrils, before it forms fibrils that accumulate in plaques, typically regarded as an AD neuropathological hallmark (He et al., 2018; Hampel et al., 2021). Among them, researchers generally believe that the most hazardous aggregation form is Aβ oligomer, which can cause pathological cascade events such as inflammation, disruption of long-term potentiation in the hippocampal region, dysfunctional synapses, neurofibrillary tangles, and neuronal death (Walsh et al., 2002; Shankar et al., 2008; Bernstein et al., 2009; Koffie et al., 2009; Mc Donald et al., 2010; Tomiyama et al., 2010; Hong et al., 2018; Li and Selkoe, 2020; Walsh and Selkoe, 2020).
APP processing can be divided into nonamyloidogenic pathway and amyloidogenic pathway (Zhang et al., 2011; Hampel and Hardy, 2021). In physiological conditions, APP is prioritized to be sequentially cleaved by α-secretase and γ-secretase, namely, nonamyloidogenic pathway, which cleaves APP within the Aβ region, leading to the production of a soluble APP N-terminal fragment (sAPPα) and a shorter, nontoxic fragment (p3; Haass et al., 1993; Hick et al., 2015). The amyloidogenic pathway is an alternative cleavage pathway for APP sequentially mediated by β-secretase and γ-secretase, resulting in the formation of sAPPβ and intact Aβ peptide (LaFerla, Green and Oddo, 2007; O’Brien and Wong, 2011; Hampel et al., 2021). The most frequently studied α-secretases in the brain are members of a disintegrin and metalloprotease (ADAM) family, in particular ADAM10 and ADAM17 (Lammich et al., 1999; Jorissen et al., 2010; Kuhn et al., 2010; Qian et al., 2016). Researches have shown that production of sAPPα or activation of α-secretase reduces β-secretase activity and amyloid-β generation, suggesting that competition exists between amyloidogenic and nonamyloidogenic processing and activating α-secretases is a potentially effective therapeutic approach for the treatment of AD (Postina et al., 2004; Obregon et al., 2012; Suh et al., 2013).
Nonpharmacological therapies, such as physical activity, cognitive training, social interaction, and vascular and metabolic risk management, are applied in conjunction with drug therapy of AD (Sabbagh et al., 2020; Scheltens et al., 2021). The enriched social network and frequent social contact are reported to protect against dementia and increase patients' cognitive reserve (Fratiglioni et al., 2000; Zhou et al., 2018a; Sommerlad et al., 2019; Cai, 2022; Sommerlad et al., 2023). Although the benefits of social stimulation for individuals with AD are well documented, the specific mechanism through which it enhances cognitive abilities remains unclear.
Previous studies have shown that social stimuli can activate multiple brain regions in rodents (Chen and Hong, 2018; Perkins et al., 2017; Tanimizu et al., 2017). It is unclear how these activities are associated with promoting the cognition of AD patients or animal models. In this study, we demonstrated that a single novel social (SNS) stimulation activated vHPC cells and increased α-secretase and sAPPα expression in WT mice through dCA2 activity and NMDAR-dependent pathways. Likewise, the changes caused by SNS stimulation are conservative in the AD model FAD4T mice. In addition, multiple novel social (MNS) stimulation to early-stage FAD4T mice effectively alleviated their cognitive and pathological defects. Our study elucidates the mechanisms by which novel social stimulation alleviates AD phenotype, providing MNS stimulation as a potential noninvasive, nonpharmacological treatment method in the early stage of AD.
Materials and Methods
Mice
All the mice utilized in the experiment were between 2 and 3 months old and had the C57BL/6 genetic background. Amigo2-Cre mice were obtained from Jackson Laboratory (catalog #030215). FAD4T mice coexpressed human APP with the Swedish (KM670/671NL) and Indian (V717F) variants, together with human mutant PS1 (M146L, L286V), driven by the mouse Thy1 promoter. The FAD4T line was maintained by breeding male hemizygous transgenic mice with female breeders. FAD4T mice were purchased from GemPharmatech (Strain ID T053302). All animals were kept in standard laboratory settings with ad libitum access to food and water as well as a 12 h light/dark cycle, a temperature range of 22–26°C, and a humidity level of 55–60%. The mice utilized in this study were all cared for in accordance with Southeast University's (Nanjing, China) authorized protocols.
Antibodies, chemicals, and other reagents
Primary antibodies include rabbit polyclonal rabbit monoclonal antibody β-amyloid (D54D2; 1:1,000, Cell Signaling Technology, catalog #8243T), rabbit monoclonal antibody anti-BACE (D10E5) (1:1,000, Cell Signaling Technology, catalog #5606S), rabbit monoclonal antibody anti-ADAM10 (1:2,000, Proteintech, catalog #25900-1-AP), rabbit monoclonal antibody anti-ADAM17 (1:1,000, Abcam, catalog #ab39162), rabbit polyclonal anti-APP (1:1,000, C-terminal; Proteintech, catalog #25524-1-AP), mouse monoclonal anti-sAPPα (IBL, catalog #11088), mouse monoclonal anti-sAPPβ (IBL, catalog #10,321), phospho-Tau (Cell Signaling Technology, catalog #49561S), and GAPDH (1:2,000, GenScript, catalog #A0162). Secondary antibodies include goat anti-rabbit (GenScript, catalog #A00098), goat anti-mouse (GenScript, catalog #A00160), and goat anti-rabbit IgG (H+L) cross-adsorbed secondary antibody (Invitrogen, catalog #A-21070). Drugs and reagents include D-AP5 (Tocris, catalog #0106), DAPI (Cayman Chemical, catalog #28718-90-3), Diamond Antifade Mountant (Thermo Fisher Scientific, catalog #P36965), Thioflavin T (Macklin, catalog #D844310), protease inhibitor cocktail and phosphatase inhibitor (Roche, catalog #1836153), and enhanced chemiluminescence (Thermo Fisher Scientific, catalog #34580).
SNS stimulation
The cages housing tested WT or FAD4T mice were divided into two groups, one group was exposed to a novel stranger mouse (S1) of similar size, age, and same gender, which was allowed to interact for 10 min as a SNS stimulus. No new mice were introduced to the other group (group-housed, GH), in which mice can only socialize with their familiar littermates.
MNS stimulation
On day 1, one group of 2.5-month-old FAD4T mice was exposed to a new stranger mouse (S1) of similar size, similar age, and same gender. They were allowed to socialize for 1 h. On the following day (day 2), another stranger mouse (S2) with similar size, similar age, and same gender was introduced and allowed to socialize for 1 h again. This process was repeated until the introduction of seventh stranger mouse (S7). The same sequence was repeated for another time between days 8 and 14. No new mice were introduced to the GH cage, in which mice can only socialize with familiar littermates. For the 7 d MNS treatment, the male S1 was introduced to the previously group-housed 2.75-month-old male mice on day 8, S2 on day 9, and so on.
Thioflavin T staining
Male mice were anesthetized using 10% chloral hydrate and quickly killed with heads severed. The brain was preserved at 4°C in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) for at least 48 h. After fixation, the brain was immediately placed into a 30% sucrose solution and kept at 4°C for 48 h. The brain was fast frozen in liquid nitrogen and maintained at −80°C before being embedded in Tissue-Tek O.C.T. Compound (Sakura) and sliced into 30 µm coronal cryostat slices at −20°C (Leica CM1950). The brain slices were then stuck on a glass slide that was coated with poly-D-lysine. For staining, slices were first rinsed in 70% ethanol for 1 min and then in 80% ethanol for another 1 min. Then, the slides were incubated in a filtered Thioflavin T (0.01% in 80% ethanol) solution for 15 min. After that, slices were stained and rinsed with 80% ethanol for 1 min, 70% ethanol for another 1 min, and thrice with distilled water. After mounting, the Thioflavin T labeled green plaques were seen with fluorescent microscopy (Leica DM5000 B).
Behavioral tests
Animals between 2.5–3 months old were adopted and before behavioral tests, mice were handled daily for 3 d and moved from the colony room to the behavioral room for at least 1 d. All trials were conducted between 9:00 A.M. and 4:00 P.M. Littermates with various treatments were housed in groups of 3–5 per cage with ad libitum access to food and water. Between each participant, a thorough cleaning of the device with 75% ethanol was performed.
Open field test
The open field test (OFT) was conducted as previously mentioned (Zhou et al., 2018a,b). A rectangular Plexiglas box with four 50-cm-high sidewalls and an open roof served as the open field equipment. The space was lit by faint in-ceiling lights that were positioned in the center. Recording began 1 min after the mice were released in the same location in the arena, close to the center, and continued for another 10 min. A computer-based video tracking system (Noldus software) tracked mouse movement in the open field. For analysis, the box was separated into the center (15 cm in diameter) and peripheral areas.
Novel object recognition test
Novel object recognition (NOR) was conducted with the same equipment as the OFT. The test was divided into two phases: training trial and testing trial. The mice were individually brought into the arena on the training day, which contained two identical objects (A and A′), and given 10 min to explore before being returned to their home cage. After 24 h, on the testing trial, object A′ was replaced with a different novel object (B) with a different color and shape. A 5 min free exploration period was given to each mouse once it was individually placed in the arena. EthoVision XT 13 was used to capture each mouse's behavior. Two experienced researchers manually scored the time (t) spent exploring the novel and familiar objects independently. The formula used to determine the discrimination index was (tnovel − tfamiliar) / (tnovel + tfamiliar). A small number of participants were eliminated from behavioral analysis if the total exploration duration was fewer than 8 s on testing day.
Three-chamber test
The three-chamber apparatus was made of a Plexiglas box (50 × 25 cm) with a removable floor and three rooms separated by two partitions with a 10 × 5 cm door to allow the mouse to travel between each chamber. Four trials were conducted to complete the task. In stage 1, a test mouse was put in the middle of the three-chamber apparatus, and two wired metal pencil boxes were placed in the upper left and upper right corners of the instrument. The mouse was allowed to explore freely for 10 min, and the time spent sniffing each pencil box was recorded. When the habituation was over, the test mouse was gently directed to the central chamber with doors on both sides closed. In stage 2, a stranger mouse (Stranger 1) was placed in any pencil box. Then, the test mouse was allowed to freely explore the arena for another 10 min. The time spent in contact with Stranger 1 and the empty pencil box was recorded. When the sociability test session was over, the test mouse was guided to the middle chamber and stayed for 10 min, with the doors of both sides closed (stage 3). In stage 4, the second stranger mouse (Stranger 2) was introduced into the rest empty pencil box, and the test mouse was allowed to explore the whole instrument for 10 min. The time of contact between Stranger 1 and Stranger 2 was recorded. Social ability index = (tS1 − tempty) / (tS1 + tempty), social memory index = (tS2 − tS1) / (tS2 + tS1).
Brain slice electrophysiology
All electrophysiological recordings were made at the Schaffer collateral–commissural circuit in the ventral hippocampus (vHPC). Briefly, the male mouse brains were rapidly dissected and stuck upside down on the platform and then horizontally sliced to 360 µm in ice-cold artificial cerebrospinal fluid (ACSF) that was saturated with 95% O2/5% CO2. The ACSF included the following (in mM): 120.0 NaCl, 3.0 KCl, 1.2 MgSO4, 1.0 NaH2PO4, 26.0 NaHCO3, 2.0 CaCl2, and 11.0 D-glucose. Then slices were incubated at 95% O2/5% CO2 saturated ACSF at 28°C for at least 2 h before being transferred to an ACSF-perfused recording chamber. The perfusion flow rate was maintained at 2 ml/min at all times. Using an infrared differential interference contrast microscope (Olympus X51), hippocampal ventral CA1 neurons were visualized. A stimulating electrode was positioned in the SC pathway 200–300 µm from the CA1 cell bodies to elicit fEPSPs at 0.067 Hz recorded in the stratum radiatum using a nearby glass micropipette filled with ACSF (access resistance, 3–6 M) for field recordings. Theta burst stimulation (TBS, 10 bursts at 5 Hz; 5 pulses at 100 Hz per burst) to induce LTP after a stable 20 min baseline period. By comparing the fEPSP slopes of the last 10 min and baseline, LTP amplitude was calculated and statistically analyzed.
Immunohistochemistry
In brief, male mice were anesthetized using 10% chloral hydrate and perfused with 0.1 M PBS and followed with 4% PFA in PBS. The brain was removed, then fixed for 16 h in 4% PFA, and then transferred to 30% sucrose in PBS for 48 h at 4°C. The brain was then embedded in Tissue-Tek O.C.T. Compound (Sakura), flash frozen in liquid nitrogen, and kept at −80°C before being cut into coronal cryostat slices of 30 µm thickness and preserved at −20°C (Leica CM1950). For immunostaining, the brain slices were stuck on a glass slide that was coated with poly-D-lysine. Slices were then permeabilized for 2 h with 0.1% Triton X-100 in PBS, blocked for 1 h with 10% FBS, and then incubated with the c-Fos antibody overnight at 4°C. After PBS washing for three times (5 min for each), the slices were incubated with secondary antibody at 37°C for 2 h and then washed in PBS for six times (5 min for each) before being stained with DAPI and mounted. Images were collected by Zeiss confocal microscopes LSM 700. Data were analyzed using ImageJ software (National Institutes of Health).
Virus injection
Adult male Amigo2-Cre mice were anesthetized with 4% isoflurane and maintained with 1% isoflurane and placed into a stereotaxic frame with a heating pad (37.5–38°C) to keep them warm. A glass pipette backfilled with mineral oil was connected to a 10 µl Hamilton microsyringe. The micropipette was slowly lowered to the target site, and virus [rAAV-EF1α-DIO-hM4D(Gi)-EYFP-WPREs, 5.13E + 12 vg/ml] was bilaterally delivered to the dorsal CA2 region (AP, −1.6 mm; ML, ±1.6 mm; DV, −1.7 mm), driven by a microinjection pump at 50 nl/min. The injection pipette was retracted with 0.01–0.02 mm after the injection finished and left for an additional 10 min. Atipamezole (10 mg/kg) was injected intraperitoneally. The animal was finally put into a recovery cage with a heated pad before being taken back to its cage and watched for 24 h. Normally, after surgery, mice were given a 6–8 d recovery period. The viruses were allowed to express at least 3 weeks prior to subsequent biochemistry and histological experiments.
Cannula implantation and cannula infusion of drugs
Adult male mice were anesthetized with 4% isoflurane and maintained with 1% isoflurane and placed in a stereotaxic apparatus with a heating pad to keep them warm. After securing the animal head in the stereotaxic apparatus, the guide cannula (RWD, catalog #62003, 26 gauge, C = 4.4 mm) was positioned in its support in a straight position, and a 2.5 cm anterior–posterior incision was made on the scalp's midline. Using the Allen Mouse Brain Atlas as a guide, the guide cannulas were bilaterally positioned in the ventral CA1 (AP, −3.25 mm; ML, ±3.16 mm; DV, −4.15 mm). Light-cured dental adhesive cement and tiny screws were used to attach the optical implants to the skull. The cannulas were then fixed with screws after being extensively covered in dental cement. Cannula caps (RWD, catalog #62102, 26 gauge, G2 = 0.2 mm) placed in the guides after the cement has completely dried prevented obstruction. Atipamezole (10 mg/kg) was injected intraperitoneally. The animal was finally put into a recovery cage with a heated pad before being taken back to its cage and watched for 24 h. Normally, after surgery, mice were given a 6–8 d recovery period. The cannula cap was taken out of the guide cannula after the animal head had been fixed in the stereotaxic apparatus, and an injection cannula (RWD, catalog #62203, 30 gauge, G1 = 0.2 mm) was gently inserted. The injection cannula was linked to a 10 µl Hamilton microsyringe that was powered by a microinjection pump via PE20 tubing that was backfilled with mineral oil that had separated from the peptide solution. A volume of 0.8 µl of the solution was injected on each side at a rate of 0.3 µl/min. To reduce dragging of injected liquid along the injection track, the injection cannula was left in place for an extra 5 min after injection.
Immunoblots
After the male mice were killed, the HPC was rapidly removed and homogenized in a Dounce homogenizer with 0.4 ml ice-cold lysis buffer, containing (in mM) 20 Tris-HCl, pH 7.5, 150 NaCl, 1 EDTA, 1 EGTA, 1% Triton X-100, 2.5 sodium pyrophosphate, 1 β-glycerophosphate, 1 Na3VO4, 20 NaF, and 1% protease inhibitor cocktail and phosphatase inhibitor (Roche), and kept at 4°C for 40 min before debris was removed by centrifugation at 14,000×g for 10 min. The protein samples were either stored or blended with 25% by volume of 5× SDS loading buffer (250 mM Tris-HCl, 10% SDS, 0.5% bromophenol blue, 50% glycerol, 5% β-mercaptoethanol, pH 7.4) to prepare them for electrophoresis on an SDS-PAGE gel and electrotransfer on PVDF filter. After that, the filter was incubated with the appropriate primary antibodies in TBST for overnight incubation at 4°C after being blocked with 5% dry milk in TBST (20 mM Tris-HCl, 9% NaCl, 1% Tween 20, pH 7.6). Following washing and incubation with appropriate secondary antibodies, the filter was washed extensively, developed using an enhanced chemiluminescence method of detection, and analyzed using AlphaEaseFC software, and protein loading was controlled by normalizing each tested protein with GAPDH on the same blot.
RNA isolation and quantitative RT-PCR
Male mouse vHPC RNA was extracted with Total RNA Isolation Kit (Vazyme Biotech, catalog #RC112-01). One microgram of the total RNA for each sample was reverse transcribed into cDNA with the cDNA Synthesis Kit (Vazyme Biotech, catalog #R111-01/02). The cDNA was then amplified with the SYBR Green Master Mix Kit (Vazyme Biotech, catalog #Q111-02/03) in the CFX96 real-time PCR system (Bio-Rad). The amplification cycle was 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s, 59°C for 1 min. The levels of target RNA were adjusted to that of GAPDH. Each reaction was run in triplicate and was analyzed following the ΔΔCt method. Primer information is as follows:
ADAM10, forward: CCTGCCATTTCACTCTGTCATTTA,
reverse: GTGCCCGGGCTCCTTCCTCTACTC;
ADAM17, forward: AGGACGTAATTGAGCGATTTTGG,
reverse: TGTTATCTGCCAGAAACTTCCC;
BACE1, forward: GGAGACCGACGAGGAATCG,
reverse: GCAAAGTTACTACTGCCCGTG;
GAPDH, forward: AGGTCGGTGTGAACGGATTTG,
reverse: TGTAGACCATGTAGTTGAGGTCA.
Software and algorithms
Prism 8.0.1 GraphPad, https://www.graphpad.com/
SigmaPlot 12.0 Systat Software, http://sigmaplot.co.uk/
IBM SPSS Statistics 27, https://www.ibm.com/
ImageJ (FIJI) GNU General Public License, https://imagej.net/software/fiji/
Origin 2018 Microcal Software, https://microcal-origin.software.informer.com/
Clampfit 10.6 Molecular Devices, https://www.moleculardevices.com/
Statistics
All the data in this research were completed in double-blind condition. All the data were analyzed using IBM SPSS Statistics 27 or SigmaPlot and presented as mean ± SEM. Detailed descriptions of the statistical procedures of each figure in this article is approved with their related table. Statistical tests include two-tailed unpaired t tests, one-way ANOVA followed by the LSD post hoc test, and two-way ANOVA followed by two-tailed paired t tests. Significant results were defined as p < 0.05 (*p < 0.05, **p < 0.01, ***p < 0.001).
Results
Novel social stimulation promotes α-secretase activity in vHPC
The HPC, one of the first and most severely affected brain regions in AD patients (de Leon et al., 1989; Skoog, 2000; Scheff and Price, 2006; Villemagne et al., 2013), responds quickly to social interaction and is involved in social cognition (Buwalda et al., 2005; Tavares et al., 2015; Alexander et al., 2016; Okuyama et al., 2016; Montagrin et al., 2018; FeldmanHall et al., 2021), suggesting that social stimulation may alleviate the pathological progression of AD by regulating hippocampal activity. To identify the key hippocampal subregion that is most sensitive to social stimulation, we first administered social stimulation to WT mice and detected cell activity in the dorsal and ventral subregions of the HPC with c-Fos staining, which is an immediate early gene and has been widely used to reflect cellular activation in the nervous system. As shown in Figure 1A–D, compared with the group-housed mice (GH; in contact with familiar mice), stimulation with a novel unfamiliar mouse (SNS stimulation) effectively activated the expression of c-Fos in the CA1 cells of vHPC (vCA1), while intact in the dorsal hippocampal CA1 (dCA1) region.
SNS stimulation activates vHPC α-secretase. A, Illustration of GH and SNS treatment. B–D, Expression of c-Fos in vHPC is increased after SNS. Representative confocal images of dCA1 (top) and vCA1 (bottom); B, comparison of c-Fos positive cells in dCA1 (C) and vCA1 (D) between GH and SNS groups. Two-tailed unpaired t test. Scale bar: 100 µm. E–H, The activity of α-secretase is upregulated in SNS group. Representative images (E) and statistical graphs of ADAM10 (F), ADAM17 (G), and BACE1 (H). One-way ANOVA followed by the Fisher's LSD post hoc test. I–L, SNS induces expression of sAPPα. Representative images (I) and statistical graphs of sAPPα (J), sAPPβ (K), and sAPPα/β (L). One-way ANOVA followed by the Fisher's LSD post hoc test. For C and D, dots represent individual slices. For F–H and J–L, dots represent individual mice. N represents individual groups. Data are presented as mean ± SEM. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. Also see Extended Data Figures 1-1–1-3. The related statistical methods and information are shown in Extended Data Table 1-1.
Figure 1-1
Social stimulation activates ADAM 10 in WT and FAD4T mice A-B. Immunostaining data showing the expression of ADAM10 protein in WT vHPC is increased after SNS. Scale bar, 100μm. Two-tailed unpaired t-test. C. qPCR data showing the mRNA level of ADAM10 in WT vHPC is intact after SNS. Two-tailed unpaired t-test. D-E. Immunostaining data showing the expression of ADAM10 protein in FAD4T vHPC is increased after MNS. Scale bar, 100μm. Two-tailed unpaired t-test. Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 1-2. Download Figure 1-1, TIF file.
Figure 1-2
Social stimulation activates ADAM 17 in WT and FAD4T mice A-B. Immunostaining data showing the expression of ADAM17 protein in WT vHPC is increased after SNS. Scale bar, 100μm. Two-tailed unpaired t-test. C. qPCR data showing the mRNA level of ADAM17 in WT vHPC is intact after SNS. Two-tailed unpaired t-test. D-E. Immunostaining data showing the expression of ADAM17 protein in FAD4T vHPC is increased after MNS. Scale bar, 100μm. Two-tailed unpaired t-test. Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 1-3. Download Figure 1-2, TIF file.
Figure 1-3
The effect of social stimulation on BACE1 A-B. Immunostaining data showing the expression of BACE1 protein in WT vHPC is intact after SNS. Scale bar, 100μm. Two-tailed unpaired t-test. C. qPCR data showing the mRNA level of BACE1 in WT vHPC is intact after SNS. Two-tailed unpaired t-test. D-E. Immunostaining data showing the expression of BACE1 protein in FAD4T vHPC is decreased after MNS. Scale bar, 100μm. Two-tailed unpaired t-test. Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 1-4. Download Figure 1-3, TIF file.
Table 1-1
Statistical methods and information refers to Figure 1. Download Table 1-1, XLSX file.
Table 1-2
Statistical methods and information refers to Figure 1-1. Download Table 1-2, XLSX file.
Table 1-3
Statistical methods and information refers to Figure 1-2. Download Table 1-3, XLSX file.
Table 1-4
Statistical methods and information refers to Figure 1-3. Download Table 1-4, XLSX file.
α-Secretase is the physiological extracellular cleaving enzyme of APP and other type Ⅰ transmembrane proteins, and its activity is related to neuronal activity in the nervous system (Hoey et al., 2009; Tampellini et al., 2009). Therefore, we speculated that social stimulation may enhance α-secretase activity in the vHPC region. To verify this hypothesis, we collected the vHPC tissues from the SNS and GH mice and examined the expression of APP-related extracellular cleaving enzymes. We observed a significant increase in the expression of α-secretase (including mature ADAM10 and ADAM17) at 60 min post-SNS, while the expression of β-secretase BACE1 remained unchanged (Fig. 1E–H; Extended Data Figs. 1-1–1-3).
Subsequently, to determine whether changes in enzyme expression resulted in alterations in the production of APP cleavage products, we compared the expression levels and ratios of the major intermediate products sAPPα and sAPPβ, respectively (Fig. 1I–L). At 1 h post-SNS, the expression level of sAPPα, not sAPPβ, was increased, leading to an elevation in the sAPPα/sAPPβ ratio. Overall, these results show that novel social stimulation is able to activate vHPC and elevate local α-secretase expression and function in WT mice.
Social induced α-secretase activation is dCA2 activity and vHPC NMDAR dependent
Next, we wondered where the vHPC received social information input. Previous studies have shown that the dorsal CA2 region of the HPC (dCA2) is the key node for social information to transfer into the HPC and mediate social memory encoding and storage via the dCA2–vCA1 circuit. Therefore, we speculated that the changes in α-secretase and sAPPα caused by SNS also depend on the activity of dCA2 neurons. We bilaterally injected recombinant adeno-associated viruses (rAAVs) expressing a floxed engineered M4-muscarinic receptor [EF1α-DIO-hM4D(Gi)-EYFP] into the dCA2 region of adult hippocampal CA2-specific Amigo2-Cre mice and injected clozapine-N-oxide (CNO, 0.5 mg/kg, i.p.) at 20 min before the social stimulation to inhibit dCA2 neuronal activity. One hour after social treatment, we collected vHPC samples for testing (Fig. 2A). As shown in Figure 2B and C, after inhibiting the activity of dCA2 neurons, SNS stimulation was no longer able to cause the increase of ADAM10 and sAPPα expression compared with those in the control group.
Social induced α-secretase activation is dCA2 activity and vHPC NMDAR dependent. A, Schematic drawing of the experimental process. Scale bar: 100 µm. B–C, Upregulation of nonamyloidogenic pathways induced by SNS is suspended with chemical inhibition of dCA2. Sample images (B) and relative protein levels (SNS/GH; C). Two-tailed unpaired t test. D, Illustration of the experimental process. E–G, Inhibition of NMDAR occludes upregulation of nonamyloidogenic pathways induced by SNS. Sample images (E) and relative protein levels (SNS/GH) of saline group (F) and APV group (G). Two-tailed unpaired t test. Dots represent individual mice. Data are presented as mean ± SEM. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. The related statistical methods and information are shown in Extended Data Table 2-1.
Table 2-1
Statistical methods and information refers to Figure 2. Download Table 2-1, XLSX file.
NMDAR is one of the most important ionotropic glutamate receptors in neurons, mediating synaptic plasticity and neuronal activation, and previous researches have demonstrated that neuronal activity upregulates α-secretase and is related to NMDAR activation (Hoey, Williams and Perkinton, 2009; Wan et al., 2012). To verify whether the upregulation of α-secretase activity induced by SNS is NMDAR activation dependent, we bilaterally injected NMDAR antagonist AP5 into vCA1 regions before SNS and collected the vHPC tissues 1 h after SNS (Fig. 2D). As the results shown in Figure 2E–G, SNS-induced upregulation of α-secretase and sAPPα was blocked by AP5.
Therefore, we conclude that novel social stimulation is able to activate vCA1 cells and upregulate the expression of α-secretase and sAPPα in WT vHPC region, which are dCA2 neural activity and vHPC NMDAR dependent.
Novel social stimulation promotes vHPC α-secretase activity in FAD4T mice
To investigate whether the novel social stimulation-induced secretase expression alteration will regulate APP processing in AD models, we utilized a new transgenic mouse model based on the Aβ hypothesis, named FAD4T, which expresses the Swedish (KM670/671NL) and Indian (V717F) mutated APP gene, as well as the PSEN1 gene with M146V and L286V mutations, making it a multitransgenic AD model. In comparison with commonly used transgenic AD models, FAD4T mice exhibit accelerated amyloid deposition and noticeable memory deficits (Huang et al., 2023). Through Thioflavin T (Tht-t) staining and electrophysiological recording, we observed mild amyloid plaque deposition and LTP impairment in the vHPC of 2.5-month-old FAD4T mice, while aggravated at 3 months (Fig. 3A–D). We also performed three-chamber social test and NOR to assess the recognition ability in FAD4T mice. These results showed that FAD4T mice exhibited impaired NOR at 2.5 months of age, but normal social memory, which disappeared until 3 months of age both in male and female FAD4T mice (Fig. 3E–K and Extended Data Fig. 3-1). These results suggest that the time window between 2.5 and 3 months is a critical period for FAD4T mice to develop AD-like pathological deterioration and indicate that we can adopt novel social stimulation methods to intervene in the deficits in FAD4T mice from 2.5 months, which mimics the early stage of AD.
FAD4T presents impaired synaptic plasticity and memory at 3 months. A, B, Plaques appeared in the vHPC of FAD4T mice at 2.5 months and significantly increased at 3 months. Representative pictures of vHPC with Tht-t staining (A) and comparison of Tht-t positive areas among WT and FAD4T (B). Arrows refer to plaque. One-way ANOVA followed by the Fisher's LSD post hoc test. Scale bar: 100 µm. C, D, LTP of FAD4T appeared to be impaired at 2.5 months old. Time course of normalized averaged slopes of fEPSPs recorded from CA1 region following TBS (to induce LTP; C) and comparison of the mean LTP amplitude among experimental groups in WT and FAD4T (D). The above panels were representative traces from baseline (1) and 60 min after TBS (2). One-way ANOVA followed by the Fisher's LSD post hoc test. Calibration: 0.2 mV, 20 ms. E–H, Social memory is impaired in 3 month-old FAD4T. Schematic drawing of three-chamber social test (E), exploration time spent at social ability stage (F), exploration time spent (G), and social memory index (H) at social novelty stage. Two-way ANOVA followed by the paired t test (F, G), and one-way ANOVA followed by the Fisher's LSD post hoc test (H). I–K, Impaired performance of FAD4T in NOR test. Illustration of NOR (I), exploration time spent in objects (J), and discrimination index (K) at testing day. Two-way ANOVA followed by the paired t test (J) and two-tailed unpaired t test (K). For B and D, dots represent individual slices. For F, G, H, J, and K, dots represent individual mice. Data are presented as mean ± SEM. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. Also see Extended Data Figure 3-1. The related statistical methods and information are shown in Extended Data Table 3-1.
Figure 3-1
Female FAD4T presents impaired social memory at 3M. A-D. Social memory is impaired in 3M-female FAD4T. Schematic drawing of three-chamber social test (A), exploration time spent at social ability stage (B), exploration time spent(C) and social memory index (D) at social novelty stage. Two-way ANOVA followed by the paired t-test (B, C), and One-way ANOVA followed by the Fisher’s LSD post hoc test (D). E-G. Impaired performance of female FAD4T in novel object recognition (NOR) test. Illustration of NOR (E), exploration time spent in objects (F) and discrimination index (G). Two-way ANOVA followed by the paired t-test (F), and Two-tailed unpaired t-test (G). Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 3-2. Download Figure 3-1, TIF file.
Table 3-1
Statistical methods and information refers to Figure 3. Download Table 3-1, XLSX file.
Table 3-2
Statistical methods and information refers to Figure 3-1. Download Table 3-2, XLSX file.
To confirm whether novel social stimuli can induce similar biological changes in FAD4T mice like those in WT mice, we first compared the expression of secretases and their APP cleavage products in 2.5-month-old WT and FAD4T mice. According to the findings, the expressions of mature ADAM10, APP, and BACE1 in FAD4T mice were slightly increased than those of their WT littermates. The extracellular APP cleavage products sAPPα and sAPPβ showed large elevations, and Aβ was only deposited at FAD4T samples (Fig. 4A,B). Next, we wondered whether novel social stimulation could alter secretase activity and APP cleavage in FAD4T mice. As we did in WT mice, we delivered FAD4T mice SNS stimulation with an unfamiliar mouse and extracted their ventral hippocampal proteins 1 h later. As shown in Fig. 4C,D, SNS effectively increased the protein expression of mature ADAM10 and sAPPα in FAD4T mice. Meanwhile, we also found that the protein levels of BACE1 and its cleavage product sAPPβ were slightly reduced in FAD4T mice, together resulting in a very significant elevation of the sAPPα and sAPPβ ratio. Although this single novel stimulus altered the enzymatic activity associated with APP cleavage, it appeared to have no significant impact on the production of Aβ (Fig. 4C,D).
SNS promotes vHPC α-secretase activity in FAD4T mice. A, B, Compared with WT, APP splicing is upregulated in FAD4T. Representative images of APP splicing-related proteins (A) and statistical graphs of relative protein levels in WT versus FAD4T (B). Two-tailed unpaired t test. C, D, α-Secretase is activated and β-secretase is downregulated after SNS in FAD4T. Representative images of APP splicing-related proteins (C) and statistical graphs of relative protein levels in FAD4T-GH versus FAD4T-SNS (D). Two-tailed unpaired t test. E, F, Inhibition of α-secretase occludes upregulation of nonamyloidogenic APP processing of FAD4T induced by SNS. Representative images of APP splicing-related proteins (E) and statistical graphs of relative protein levels in FAD4T-GI254-GH versus FAD4T-GI254-SNS (F). Two-tailed unpaired t test. Dots represent individual mice. Data are presented as mean ± SEM. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. The related statistical methods and information are shown in Extended Data Table 4-1.
Table 4-1
Statistical methods and information refers to Figure 4. Download Table 4-1, XLSX file.
To examine whether the SNS-induced secretases and APP cleavage product alteration were caused by the increased activity of α-secretase, we bilaterally injected the selective α-secretase inhibitor GI254023X (GI254, 100 µM, 1 µl) into vHPC 20 min before social stimulation and collected the vHPC tissues for testing 1 h after the stimulation. The results showed that after inhibiting the activity of α-secretase, SNS could no longer induce changes in the levels of ADAM10 and sAPPα/β ratio (Fig. 4E,F). Interestingly, the level of sAPPα was significantly lower than that of the GH group, indicating that in the presence of α-secretase inhibitors, social stimuli may inhibit the nonamyloidogenic cleavage pathway of APP or promote sAPPα degradation.
In summary, to investigate whether novel social stimulation can alter the APP cleavage pathway in AD model mice, we introduced the FAD4T mouse model and identified the critical period of exacerbation of AD symptoms in this model. Subsequently, we found that SNS stimulation was sufficient to promote α-secretase activity and enhance the nonamyloidogenic cleavage pathway, while inhibiting the amyloid cleavage pathway of APP in the vHPC tissues of 2.5-month-old FAD4T mice. Additionally, we proved that the alterations of enzyme expression and APP cleavage products were dependent on α-secretase activity.
MNS stimulation restricts Aβ generation in FAD4T mice
Considering that the course of AD is a developmental process, a single social stimulus may not be sufficient to reverse pathological deterioration within the 2.5–3 months time window. We adopted 14 d MNS stimulation strategies, that is, giving one unfamiliar mouse to the trained mice for social stimulation every day for 14 d, one day per mouse for 1 h. Because the social memory of mice can only be maintained for a couple of days (Wu et al., 2021), we used seven unfamiliar mice (S1–S7) for training from day 1 to day 7, and on days 8–14, we used the seven mice for the second round of training as well. On the other hand, the GH mice have also been raised in groups but can only socialize with familiar mice in the same litter every day. On the 15th day, we collected the vHPC tissues of two groups of mice for immunoblotting (Fig. 5A).
MNS promotes nonamyloidogenic pathway in FAD4T mice. A, Schematic diagram of the experimental timeline. B, C, α-Secretase is activated and β-secretase is downregulated after SNS in FAD4T. Representative images of APP splicing-related proteins (B) and graphs of relative protein levels (C) in FAD4T-GH and FAD4T-MNS. Two-tailed unpaired t test. D, E, MNS reduces plaques of FAD4T mice. Representative pictures of vHPC with Tht-t staining (D) and comparison of Tht-t positive area (E). Arrows refer to plaques. Two-tailed unpaired t test. Scale bar: 100 µm. F, G, MNS increases LTP amplitude in FAD4T. Time course of normalized averaged slopes of fEPSPs recorded from CA z1 region following TBS (F) and comparison of the mean LTP amplitude between two groups (G). The above panels were representative traces from baseline (1) and 60 min after TBS (2). Two-tailed unpaired t test. Calibration: 0.2 mV, 20 ms. Dots represent individual mice. Data are presented as mean ± SEM. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001. Also see Extended Data Figure 5-1. The related statistical methods and information are shown in Extended Data Table 5-1.
Figure 5-1
MNS stimulation reduces Aβ in FAD4T mice. A. Schematic diagram of the experimental timeline. B-D. Immunostaining data showing the expression of Aβin FAD4T vHPC is decreased after MNS (7D) and MNS (14D). Scale bar, 100μm. One-way ANOVA followed by the Fisher’s LSD post hoc test. Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 5-2. Download Figure 5-1, TIF file.
Table 5-1
Statistical methods and information refers to Figure 5. Download Table 5-1, XLSX file.
Table 5-2
Statistical methods and information refers to Figure 5-1. Download Table 5-2, XLSX file.
We found that MNS had notable effects on secretase activity and their cleavage products when compared with those in GH mice. Specifically, MNS stimulation effectively activated α-secretase, while inhibiting β-secretase activity, leading to an increase in sAPPα/β ratio (Fig. 5B,C; Extended Data Figs. 1-1–1-3). In addition, we also measured the levels of Aβ generation and aggregation. These results exhibited that the level of Aβ oligomers and the number of plaques were both significantly reduced in the MNS group compared with those in the age-matched GH FAD4T mice (Fig. 5B–E). To investigate the relationship between MNS and the rate of Aβ plaque formation, we trained the 2.75-month-old FAD4T male mice with 7 d MNS and performed Aβ immunostaining. The results showed that the intensity and numbers of Aβ plaques were lower than those in the GH group but significantly higher than those in the 14 d MNS group (Extended Data Fig. 5-1), suggesting a negative correlation between the intensity of the social stimulus and the rate of Ab plaque formation between 2.5- and 3-month-old FAD4T mice. Surprisingly, we found that TBS induced significant LTP in the vHPC slices of MNS group mice, but not in GH group mice. It is interesting that before MNS stimulation, the 2.5-month-old FAD4T mice exhibited significant TBS-LTP deficiency, indicating that MNS stimulation reversed the LTP loss in AD mice (Fig. 5F,G).
These results showed that MNS stimulation effectively altered the activities of APP extracellular secretases, restricted Aβ generation, and reversed the LTP deficit in FAD4T mice, suggesting it may have the potential to restore cognitive impairment in AD mouse models.
MNS stimulation restores cognition impairment in FAD4T mice
To investigate whether multiple social stimuli can improve the cognitive behavior of early-stage AD mice, we conducted behavioral tests on the second day after completing MNS training. Firstly, to eliminate the interference of multiple stimuli on the mental state of mice, we conducted the OFT to measure the anxiety and autolocomotion ability of mice. The results showed that both the male and female mice stimulated by MNS had comparable autonomous motor ability and anxiety states compared with GH mice (Fig. 6A–E; Extended Data Fig. 6-1A–E). Subsequently, we conducted the three-chamber social test and NOR and found that the GH group FAD4T mice exhibited cognitive impairment in both behavioral patterns, while the MNS-trained mice showed significant social memory and NOR ability in both male and female groups, indicating that MNS training effectively intervened in the cognitive deficits in early AD model mice (Fig. 6F–L; Extended Data Fig. 6-1F–L).
MNS stimulation rescues social and NOR impairment in FAD4T mice. A–E, MNS stimulation does not affect motor ability and anxiety level of FAD4T. Illustration of OFT (A) and comparison of distance (B), speed (C), time in center area (D), and enter times in center area (E) between FAD4T-GH and FAD4T-MNS. Two-tailed unpaired t test. F–I, Social memory of FAD4T is improved with MNS. Illustration of three-chamber test (F) and exploration time spent in social ability stage (G) and exploration time spent (H) and social memory index (I) at social novelty stage. Two-way ANOVA followed by two-tailed paired t test (G, H) and two-tailed unpaired t test (I). J–L, MNS stimulation improved the performance of FAD4T in NOR test. Illustration of NOR test(J) and comparison of exploration time toward novel and familiar object (K) and discrimination index (L) between FAD4T-GH and FAD4T-MNS. Two-way ANOVA followed by two-tailed paired t test (K) and two-tailed unpaired t test (L). Dots represent individual mice. Data are presented as mean ± SEM. ***p < 0.001. Also see Extended Data Figures 6-1 and 6-2. The related statistical methods and information are shown in Extended Data Table 6-1.
Figure 6-1
MNS stimulation rescues social and novel object recognition impairment in female FAD4T mice. A-E. MNS stimulation doesn’t affect motor ability and anxiety level of female FAD4T. Illustration of OFT (A) and comparison of distance (B), speed (C), time in center area (D) and enter times in center area (E) between FAD4T-GH and FAD4T-MNS. Two-tailed unpaired t-test. F-I. Social memory of female FAD4T is improved with MNS. Illustration of three-chamber test (F) and exploration time spent in social ability stage (G) and, exploration time spent (H) and social memory index (I) at social novelty stage. Two-way ANOVA followed by two-tailed paired t-test (G, H), and two-tailed unpaired t-test (I). J-L. MNS stimulation improved the performance of female FAD4T in novel object recognition (NOR) test. Illustration of NOR test (J) and comparison of exploration time toward novel and familiar object (K) and discrimination index (L) between FAD4T-GH and FAD4T-MNS. Two-way ANOVA followed by two-tailed paired t-test (K), and two-tailed unpaired t-test (L). Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 6-2. Download Figure 6-1, TIF file.
Figure 6-2
The effect of social stimulation on phospho-Tau in WT and FAD4T mice A-B. SNS (60min) induces reduced expression of phospho-Tau in WT. One-way ANOVA followed by the Fisher’s LSD post hoc test. C-D. No significant difference of phospho-Tau expression between WT and FAD4T. Two-tailed unpaired t-test. E-F. No significant difference of phospho-Tau expression between FAD4T-GH and FAD4T-SNS. Two-tailed unpaired t-test. G-H. MNS failed to induce reduced expression of phospho-Tau in FAD4T. Two-tailed unpaired t-test. Dots represent individual mice. Data are presented as mean ± S.E.M. ***p< 0.001. The related statistical methods and information are shown in Table 6-3. Download Figure 6-2, TIF file.
Table 6-1
Statistical methods and information refers to Figure 6. Download Table 6-1, XLSX file.
Table 6-2
Statistical methods and information refers to Figure 6-1. Download Table 6-2, XLSX file.
Table 6-3
Statistical methods and information refers to Figure 6-2. Download Table 6-3, XLSX file.
Discussion
Social communication has been used as an adjunctive therapy for AD, but its mechanism is not clear. In this study, we discovered that, in contrast to interaction with familiar mice, a SNS stimulation from unfamiliar mice enhanced the activity of the α-secretase in the vHPC region in both WT and AD model mice and as a result, encouraged the APP nonamyloidogenic pathway. Furthermore, MNS stimuli that were administered to AD model mice during the early stages of the disease were successful in reducing Aβ production and dementia symptoms. Our research suggests that novel social communication, which depends on the α-secretase activation, may work as a noninvasive, nonpharmacological treatment for AD.
In mammals, social stimulation activates numerous brain regions, including the entorhinal cortex, HPC, prefrontal cortex, etc., among which the HPC is considered essential for social memory formation and short-term memory storage and is one of the most vulnerable areas of AD patients (Scheff and Price, 2006; Santini et al., 2013; Kingsbury et al., 2019; Donegan et al., 2020; Liu et al., 2020; Watarai et al., 2021). Through c-Fos labeling, we found that social stimulus triggered cell transcription in the vHPC region rather than the dorsal hippocampal region. However, the elevation of vHPC ADAM10/17 and sAPPα caused by social stimulation were blocked by dCA2 inhibition (Fig. 2A–C), suggesting different roles of dorsal and ventral HPC in social information processing. Previous studies have shown that the activity of α-secretase is regulated by neuronal activity, and in some cases, by NMDA receptors (Hoey et al., 2009; Tampellini et al., 2009; Kim et al., 2010; Wan et al., 2012). In line with this, social stimulation in our tests increased vHPC cell activation as well as the activity/maturation of the α-secretase family members ADAM10/17, which could be inhibited by the NMDAR inhibitor AP5 (Fig. 2D–G). Comparing the increase of mature ADAM10/17 protein levels in whole vHPC tissue, the number of c-Fos positive cells is relatively sparse. We believe that after social stimulation, a large number of cells in the vHPC region undergo NMDAR activation and subsequent activation of mature ADAM10/17 proteins. Among these activated cells, only a small portion may engage in transcription, information storage, and transformation into memory engram cells.
Previous sociological research findings have shown that various social stimuli are able to alleviate the cognitive decline rate of AD patients (Fratiglioni et al., 2000; Fancourt et al., 2020; Sommerlad et al., 2023). Our data emphasized that experiencing novel social communication effectively reduced the pathological manifestations of AD model mice. It was noteworthy that after MNS stimulation, FAD4T mice exhibited intact social and novel object memories, whereas FAD4T mice stimulated with familiar mice exhibited significant cognitive deficits. These results suggest that novel social interaction is a key fact that effectively reduces the pathological manifestations of AD model mice. Meanwhile, in protein testing, we found that mice undergoing MNS stimulation exhibited increased α-secretase and sAPPα, while decreased sAPPβ and Aβ accumulation. Noteworthy is the fact that in FAD4T mice, after experiencing MNS stimuli, the protein levels of BACE1 and sAPPβ were significantly reduced (Fig. 5B,C; Extended Data Fig. 1-3D–E). These findings were consistent with previous studies that increased α-secretase function and sAPPα were able to inhibit BACE1 and prevent Aβ formation (Postina et al., 2004; Obregon et al., 2012; Suh et al., 2013; Peters-Libeu et al., 2015). Considering the competitive relationship between α-secretase and β-secretase in the formation of Aβ (Hampel and Hardy, 2021), we speculate that the function of MNS stimulation in alleviating the pathological process of AD might be dependent on the enhancement of α-secretase function, like the results obtained by previous genetic and pharmacological studies that inhibited Aβ generation by enhancing α-secretase function (Postina et al., 2004; Kuhn et al., 2010).
Furthermore, we found that MNS stimulation might also have a therapeutic effect on early AD symptoms. We detected mild cognitive and synaptic plasticity defects in FAD4T mice aged 2.5 months (Fig. 3; Extended Data Fig. 3-1). After MNS training, the mice exhibited normal NOR and TBS-induced hippocampal LTP (Figs. 5, 6; Extended Data Fig. 6-1), demonstrating that MNS stimuli led to the restoration of animal cognitive capacity and neuronal function. We speculated that this would be due to the inherent Aβ clearance ability in the AD animals (Tarasoff-Conway et al., 2015; Rasmussen et al., 2018), together with the inhibition of Aβ production, resulting in reduced Aβ accumulation and neural function recovery.
It should be noted that, in addition to the enhanced nonamyloidogenic pathway of APP in vHPC, there may be other neural mechanisms at play. For instance, social stimulation and activation of α-secretase in the vHPC could also trigger other cellular mechanisms, such as tau activity alteration (Extended Data Fig. 6-2), social memory encoding, cell transcription, and hydrolysis of other transmembrane proteins (Sahin et al., 2004; Suzuki et al., 2012; Okuyama et al., 2016; Qian, Shen and Wang, 2016; Yuan et al., 2017). These mechanisms may modulate neural activity and subsequent AD pathological processes over time. Furthermore, social stimulation has been shown to activate various brain regions, including the entorhinal cortex, medial septum, and prefrontal cortex (Perkins et al., 2017; Tanimizu et al., 2017; Chen and Hong, 2018; Gangopadhyay et al., 2021), which could engage different cellular mechanisms involved in the regulation of AD progression. Moreover, similar to previous studies on AD treatment involving physical training and different diets, social stimulation may also have an impact on body systems beyond the nervous system (Valenzuela et al., 2020; Agarwal et al., 2023). And it should be noted that FAD4T mice used in this study exhibit rapid APP processing, which is an advantage but also a limitation and cannot reflect the situation well in AD patients.
In summary, we found that SNS or MNS stimuli were able to alleviate or even treat the pathological phenotypes of early AD model mice. These improvements were attributed to the activation of α-secretase and the inhibition of the amyloidogenic pathway of APP in vHPC. Our findings provide insights into the mechanisms through which social stimulation enhances cognition in AD and highlight the preventive and therapeutic potential of novel social communication and α-secretase activation during the early stages of the disease.
Footnotes
This work was supported by grants from STI2030-Major Projects (2021ZD0204000, W.X. and A.L.; 2022ZD0205900, A.L.), Natural Science Foundation of China (NSFC 91632201, W.X.; NSFC 31970958, A.L.), Natural Science Foundation of Jiangsu Province (BK20211561, A.L.), Basic Research Project of Leading Technology of Jiangsu Province (BK20192004, A.L.), Shenzhen Science and Technology Innovation Foundation (2021Szvup028, A.L.), Guangdong Key Project (2018B030335001, W.X.), NSFC-Guangdong Joint Fund (U20A6005, J.G.), the Canadian Institutes of Health Research (CIHR PJT155959, CIHR PJT168922, Z.J.), and Canadian Natural Science and Engineering Research Council (NSERC RGPIN341498, RGPIN06295, Z.J.).
↵*Q.R. and S.W. contributed equally to this work.
The authors declare no competing financial interests.
- Correspondence should be addressed to An Liu at liu_an{at}seu.edu.cn.