Abstract

   High‐salt (HS) diet is an established risk factor for cognitive
   impairment, but the underlying mechanisms remain unclear. This study
   reveals that HS diet reduces SHANK1, a key postsynaptic scaffolding
   protein, via downregulation of the PKA/CREB pathway, leading to
   synaptic dysfunction and cognitive deficits in rats. RNA sequencing of
   HS‐fed rat hippocampi showed downregulation of cAMP signaling and
   SHANK1 expression. Pharmacological inhibition of PKA/CREB reduced
   SHANK1 levels and impaired dendritic structure and synaptic function,
   while PKA activation restored CREB activity and SHANK1 expression,
   reversing HS‐induced deficits. Notably, CREB activation is essential
   for SHANK1 regulation, as a CREB mutant (S133A) blocked the effects of
   PKA activation, and a constitutively active CREB (S133D) prevented
   SHANK1 downregulation. These findings highlight the PKA/CREB/SHANK1
   pathway as a potential therapeutic target for HS‐induced cognitive
   dysfunction.

   Keywords: high‐salt diet, SHANK1, PKA/CREB, synaptic dysfunction,
   cognitive deficits
     __________________________________________________________________

   The study identifies a novel mechanistic link between long‐term HS diet
   and cognitive impairment, wherein PKA/CREB axis inactivation leads to
   SHANK1 reduction, synaptic damage, and cognitive deficits.

   graphic file with name ADVS-12-e02099-g006.jpg

1. Introduction

   High salt intake remains widespread globally, despite ongoing public
   health efforts to raise awareness of its associated health risks.^[
   [44]^1 , [45]^2 , [46]^3 ^] Accumulating evidence from rodent models
   strongly suggests that 4–12 weeks of HS intake causes cognitive
   impairments, in addition to cardiovascular complications.^[ [47]^4 ,
   [48]^5 , [49]^6 , [50]^7 ^] A 3‐year prospective study involving 1262
   adults found that lower dietary sodium intake was associated with a
   reduced incidence of cognitive impairment.^[ [51]^8 ^] However, the
   precise molecular mechanisms underlying HS‐induced cognitive defects
   remain poorly understood.

   HS diet is an independent risk factor for dementia and has been linked
   to cerebral small vessel disease, which underlies vascular cognitive
   impairment.^[ [52]^4 , [53]^5 , [54]^9 ^] Recent findings by Facaro
   et al. reported that the HS diet induces gut‐mediated adaptive immune
   activation, leading to increased IL‐17 levels and downregulation of
   endothelial nitric oxide synthase. These results in nitric oxide
   deficiency, decreased calpain nitrosylation, and subsequent activation
   of calpain and CDK5, promoting cognitive deficits via tau
   hyperphosphorylation.^[ [55]^10 ^] Interestingly, tau phosphorylation
   in the neocortex was observed to increase after 4 weeks of HS intake,
   whereas in the hippocampus, it was not evident until 12 weeks. Our
   previous studies demonstrated significant cognitive deficits following
   8–9 weeks of chronic HS intake, yet these were not associated with tau
   phosphorylation, Aβ deposition, or neuronal loss in the hippocampus.^[
   [56]^11 ^] Given that dementia, including Alzheimer's disease (AD), is
   characterized by synaptic dysfunction, formation of senile plaques,
   hyperphosphorylated tau tangles, neuroinflammation, and apoptotic cell
   death.^[ [57]^12 , [58]^13 , [59]^14 , [60]^15 , [61]^16 , [62]^17 ^]
   This suggests that synaptic dysfunction is preexisting in Aβ
   aggregation and tau pathology and may play a key role in HS
   diet‐induced cognitive impairments.

   SHANK is a multidomain scaffold protein that mediates the integration
   of multiple synaptic proteins.^[ [63]^18 , [64]^19 , [65]^20 ^] The
   specific localization of SHANK proteins at postsynaptic sites of
   excitatory synapses in the brain suggests their roles in organizing
   cytoskeletal and signaling complexes at specialized cell junctions.^[
   [66]^18 , [67]^21 ^] SHANK proteins are implicated in the
   pathophysiology of autism spectrum disorders and are encoded by the
   SHANK1, SHANK2, and SHANK3 genes.^[ [68]^20 , [69]^21 , [70]^22 ,
   [71]^23 , [72]^24 , [73]^25 ^] However, the physiological role of SHANK
   proteins and their involvement in cognitive impairments, particularly
   in the context of HS diet, remains largely unexplored.

   Based on the theoretical background and our previous findings, HS diet
   induces hippocampal‐dependent spatial memory impairment, accompanied by
   the inhibition of cAMP‐response element‐binding protein (CREB) activity
   and decreased phosphorylation at Ser133 of CREB.^[ [74]^11 ^] It is
   well‐established that members of the CREB family are key regulators of
   gene expression in response to cAMP signaling, via phosphorylation by
   protein kinase A (PKA). CREB is a nuclear protein that modulates the
   transcription of genes containing cAMP‐responsive elements within their
   promoters. Increased intracellular cAMP concentrations trigger the
   phosphorylation and activation of CREB, a key transcription factor that
   regulates numerous memory‐related processes.^[ [75]^26 , [76]^27 ,
   [77]^28 , [78]^29 , [79]^30 , [80]^31 ^]

   In this report, we demonstrate that SHANK1 reduction leads to
   HS‐related cognitive deficits. Additionally, downregulation of SHANK1
   exacerbates synaptic dysfunction and cognitive deficits, while
   upregulation of the PKA/CREB pathway increases SHANK1 expression,
   thereby improving synaptic damage and mitigating cognitive deficits.
   This study is the first to report on the dysfunction of the
   PKA/CREB/SHANK1 axis in HS diet‐induced learning and memory
   impairments, suggesting that enhancing PKA activity to increase SHANK1
   levels may represent a promising therapeutic approach for treating
   HS‐induced cognitive dysfunction.

2. Results

2.1. High‐Salt Diet Induces Synaptic Dysfunction and Cognitive Impairments

   High salt is a well‐established risk factor for cognitive decline.^[
   [81]^7 , [82]^32 , [83]^33 ^] In this study, 2‐month‐old SD rats were
   fed an HS diet (8% NaCl) for 9 weeks to establish an HS model.^[
   [84]^10 , [85]^11 ^] Behavioral, electrophysiological, and biochemical
   analyses were performed (Figure [86]1A). A series of behavioral tests
   were conducted to assess cognitive function. The open field test showed
   no significant differences in total distance covered (Figure [87]S1A,
   Supporting Information) and the number of zone crossings (Figure
   [88]S1B, Supporting Information) between groups, indicating that HS
   intake did not affect motor function. Social ability, assessed via the
   three‐chamber social test, revealed a significant reduction in the
   frequency of contact with unfamiliar rats in the HS group (Figure
   [89]S1C, Supporting Information), though no differences were observed
   in the time spent in the novel box (Figure [90]S1D, Supporting
   Information). Elevated plus maze (EPM) test results show that the HS
   diet significantly increased the open arm entry ratio without affecting
   the duration of the open arm ratio (Figure [91]S1E,F,Supporting
   Information). These results suggest that a long‐term HS diet leads to
   anxiety and social disorders in rats.

Figure 1.

   Figure 1
   [92]Open in a new tab

   High‐salt diet induces synaptic dysfunction and cognitive impairments
   A) Experimental design schematic. Healthy male SD rats were randomly
   assigned to two groups: the control group received a normal diet, while
   the model group was subjected to HS diet (8% NaCl) for 9 weeks.
   Following dietary treatment, behavioral, electrophysiological, and
   biochemical assessments were conducted. B) The novel object recognition
   test showed the recognition index at 2 h and 24 h (n  = 19), (C). D–F)
   Morris water maze test: swimming speed (D), the spatial memory was
   tested on the 6th day by removing the platform the number of crossing
   the position of the target platform E) and the time spent in the target
   quadrant F) were measured, (n = 19). Fear conditioning test showed
   freezing duration was measured during the 3 min memory test at 2 h G)
   or 24 h H) after conditioning, (n  = 19). I, J) The firing rate of AP
   was measured, (n = 18–24 neurons from 3 rats). K) Hippocampal CA3‐CA1
   LTP was recorded by the MED64 system. Normalized CA3‐CA1 fEPSP mean
   slope recorded from the CA1 dendritic region in acute hippocampal
   slices. Quantitative analyses for fEPSP measured 60–90 min after HFS
   relative to baseline L), (n = 3 slices from 3 rats). M) For pairwise
   pulse ratio experiments, paired stimuli with a 50 ms interval were
   delivered and the ratio was calculated as the glutamate receptor
   amplitude ratio (EPSC2 / EPSC1) N). Rat hippocampal neurons were
   subjected to whole‐cell patch‐clamp recording of mEPSCs O) amplitude
   and frequency of mEPSCs of control and HS neurons P), (n = 9–11 neurons
   from 3–4 rats). The miniature inhibitory postsynaptic potential (mIPSC)
   of the GAMA receptor was recorded by whole‐cell patch‐clamp Q),
   amplitude and frequency of mIPSCs of control and HS neurons R). (n =
   12–14 neurons from 3–4 rats). S–W) The representative images showing
   transmission electron microscope (Scale bar, 0.5µm) (S), Quantitative
   analysis of the synaptic density (n = 10 from 3 rats per group) (T),
   the length of postsynaptic density (U), the width of postsynaptic
   denses (V) and the release of presynaptic vesicles (W) were measured
   (n  = 40–50 synapses from 3 rats). Data were expressed as mean ± SD, ^*
   p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001 vs control.
   Statistical details were provided in [93]Supporting Information.

   To evaluate learning and memory, we performed the novel object
   recognition test, the Morris water maze (MWM), and fear conditioning
   experiments. In the novel object recognition test, rats in the HS group
   had a significantly reduced recognition index at 2 and 24 h for novel
   objects compared to controls (Figure [94]1B,C). The MWM results showed
   no differences in swimming speed between groups (Figure [95]1D). On day
   6 after the platform was removed, HS rats exhibited significantly fewer
   platform crossings (Figure [96]1E) and spent less duration in the
   target quadrant (Figure [97]1F). In the fear conditioning test,
   freezing time in the HS group was significantly reduced at both 2 h
   (Figure [98]1G) and 24 h post‐conditioning (Figure [99]1H). These data
   demonstrate that the HS diet impairs memory in rats.

   The hippocampus, a critical region of the brain involved in memory and
   spatial navigation, is essential for synaptic plasticity, the cellular
   basis of learning and memory.^[ [100]^34 , [101]^35 , [102]^36 ^]
   Anatomically, the hippocampus consists of distinct subregions: cornu
   ammonis area 1(CA1), cornu ammonis area 3(CA3), and the dentate gyrus
   (DG), with the CA1 region particularly vulnerable to injury and
   stress.^[ [103]^34 , [104]^37 , [105]^38 ^] To explore the mechanisms
   underlying HS diet‐induced cognitive deficits, we examined hippocampal
   CA1 neuronal activity, HS rats showed reduced action potential (AP)
   firing (Figure [106]1I,J). Additionally, we examined hippocampal CA1
   synaptic plasticity. Long‐term potentiation (LTP) experiments showed
   that HS‐fed rats exhibited a significant reduction in the slope of
   field excitatory postsynaptic potentials (fEPSP) following
   high‐frequency stimulation (HFS) compared to controls
   (Figure [107]1K,L). To investigate the underlying synaptic transmission
   defects, the paired‐pulse ratio (PPR) was assessed to evaluate
   presynaptic vesicle release.^[ [108]^39 , [109]^40 ^] PPR analysis
   revealed no significant differences between the two groups at
   inter‐stimulus intervals of 50 ms (Figure [110]1M,N), as well as at 100
   ms, and 150 ms (Figure [111]S2A,B,Supproting Information). Whole‐cell
   patch‐clamp recordings were performed to measure miniature excitatory
   postsynaptic currents (mEPSC) (Figure [112]1O) and miniature inhibitory
   postsynaptic currents (mIPSC) (Figure [113]1Q) in the CA1 region. HS
   intake severely impaired excitatory synaptic transmission
   (Figure [114]1P) but had no effect on inhibitory synaptic transmission
   (Figure [115]1R). Moreover, transmission electron microscopy analysis
   revealed that the number of excitatory synapses per area in the CA1
   region was significantly lower in HS‐fed rats compared to controls
   (Figure [116]1S, T). The length of the postsynaptic density (PSD) in
   individual excitatory synapses was also significantly reduced in the HS
   group (Figure [117]1U), with no significant changes in PSD width
   (Figure [118]1V) or synaptic vesicle count (Figure [119]1W). These
   findings confirm that a long‐term HS diet induces synaptic dysfunction
   and cognitive deficits.

2.2. High‐Salt Diet Significantly Decreases the Expression of SHANK1

   To further investigate the molecular mechanisms underlying the synaptic
   transmission deficits observed in HS‐fed rats, we performed RNA
   sequencing analysis. A volcano plot (Figure [120]S3A, Supporting
   Information) and heatmap (Figure [121] 2A) illustrate the differential
   expression of genes (DEGs) between the HS diet and control groups. Gene
   Ontology (GO) enrichment (Figure [122]S3B, Supporting Information) and
   KEGG pathway analyses (Figure [123]2B, [124]S3C,Supporting Information)
   revealed significant changes in synapse‐related pathways. Visualization
   of the top enriched cellular components and core enrichment genes
   identified from GO and KEGG analyses are shown (Figure [125]S3D, E,
   Supporting Information). A Venn diagram comparing DEGs from our RNA‐seq
   data and the [126]GSE36980 dataset revealed 14 co‐upregulated and 40
   co‐downregulated genes (Figure [127]2C). We further examined the
   intersection between these co‐upregulated and co‐downregulated genes
   (identified in Figure [128]2C) and the core enrichment genes in
   synapse‐related KEGG pathways, specifically the synaptic vesicle cycle,
   glutamatergic synapse, and GABAergic synapse pathways, in the high‐salt
   (HS) diet rat model. This overlap was visualized in a Venn diagram,
   identifying SHANK1 and DNM3 as key genes (Figure [129]2D). The
   protein‐protein interaction (PPI) network of core enrichment genes in
   the synaptic vesicle cycle, glutamatergic synapse, and GABAergic
   synapse pathways (Figure [130]2E) shows that SHANK1 plays a central
   role. Our data showed that SHANK1 and SHANK2 are significantly reduced,
   but there were no differences in SHANK3 in RNA sequencing
   (Figure [131]2F). Quantitative PCR (qPCR) results corroborated these
   findings, demonstrating a significant reduction in SHANK1 mRNA levels,
   with no notable differences in SHANK2 and SHANK3 expression
   (Figure [132]2G). This was further confirmed by immunofluorescence
   analysis (Figure [133]2H) and western blotting (Figure [134]2I), both
   of which showed a marked decrease in SHANK1 protein levels in the
   hippocampal CA1 region of HS‐fed rats. These results suggest that HS
   diet specifically reduces SHANK1 expression, probably contributing to
   synaptic dysfunction in the hippocampus.

Figure 2.

   Figure 2
   [135]Open in a new tab

   High‐salt diet significantly decreases the expression of SHANK1 A)
   Heatmap illustrating the expression levels of differentially expressed
   genes (DEGs) identified from RNA‐seq analysis of a HS diet rat model.
   B) KEGG pathway enrichment analysis results (|NES| > 0.5, p.adjust <
   0.05, q‐value < 0.25). The core enrichment genes in the identified
   pathways are presented, including their average expression, logFC, and
   p‐value. C) Venn diagram showing the overlap between DEGs (p‐value <
   0.05) in the [136]GSE36980 dataset and RNA‐seq results from the HS diet
   rat model. A total of 14 co‐upregulated and 40 co‐downregulated genes
   were identified. D) Venn diagram illustrating the intersection between
   the co‐upregulated/downregulated genes identified in Figure [137]2C and
   the core enrichment genes in synapse‐related KEGG pathways (synaptic
   vesicle cycle, glutamatergic synapse, and GABAergic synapse pathways)
   in the HS diet rat model. SHANK1 and DNM3 were identified. E) PPI
   network of the core enrichment genes in the synaptic vesicle cycle,
   glutamatergic synapse, and GABAergic synapse pathways. F) SHANK1
   /SHANK2/SHANK3 expression levels in RNA sequencing were measured. G)The
   mRNA levels of SHANK1 /SHANK2/SHANK3 were measured (n = 4). The
   immunofluorescence H) and Western blotting I) were measured to the
   SHANK1 level of the hippocampus CA1 (n = 3). Data were expressed as
   mean ± SD, ^* p < 0.05,^** p < 0.01 vs control. Statistical details
   were provided in [138]Supporting Information.

2.3. Down‐Regulation of SHANK1 Leads to Cognitive Impairments and Synaptic
Dysfunction

   To evaluate the consequences of reduced SHANK1 in the hippocampus, we
   analyze the [139]GSE36980 dataset. Bioinformatics analysis revealed
   that the expression of SHANK1 was reduced across various brain regions,
   including the frontal cortex (FC), temporal cortex (TC), and
   hippocampus (HP) in Alzheimer's disease (AD), based on the
   [140]GSE36980 dataset (Figure [141]S4A, Supporting Information). Using
   the [142]GSE36980 dataset, correlation analysis further examined the
   expression levels of SHANK1, SHANK2, and SHANK3 in relation to age and
   gender (Figure [143]S4B, Supporting Information). Regarding the issue
   of sex‐specific effects, we performed a sex‐stratified analysis within
   the [144]GSE36980 dataset and results show that the correlation between
   SHANK1 expression and age is particularly strong in males (Figure
   [145]S4C, Supporting Information). Since the [146]GSE36980 dataset
   lacks detailed clinical and pathological data on Alzheimer's disease
   patients, we are unable to directly assess the correlation between
   SHANK1 and clinical disease severity within this dataset. Correlation
   analysis in [147]GSE84422 reveals a negative association between SHANK1
   expression and both neurofibrillary tangle density and clinical
   dementia rating, suggesting a potential link between SHANK1 levels and
   disease severity in Alzheimer's disease (Figure [148]S4D, Supporting
   Information). To assess the functional consequences of SHANK1
   downregulation on cognition, we injected AAV‐vector or AAV‐Sh‐SHANK1
   bilaterally into the hippocampal CA1 region of 8‐week‐old SD rats.
   Behavioral, electrophysiological, and biochemical assessments were
   subsequently conducted (Figure [149]3A). Western blotting confirmed a
   significant SHANK1 reduction in the Sh‐SHANK1 group (Figure [150]3B,C).
   A series of behavioral tests were performed to evaluate cognition. The
   open field test showed no significant differences in total distance
   traveled (Figure [151]S5A, Supporting Information) or zone crossing
   (Figure [152]S5B, Supporting Information) between the control and
   Sh‐SHANK1 groups. However, in the elevated plus maze test, the
   Sh‐SHANK1 rats exhibited a significant increase in the open arm entry
   ratio without affecting the duration in the open arm ratio (Figure
   [153]S5C,D, Supporting Information), indicating heightened anxiety‐like
   behavior. Cognitive function was assessed using the Morris water maze
   and fear conditioning tests. In the Morris water maze, Sh‐SHANK1 rats
   demonstrated significantly longer latency to find the hidden platform
   during training (Figure [154]3E), with no significant differences in
   swimming speed (Figure [155]3D). On the probe trial, time spent in the
   target quadrant (Figure [156]3F) and platform crossing (Figure [157]3G)
   were significantly reduced in the Sh‐SHANK1 group. In the fear
   conditioning test, freezing time was markedly decreased in the
   Sh‐SHANK1 group at both 2 h (Figure [158]3H) and 24 h (Figure [159]3I),
   indicating impaired memory deficits.

Figure 3.

   Figure 3
   [160]Open in a new tab

   Down‐regulation of SHANK1 leads to synaptic damage and cognitive
   deficits A)Experimental design sketch. Healthy SD rats were randomly
   divided into 2 groups. In the control group, rats were injected with
   AAV/vector virus in the hippocampal CA1 region. In the model group,
   rats were injected with AAV‐Sh‐SHANK1 virus in the hippocampal CA1
   region. Following the treatment, behavioral, electrophysiological, and
   biochemical tests were performed after 4 weeks. B) Brain tissues
   (hippocampus CA1 region) were homogenized, and SHANK1 protein levels
   were detected by immunoblotting. Actin was used as a loading control.
   Quantitative analysis of the SHANK1, (n = 3), (C). D–G) The Morris
   water maze test: the swimming speed (D) and the latency to find the
   hidden platform (E) from day 1 to day 6, the spatial memory was tested
   on the 7th day by removing the platform, and the time spent in the
   target quadrant (F) and the number of crossing the position of the
   target platform (G) were measured, (n = 12–15). (H, I) Fear
   conditioning test showed freezing duration was measured during the
   3‐min memory test at 2 h H) or 24 h I) after conditioning, (n  =
   12–15). J, K) The firing rate of action potential was reduced, (n =
   18–24 neurons from 3 rats). L, M) Hippocampal CA3‐CA1 LTP was recorded
   by the MED64 system. Normalized CA3‐CA1 fEPSP mean slope recorded from
   the CA1 dendritic region in acute hippocampal slices. Quantitative
   analyses for fEPSP measured 60–90 min after HFS relative to baseline (n
   = 3 slices from 3 rats). Rat hippocampal neurons were subjected to
   whole‐cell patch‐clamp recording of mEPSCs N), amplitude and frequency
   of mEPSCs of control and HS neurons (O), (n = 10–14 neurons from 3
   rats). P) The representative images of Golgi staining, and dendritic
   complexity were evaluated by sholl analysis,(n = 10 neurons from 3
   rats), (Q). R) the density of dendritic spines (n = 30 neurons from 3
   rats) and mushroom‐type spines (n = 30 neurons from 3 rats), (S) were
   tested. T–X) The representative images showing transmission electron
   microscope (Scale bar, 0.5µm) (T), Quantitative analysis of the
   synaptic density (n = 9 from 3 rats per group) (U), the length of
   postsynaptic density (V) and width of postsynaptic denses (W) and the
   release of presynaptic vesicles (X) were measured (n  = 40–45 synapses
   from 3 rats). Data were expressed as mean ± SD, ^* p < 0.05, ^** p <
   0.01, ^*** p < 0.001, ^**** p < 0.0001 vs control. Statistical details
   were provided in [161]Supporting Information.

   To further investigate the role of SHANK1 deficiency in the induction
   of learning and memory impairment. SHANK1 knockdown significantly
   diminished AP firing rates (Figure [162]3J,K), which suggests SHANK1
   downregulation reduced neuronal activity. Next, we evaluated the effect
   of SHANK1 knockdown on synaptic function. Electrophysiological
   recordings showed that the Sh‐SHANK1 rats exhibited a significantly
   reduced slope of fEPSP following HFS, compared to control rats
   (Figure [163]3L,M). Whole‐cell patch‐clamp recordings in the CA1 region
   revealed a substantial reduction in both the amplitude and frequency of
   mEPSC in Sh‐SHANK1 rats (Figure [164]3N,O). Golgi staining further
   demonstrated reduced dendritic complexity in Sh‐SHANK1 neurons, as
   evidenced by Sholl analysis (Figure [165]3P, Q), along with decreased
   spine density and a reduction in the mature mushroom‐type spines
   (Figure [166]3R, S). Transmission electron microscopy confirmed a
   significant reduction in the number of excitatory synapses per unit
   area in the CA1 region (Figure [167]3T,U), as well as a decrease in
   both the length (Figure [168]3V) and width (Figure [169]3W) of the
   postsynaptic density. However, no significant differences were observed
   in the number of synaptic vesicles between the two groups
   (Figure [170]3X). Additionally, primary hippocampal neurons were
   transfected with AAV‐GFP‐vector or AAV‐GFP‐Sh‐SHANK1 (Figure [171]S6A,
   Supporting Information). AAV/GFP labeling revealed a reduction in
   dendritic complexity and total dendritic length in Sh‐SHANK1 neurons
   compared to controls (Figure [172]S6A–C, Supporting Information),
   suggesting that SHANK1 deficiency results in morphological
   abnormalities of neurons. In summary, these findings strongly
   demonstrate that SHANK1 downregulation leads to synaptic damage and
   cognitive impairments, further implicating SHANK1 as a critical player
   in synaptic and cognitive function.

2.4. High‐Salt Diet Downregulates the PKA/CREB Signaling Pathway, Leading to
SHANK1 Reduction

   Our previous research demonstrated that HS diet induced
   hippocampal‐dependent spatial memory impairments, accompanied by
   decreased CREB phosphorylation at Ser133, and inhibited CREB
   activity.^[ [173]^11 ^] RNA sequencing analysis suggested that the HS
   diet suppressed the cAMP pathway (Figure [174]2B). It is
   well‐established that members of the CREB family are pivotal regulators
   of gene expression in response to cAMP, following phosphorylation by
   cAMP‐dependent protein kinase (PKA).^[ [175]^41 ^] However, it is not
   clear whether and how PKA/CREB pathway abnormalities and SHANK1
   deficiency co‐contribute to cognitive dysfunction caused by HS diet.
   Therefore, we first investigated the cAMP/PKA signaling and found
   significantly lower cAMP levels in the hippocampus of HS rats compared
   to controls (Figure [176]4A), alongside reduced PKA kinase activity
   (Figure [177]4B). Western blotting also confirmed a significant
   reduction in phosphorylated CREB‐Ser133 (p‐CREB) in the HS group
   (Figure [178]4C, D). Since cAMP elevation promotes PKA activation and
   induces CREB phosphorylation into the nucleus, which regulates gene
   transcription and is critical for memory function, CREB and SHANK1 mRNA
   levels are significantly reduced in the hippocampus of HS rats, we
   hypothesize that HS‐induced cognitive impairment and reduced SHANK1 are
   mediated by reduced CREB transcriptional activity in the cAMP/PKA
   pathway. Using JASPAR software, CREB is predicted to be a potential
   transcription factor for SHANK1 (Figure [179]S7, Supporting
   Information). Dual‐luciferase reporter assays confirmed a positive
   regulatory relationship between CREB and SHANK1 (Figure [180]4E). To
   further verify this, chromatin immunoprecipitation (ChIP) assays were
   performed, revealing reduced CREB binding to sites 4 (596–797 bp) and 8
   (1356–1528 bp) within the 2000‐bp upstream non‐coding region of SHANK1
   in HS rats, compared to controls (Figure [181]4F). These results
   suggest that HS diet inhibits CREB activity, leading to reduced SHANK1
   transcription.

Figure 4.

   Figure 4
   [182]Open in a new tab

   A) High‐salt diet downregulated the PKA/CREB signaling pathway, leading
   to SHANK1 reduction The level of cAMP was detected by Elisa Kit (n =
   6). B) The PKA kinase activity was detected by Elisa Kit (n = 6). C)
   Brain tissues (hippocampus CA1 region) were homogenized, and pCREB/CREB
   protein levels were detected by immunoblotting. Actin was used as a
   loading control. Quantitative analysis of the pCREB/CREB, (n = 3), (D).
   E) The dual luciferase gene report assay was tested to the relationship
   between the CREB and SHANK1, (n = 3). F) Chromosome immunoprecipitation
   assay was showed that the relationship of CREB and SHANK1, (n = 3). G)
   Primary hippocampal neurons were treated with PKA inhibitor s1582 at
   10, 25, and 50 µm. The PKA activity was measured, (n = 5–6). Western
   blotting H) was performed by using antibodies against SHANK1, CREB, and
   p‐CREB. Quantitative analysis of the protein levels (I, J) (n = 3). K)
   mRNA levels of SHANK1 were measured, (n = 3). (L, M) After the PKA
   inhibitor was administered to hippocampal primary neurons,
   cycloheximide (CHX) was added for 3 h, 6 h, and 9 h later, the
   degradation of SHANK1 was detected by western blotting (L), and
   quantitative analysis of the degradation level, (n = 3), (M). N)
   Immunofluorescence showed the representative images from hippocampal
   primary neurons treated with s1582 10, 25, and 50 µm for 24 h. Sholl
   analysis of dendrite complexity(O), (n = 10 neurons), quantification
   for the dendritic length (P) and the SHANK1 puncta/10µm dendrites (Q)
   (n = 15 neurons). By injection s1582 into hippocampus CA1, Fear
   condition was used to measure the contextual memory: freezing time
   during the 3 min memory test done at 2 h (R) or 24 h (S), (n = 7–8). T)
   The levels of CREB, p‐CREB, SHANK1 proteins were detected by western
   blotting, and quantitative analysis of the protein levels, (n = 3).
   Data were expressed as mean ± SD, ^* p < 0.05, ^** p < 0.01, ^*** p <
   0.001, ^**** p < 0.0001vs Control or s1582. Statistical details were
   provided in Supporting information.

   To investigate the regulation of SHANK1 by the PKA/CREB pathway in
   vitro, primary hippocampal neurons were treated with s1582, a PKA
   inhibitor. The CCK‐8 assay was used to screen the effect of s1582 on
   cells and establish the concentration gradient. The results are
   presented in Figure [183]S8 (Supporting Information). A significant
   decrease in cell viability was observed in the group treated with 100
   µm s1582 at 24 h. Therefore, we chose 10, 25, and 50 µm concentrations
   to treat primary hippocampal neurons for 24 h. s1582 significantly
   decreased PKA kinase activity in dose‐dependent manner (Figure [184]4G)
   and reduced CREB Ser133 phosphorylation and SHANK1 protein levels
   (Figure [185]4H–J), as well as SHANK1 mRNA expression (Figure [186]4K).
   This suggests that inhibition of PKA downregulates the CREB/SHANK1
   pathway. Additionally, to assess whether PKA inhibition affected SHANK1
   protein degradation, primary hippocampal neurons were treated with
   cycloheximide (CHX) following s1582 treatment. Western blotting showed
   that s1582 did not influence SHANK1 degradation (Figure [187]4L, M).
   s1582 treatment significantly decreased dendritic complexity
   (Figure [188]4N, O), total dendritic length (Figure [189]4P), and
   SHANK1 puncta in hippocampal neurons (Figure [190]4Q). Additionally, we
   bilaterally injected the s1582 into the hippocampal CA1 region in
   2‐month‐old male SD rats, and behavioral testing commenced 48 h
   post‐injection. When s1582 was injected into the hippocampal CA1 region
   in SD rats at concentrations of 10, 25, and 50 µm (2µL), PKA kinase
   activity was significantly reduced (Figure [191]S9A, Supporting
   Information). Behavioral assays revealed that the open field test
   showed no significant differences in total distance covered (Figure
   [192]S9B, Supporting Information) and the reduced number of zone
   crossings at s1582 50 µm (Figure [193]S9C, Supporting Information). The
   novel object recognition test showed a significant reduction in the
   recognition index in the s1582‐treated group at 24 h (Figure [194]S9E,
   Supporting Information), without change at 2 h (Figure [195]S9D,
   Supporting Information). The fear conditioning test indicated no change
   in freezing behavior at 2 h post‐conditioning (Figure [196]4R), but a
   significant reduction was observed at 24 h (Figure [197]4S). Moreover,
   s1582 injected into the hippocampal CA1 region in SD rats led to a
   significant dose‐dependent decrease in SHANK1 protein expression and
   phosphorylated CREB (Figure [198]4T). Collectively, these findings
   suggest that inhibition of PKA/CREB pathway leads to SHANK1 reduction,
   which in turn may contribute to synaptic and cognitive dysfunction.

2.5. PKA Activation Promotes SHANK1 Expression and Rescues HS‐Related
Cognitive Impairments

   The above findings indicate that the reduction of SHANK1 is involved in
   HS diet‐induced synaptic dysfunction and cognitive deficits through PKA
   inactivation, suggesting that PKA activation may potentially mitigate
   HS‐related cognitive impairments. To test this hypothesis, 8‐week‐old
   male SD rats were fed with HS diet for 9 weeks, followed by bilateral
   hippocampal CA1 injections of s7858, a PKA agonist, at concentrations
   of 10, 20, and 30 µm (2µL). Behavioral, electrophysiological, and
   biochemical assessments were subsequently performed (Figure [199]5A).
   s7858 administration significantly elevated PKA activity
   (Figure [200]5B), as well as SHANK1 mRNA and protein levels
   (Figure [201]5C–E), and phosphorylation of CREB at Ser133
   (Figure [202]5D, F) in the hippocampus of HS rats. These findings
   further support that PKA/CREB activation upregulates SHANK1
   transcription. Behavioral tests revealed no significant differences
   between groups in the open field test (Figure [203]5G). However, in the
   novelty recognition test, the HS + s7858 group exhibited a
   significantly higher recognition index for novel objects compared to
   the HS group at 24 h (Figure [204]5H). Moreover, fear conditioning
   tests showed that freezing behavior in the HS + s7858 group was
   significantly increased at both 2 h (Figure [205]5I) and 24 h
   (Figure [206]5J), compared to the HS group. Electrophysiological
   studies further confirmed that PKA activation enhanced the slope of
   fEPSP following high‐frequency stimulation relative to the HS group
   (Figure [207]5K, L). Golgi staining analysis also showed S7858 can
   alleviate HS‐induced dendritic damage, as assessed by Sholl analysis,
   in the HS + s7858 group (Figure [208]5M, N). Additionally, PKA
   activation elevated dendritic spine density and mushroom‐type spines
   (Figure [209]5O–Q). These findings strongly support the hypothesis that
   PKA activation promotes SHANK1 expression and rescues cognitive
   deficits and synaptic dysfunction caused by the HS diet.

Figure 5.

   Figure 5
   [210]Open in a new tab

   PKA activation promotes SHANK1 expression and rescues HS‐related
   cognitive impairments A) Experimental design sketch. After 9 weeks HS
   diet, s7858(PKA agonist) 2µL (10, 20, 30 µm) were injected into
   hippocampal CA1. B) The PKA kinase activity was detected by Elisa Kit,
   (n = 4). C) mRNA levels of SHANK1 were measured by qPCR,(n = 3). D–F)
   Brain tissues (hippocampus CA1 region) were homogenized, SHANK1 and
   pCREB/CREB levels were detected by immunoblotting. Actin was used as a
   loading control. Quantitative analysis of the SHANK1 (E) and pCREB/CREB
   (F), (n = 3). G) Open field test showed spontaneous exercise(n = 7–10).
   H) The novel object recognition test showed the recognition index at 24
   h,(n = 7–10). I,J) Fear condition was performed to assess: Freezing
   time during the 3‐min memory test at 2 h (I) or 24 h,(n = 7–10), (J).
   K, L) Hippocampal CA3‐CA1 LTP was recorded by the MED64 system.
   Normalized CA3‐CA1 fEPSP means slope recorded from the CA1 dendritic
   region in acute hippocampal slices. (L) Quantitative analyses for fEPSP
   measured 60–90 min after HFS relative to baseline (n = 3 slices from 3
   rats). M) Golgi staining was performed to show spines density and
   maturation in the hippocampal CA1 region. N) The representative images
   of dendritic trees in the CA1 of rats in three groups, (bar = 100 µm).
   Sholl analysis (N) and dendritic complexity analysis were performed to
   evaluate the dendritic complexity (n = 10 neurons from 3 rats). O)
   Representative images of dendritic spines. P) The quantitative analysis
   for the spine density (per 10 µm) and Q) percentage of mushroom‐type
   spines (n = 25 neurons from 3 rats). Data were expressed as mean ± SD,
   ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001 vs HS
   group. Statistical details were provided in [211]Supporting
   Information.

2.6. CREB is Essential for PKA Regulation of SHANK1 Expression and Neuronal
Dendritic Growth

   To further elucidate the role of CREB in the transcriptional regulation
   of SHANK1 and its impact on synaptic function, we employed two CREB
   mutations: a phosphorylation mimic (S133D, which simulates an activated
   state) and a non‐phosphorylation mimic (S133A, which simulates an
   inactivated state) at the Ser133 site (Figure [212]6A). These mutated
   CREB constructs were transfected into primary hippocampal neurons. qPCR
   analysis revealed that CREB‐S133D significantly upregulated SHANK1
   expression, whereas the S133A mutation suppressed SHANK1 mRNA levels
   (Figure [213]6B). Western blotting further confirmed these findings,
   showing increased SHANK1 protein levels in the CREB‐S133D group, while
   SHANK1 expression was reduced in the CREB‐S133A group (Figure [214]6C,
   D).

Figure 6.

   Figure 6
   [215]Open in a new tab

   CREB is essential for PKA regulation of SHANK1 expression and neuronal
   dendritic growth A) The two mutated viruses (AAV‐CREB‐S133D/S133A) were
   transfected in primary hippocampal neurons. The SHANK1 mRNA was
   measured by qPCR, (n = 3) (B). C) Western blotting was employed to
   detect the SHANK1 protein level, actin was used as a loading control.
   Quantitative analysis of SHANK1 (D), (n = 4). E) Representative
   immunofluorescence images showed the total length of dendrites in
   neurons. Scale bar = 20 µm), Sholl analysis (F), quantitative analyses
   of dendritic length G) and the SHANK1 puncta/10µm dendrites H) (n = 15
   neurons). I) Primary hippocampal neurons were treated with PKA
   inhibitor/s1582 and AAV‐CREB‐S133D, representative immunofluorescence
   images were shown (Scale bar = 20 µm), Sholl analysis J), quantitative
   analyses of dendritic length K) and the SHANK1 puncta/10µm dendrites
   (L), (n = 15‐16 neurons). M) The SHANK1 mRNA was measured by qPCR, (n =
   5). N)Western blotting was employed to detect the SHANK1 protein
   levels, actin was used as a loading control. Quantitative analysis of
   SHANK1 O), (n = 4). All data represent mean ± SD, ^* p < 0.05, ^** p <
   0.01, ^*** p < 0.001, ^**** p < 0.0001 vs Control or s1582 group.
   Statistical details were provided in [216]Supporting Information.

   To examine the impact of these mutations on dendritic morphology,
   primary hippocampal neurons were transfected with
   AAV‐GFP‐CREB‐S133D/S133A (Figure [217]6E), and GFP labeling was used to
   visualize dendritic structure. CREB‐S133D significantly enhanced
   dendritic complexity (Figure [218]6F), total dendritic length
   (Figure [219]6G), and SHANK1 puncta (Figure [220]6H), while CREB‐S133A
   reduced dendritic complexity and SHANK1 expression (Figure [221]6F‐H).
   In line with this, primary hippocampal neurons were treated with PKA
   inhibitor/s1582 and AAV‐CREB‐S133D. Immunofluorescence results also
   showed that the dendritic complexity, total dendritic length, and
   SHANK1 puncta were significantly increased in the s1582 + S133D group
   compared to the s1582 group (Figure [222]6I‐L). qPCR (Figure [223]6M)
   and western blotting (Figure [224]6N, O) results showed a significant
   increase in SHANK1 levels in the s1582 +S133D group compared to the
   s1582 group alone. In addition, s7858 fails to increase SHANK1 levels
   in neurons transfected with AAV/CREB S133A (Figure [225]S10A, B,
   Supporting Information). These data further suggest that CREB is
   required for PKA regulation of SHANK1 expression and neuronal dendritic
   growth. In addition, Rats’ primary hippocampal neurons were exposed to
   various concentrations (0.92%,0.95%) of NaCl for 0.5 h at 37°C to
   simulate a high‐salt environment in vitro (Figure [226]S11A, Supporting
   Information), in the 0.92% and 0.95% NaCl simulated high salt models,
   neuronal dendrite loss(Figure [227]S11B, Supporting Information), CREB
   site 133 dephosphorylation and reduced SHANK1 levels were observed
   (Figure [228]S11C, D, Supporting Information). However, transfection of
   the mutant virus CREB‐S133D under 0.92% NaCl conditions alleviated
   neuronal damage(Figure [229]S11E,F, Supporting Information) and
   increased SHANK1 levels(Figure [230]S11G,H, Supporting Information).
   So, these results suggest that CREB‐S133D could rescue the HS‐induced
   SHANK1 deficiency and neuronal damage.

3. Discussion

   Globally, dementia affects 6 to 50 million people, with recent
   publications reporting higher estimates, likely due to increased
   prevalence or improved diagnosis.^[ [231]^42 ^] Neurocognitive
   disorders affect tens of millions worldwide, imposing a significant
   societal burden.^[ [232]^43 , [233]^44 ^] However, there are no
   effective clinical drugs or interventions due to the complex etiology
   of cognitive dysfunction. Long‐term HS diet is a risk factor for
   cognitive deficits, but its pathogenesis remains unclear. In our study,
   we identified that high salt intake in rats leads to the abnormal
   downregulation of SHANK1 protein. Our findings show that PKA
   inactivation in HS rats leads to CREB dephosphorylation, contributing
   to SHANK1 reduction. CREB dephosphorylation by a PKA inhibitor led to
   decreased SHANK1 levels, dendritic damage, and subsequent cognitive
   impairments. Notably, PKA activation restored CREB phosphorylation and
   SHANK1 expression, significantly reversing synaptic and cognitive
   impairments induced by long‐term HS diet. These results suggest that
   SHANK1 plays a key role in the pathological and behavioral changes
   associated with HS diet and indicate SHANK1 as a potential therapeutic
   target for HS‐related cognitive deficits.

   To identify key driving factors of HS‐mediated cognitive dysfunction,
   we identified SHANK1 as a critical molecule. SHANK is a scaffold
   protein located in excitatory synapses and is essential for the normal
   development and function of synapses.^[ [234]^21 , [235]^45 ^] The
   SHANK protein family, encoded by the SHANK1, SHANK2, and SHANK3 genes,
   undergoes alternative promoter usage and splicing, generating multiple
   mRNA transcripts and protein isoforms.^[ [236]^20 ^] Although SHANK
   gene mutations have been extensively linked to autism spectrum
   disorders, the physiological role of SHANK proteins in other contexts,
   such as cognitive decline, remains underexplored. In the current study,
   SHANK1 not SHANK2 and SHANK3, was significantly decreased in long‐term
   HS rats. Our findings enhance the understanding of SHANK1 in synaptic
   function and provide new insights into how its dysregulation
   contributes to cognitive impairments associated with long‐term
   high‐salt diet. This suggests SHANK1's pivotal role in the synaptic and
   cognitive abnormalities observed in HS model.

   Recent findings by Facaro et al. reported that tau phosphorylation in
   the hippocampus was not evident until 12 weeks. Our studies
   demonstrated significant cognitive deficits following 9 weeks of
   chronic HS intake, yet these were not associated with tau
   phosphorylation, Aβ deposition, or neuronal loss in the hippocampus.
   Anti‐amyloid therapies have shown modest disease‐modifying benefits.
   Recent anti‐amyloid monoclonal antibodies have shown modest
   disease‐modifying effects in early Alzheimer's disease, reducing
   amyloid‐β plaques and slowing cognitive decline in clinical trials.^[
   [237]^46 , [238]^47 , [239]^48 ^]

   Given that dementia, including Alzheimer's disease, is characterized by
   synaptic dysfunction, formation of senile plaques, hyperphosphorylated
   tau tangles, neuroinflammation, and apoptotic cell death. This suggests
   that synaptic dysfunction is preexisting in Aβ aggregation and tau
   pathology and may play a key role in HS diet‐induced cognitive
   impairments. Our study in 8% HSD‐fed SD rats reveals early dendritic
   damage occurring before AD‐like pathology, suggesting that this model
   captures a preclinical window of vulnerability. Importantly, we propose
   that timely intervention targeting the PKA‐CREB‐SHANK1 axis during this
   early phase could prevent subsequent AD pathogenesis. While our current
   data demonstrate HS‐induced synaptic damage and cognitive impairment
   without tau hyperphosphorylation or Aβ accumulation.

   Our findings indicate that HS diet reduces SHANK1 via downregulation of
   the PKA/CREB pathway, leading to synaptic dysfunction and cognitive
   deficits in rats, we believe that reducing dietary intake of salt for
   an extended period of time improves dendritic complexity resulting in a
   return of cognitive improvements. In addition, we advocate for a
   low‐salt diet to reduce the risk of developing dementia. However, the
   PKA/CREB/SHANK1 pathway serves as a potential therapeutic target for
   cognitive dysfunction in individuals with a high‐salt diet.

4. Limitation

   In the present study, a limitation of our study was that a long‐term HS
   diet reduced cAMP levels in the hippocampus, which we do not fully
   understand the mechanism of this part. Recent studies have found
   whether GPER1 activation improves hypertension and kidney damage in
   female Dahl salt‐sensitive rats fed a high salt diet.^[ [240]^49 ^]
   High salt conditions may indirectly modulate cAMP levels in neurons via
   several GPCR‐dependent mechanisms. First, HS could alter the
   conformational stability or ligand‐binding affinity of certain GPCRs,
   especially those sensitive to ionic strength, potentially disrupting
   their coupling to Gαs/Gαi proteins and subsequent cAMP production.^[
   [241]^50 , [242]^51 ^] Second, HS might interfere with the heparan
   sulfate proteoglycan‐dependent clustering of GPCRs on synaptic
   membranes, impairing localized cAMP signaling microdomains.^[ [243]^52
   ^] Third, the elevated Na^+ concentrations under HS could perturb
   Na^+‐sensitive adenylate cyclase and phosphodiesterase, indirectly
   dysregulating cAMP turnover.^[ [244]^53 ^] Notably, SHANK1
   downregulation, observed in our HS model, might exacerbate these
   effects by destabilizing GPCR‐scaffolding complexes, further uncoupling
   receptors from downstream cAMP effectors.^[ [245]^20 , [246]^54 ^]
   However, how SHANK1 deficiency disrupts synaptic structure is another
   limitation. Lack of SHANK1 impairs actin‐cytoskeleton interactions (via
   βPIX), compromises trans‐synaptic adhesion (through the neuroligin‐GKAP
   complex), and displaces key glutamate receptors along with their
   downstream signaling effectors.^[ [247]^55 , [248]^56 ^] These
   disruptions collectively contribute to the observed dendritic
   simplification and spine loss, while also potentially exacerbating
   HS‐induced synaptic dysfunction through impaired GPCR scaffolding.^[
   [249]^20 ^] In the current study, we adopted an 8% high‐salt diet,
   following the methodology established by Faraco et al,^[ [250]^10 ^] to
   reliably induce neurovascular and cognitive dysfunction within a
   controlled experimental timeframe. However, a moderate level of sodium
   intake over a longer duration would better represent the slow,
   progressive nature of cognitive decline more typically observed in
   dementia patients. Therefore, a 2%–4% high‐salt dietary intervention
   merits additional study to better approximate population‐level sodium
   exposure.

5. Conclusion

   Our study identifies a novel mechanistic link between long‐term HS diet
   and cognitive impairment, wherein PKA/CREB axis inactivation leads to
   SHANK1 reduction, synaptic damage, and cognitive deficits (Figure
   [251]7 ). Given the detrimental effects of PKA/CREB/SHANK1 signaling
   inhibition on synaptic and cognitive functions in the context of HS
   diet, targeting the upregulation of this axis may offer a promising
   therapeutic strategy for addressing HS‐induced cognitive dysfunction.

Figure 7.

   Figure 7
   [252]Open in a new tab

   Schematic illustration of the PKA/CREB/SHANK1 axis and its role in
   synaptic damage and cognitive impairments in high‐salt diet rats. The
   HS diet downregulates PKA activity, triggering a reduction in CREB
   phosphorylation. This leads to decreased translocation of the CREB
   complex into the nucleus and diminished binding to the transcription
   factor of SHANK1, ultimately resulting in synaptic damage and cognitive
   impairments.

6. Experimental Section

Reagents

   Antibodies and Chemicals used in this study were provided in
   [253]Supporting Information.

Animals

   Male Sprague‐Dawley (SD) rats were purchased from the Tongji Animal
   Center at Huazhong University of Science and Technology and allowed to
   acclimatize in the experimental animal facility for one week. All
   animal experiments received approval from the Institutional Animal Care
   and Use Committee (IACUC) of Huazhong University of Science and
   Technology (IACUC NO.4486).

   In our study, 2‐month‐old male SD rats were randomly divided into two
   groups: one group was fed a normal diet (0.4% NaCl) as a control, while
   the other group received HS diet (8% NaCl) for 9 weeks. Behavioral
   tests and electrophysiological and merigenetic tests were used to
   detect the relevant indicators after 9 weeks (Figure [254]1A). To
   further investigate the role of SHANK1, bilateral intra‐hippocampal CA1
   injections of AAV‐Sh‐SHANK1 virus were administered to explore its
   impact on cognitive function. Behavioral test and electrophysiological
   and merigenetic tests were used to detect the relevant indicators after
   4 weeks (Figure [255]3A). Additionally, either ddH[2]O or the PKA
   inhibitor (s1582) was bilaterally injected into the hippocampal CA1
   region. in 2‐month‐old male SD rats, behavioral testing commenced 48 h
   post‐injection, sequentially including open field test (day 1), novel
   object recognition (days 2–3), and fear conditioning (days 4–5). Rats
   were euthanized following the final behavioral test for subsequent
   molecular analyses.

   For further evaluation of PKA's role in HS‐related cognitive deficits,
   rats were divided into five groups: a control group on a normal diet
   (0.4% NaCl), a second group on a HS diet (8% NaCl) for 9 weeks, and
   other experimental groups on HS diet (8% NaCl) for 9 weeks with
   bilateral intra‐hippocampal CA1 injections of either ddH[2]O or PKA
   agonist (2 µL at concentrations of 10, 20, and 30 µm). Behavioral test
   and electrophysiological and merigenetic tests were used to detect the
   relevant indicators after 48 h (Figure [256]5A).

Feed Preparation

   The feed was prepared by Beijing HFK BIOSCIENCE CO., LTD. Feed Main
   Ingredients: Corn, soybean meal, wheat bran, wheat middlings, fish
   meal, soybean oil, stone powder, calcium hydrogen phosphate, salt,
   choline chloride, lysine, multiple vitamins, and multiple mineral
   elements.

   Nutritional Composition: Crude protein 19.2%, crude fat 4.6%, crude
   fiber 4.0%, crude ash 6.3%, moisture 8.8%, calcium 1.19%, phosphorus
   0.87%, lysine 11.1 g kg^−1, methionine 4.5 g kg^−1, cystine 6.4
   g kg^−1, nitrogen‐free extract 55.9%. The caloric value was 3.42
   kCal g^−1, with protein accounting for 22.47%, fat for 12.11%, and
   carbohydrates for 65.42%.

   High‐Salt Feed: The salt concentration was increased to 8%, compared to
   the normal dietary salt concentration of 0.4%. All rats drank tap water
   normally.

Stereotactic Surgery

   The experimental rats were acclimatized to the operating room one day
   before surgery. Anesthesia was induced using isoflurane, and the rats
   were positioned under a stereotactic apparatus. After disinfection with
   iodophor and 75% ethanol, a midline scalp incision was made between the
   ears. Bilateral stereotaxic drilling was performed at coordinates 4.2
   mm posterior, 3 mm lateral, and 2.8 mm ventral to the bregma. AAV9‐CMV
   bGlobin‐MCS‐EGFP‐3FLAG‐WPRE‐hGH or Sh‐SHANK1 (2 µL, 2.0 × 10^12 viral
   genomes per milliliter) was injected into the hippocampal CA1 region at
   a rate of 0.125 µL/min using a stereotactic apparatus. Similarly, the
   PKA inhibitor (s1582) or agonist (s7858) was injected into the
   hippocampal CA1 region. Following the injection, the needle was
   retained in place for 10 min before being slowly withdrawn. The
   incision was sutured, and the rats were kept near a heater until they
   regained consciousness before being returned to the animal room.

Transmission Electron Microscope

   Rats were anesthetized with isoflurane, and tissue from the hippocampal
   CA1 region was rapidly dissected under a light microscope. Tissues were
   cut into approximately 10 mm^2 pieces with a thickness of less than 5
   mm. Prior to fixation, samples were rinsed with normal saline or PBS to
   remove blood and mucus. Tissues from each group were fixed in 2.5%
   glutaraldehyde at room temperature for 1 h, followed by fixation with
   1% OsO[4] in 0.1M PBS at room temperature for 2 h, then dehydrated and
   embedded. Images were captured using a light microscope, and semi‐thin
   slices of 2 µm were prepared and stained with 1% toluidine blue in 1%
   sodium borate. The CA1 region was identified, and selected semi‐thin
   sections were further cut into ultra‐thin sections using a Leica
   microtome, and then stained with uranyl acetate and lead citrate.
   Ultra‐thin sections were examined using a Hitachi HT7800 transmission
   electron microscope. The synaptic density, presynaptic vesicles, and
   the length and width of postsynaptic densities were measured.

Primary Hippocampal Neuron Culture

   Primary hippocampal neurons were isolated from Sprague‐Dawley rat
   embryos at gestational days 17–18. Following isoflurane anesthesia, the
   brain was quickly excised and the hippocampus was dissected. The blood
   vessels on the hippocampal surface were carefully removed using
   ophthalmic scissors, and the tissue was minced in Hank's buffered
   saline solution before being suspended in 0.25% (v/v) trypsin solution
   for 9 min at 37°C. Neurons were plated in 6‐well and 12‐well plates
   coated with 100 µg mL^−1 poly‐D‐lysine and incubated at 37°C with 5%
   CO[2] for 4 h. The medium was then replaced with neurobasal medium
   supplemented with 2% (v/v) B‐27 and 1 × GlutaMAX. Half of the
   neurobasal medium was refreshed every three days. Primary hippocampal
   neurons were cultured for one week before treatment. Depending on the
   experimental design, neurons were treated with
   AAV‐GFP‐Vector/AAV‐Sh‐SHANK1/AAV‐CREB‐S133A/AAV‐CREB‐S133D viruses, PKA
   inhibitor, PKA agonist, and cycloheximide. Neurons were subsequently
   collected and lysed in RIPA buffer, or RNA was extracted using Trizol
   for further bioassays, or they were fixed with 4% paraformaldehyde for
   immunofluorescence imaging. All cell culture reagents were obtained
   from Thermo Fisher Technologies. Dendritic complexity was assessed
   using Sholl analysis, and dendritic length was analyzed using a
   semi‐automated protocol with Imaris software.

Western Blotting

   Hippocampal tissue was removed from the −80°C freezer and placed on
   ice. The hippocampal CA1 region or primary hippocampal neurons were
   isolated and homogenized in a lysis buffer for western blotting. The
   protein concentration was determined using a bicinchoninic acid (BCA)
   protein assay kit. Samples were boiled in a water bath for 10 min to
   promote protein denaturation, then separated on a 10% SDS‐PAGE gel. The
   running time was adjusted based on the molecular weight of the target
   protein, and the gel was transferred to a nitrocellulose (NC) membrane.
   The membrane was incubated in 5% skim milk for 0.5 to 1 h, followed by
   incubation with the primary antibody at 4°C overnight. The NC membrane
   was then incubated with IRDye TM (800 CW) anti‐mouse/anti‐rabbit IgG or
   HRP‐conjugated goat anti‐mouse/anti‐rabbit IgG at 25°C for 1 h, and
   images were captured using the Odyssey infrared imaging system (LI‐COR
   Biosciences, USA) or ChemiScope (Clinx Science Instruments Co. Ltd.).
   Image analysis software was utilized to quantify visual bands.

Cycloheximide Test

   Primary hippocampal neurons were cultured for 7 days. For cycloheximide
   (CHX) treatment, CHX was diluted in the cell culture medium to a final
   concentration of 0.1 µm and administered for 0, 3, 6, or 9 h. Following
   treatment, cells were collected and lysed in RIPA buffer, and protein
   levels were measured via western blotting.

Cell Viability Analysis

   The survival rate of primary hippocampal neurons was assessed following
   the manufacturer's instructions using the Cell Counting Kit‐8 (CCK‐8).
   The cells were seeded in 96‐well plates and treated with different
   concentrations of the PKA inhibitor s1582 (50, and 100 µm) for 6, 12,
   and 24 h. A total of 10 µL of CCK‐8 solution was added to 90 µL of
   medium. After incubation for 2 h at 37°C and 5% CO[2], absorbance was
   measured at 450 nm using a microplate reader (BioTek, China).

Real‐Time Quantitative PCR

   Total RNA was extracted using Trizol reagent (Invitrogen, Carlsbad, CA,
   USA) according to the manufacturer's instructions and
   reverse‐transcribed into cDNA using a reverse transcription kit
   (Takara, Dalian, China). A total of 100 ng of cDNA was utilized for
   real‐time PCR. The reaction mixture was composed as follows: 2 × SYBR
   Green Mix (5 µL), cDNA (100 ng, 1 µL), forward primer (1µL), reverse
   primer (1µL), and ddH[2]O (2 µL). The primers for the target genes were
   designed as follows:
     * SHANK1:

   Sense: CCTGCGTTCCAAGTCTATGA

   Antisense: TTGTTGAGTGCCATCTGATA
     * SHANK2:

   Sense: CAGCCGCTATTGCCTACC

   Antisense: TCTTGTCGTCAGCCCTCC
     * SHANK3:

   Sense: ACAGGACTACCCGCCTAACC

   Antisense: CCGCCATTGCGAAGAAC
     * GAPDH:

   Sense: TGCCTTCTCTTGTGACAAAGTGG

   Antisense: CATTGCTGACAATCTTGAGGGAG

   The PCR cycling conditions were set as follows: 95°C for 30 s, followed
   by 40 cycles of 95°C for 5 s, 60°C for 30 s, and 72°C for 30 s. Melting
   curve analysis was performed after each experiment. Amplification and
   analysis were conducted using the StepOnePlus real‐time PCR detection
   system (Life Technologies, NY, USA).

Chromatin Immunoprecipitation Quantitative PCR (ChIP‐qPCR)

   ChIP was performed using the Simple ChIP® Plus Sonication Chromatin IP
   Kit (CST, #56383). Fresh or frozen tissue samples (100 mg) were weighed
   on ice, cut into 1–2 mm pieces, and 1 mL of pre‐cooled PBS with
   protease inhibitor cocktail (PIC) was added to each sample.
   Subsequently, 28 µL of 37% formaldehyde solution was added for
   crosslinking at room temperature for 20 min. Crosslinking was
   terminated by adding 100 µL of 10X glycine solution, gently mixing, and
   incubating on ice for 5 min. Nuclear lysis buffer (1 mL) was prepared
   for chromatin processing and fragmentation. Samples were centrifuged at
   21 000 g for 10 min at 4°C, and the supernatant containing cross‐linked
   chromatin was transferred to a new tube. Chromatin immunoprecipitation
   (IP) was conducted, and chromatin was eluted from the antibody/protein
   G microbeads before decrosslinking. To enhance DNA purity, it was
   purified using a spin column, and ChIP results were quantitatively
   analyzed by PCR. For SHANK1, an upstream non‐coding sequence of 2000 bp
   was designed using Gene Software, and divided into 11 segments for
   RT‐qPCR analysis. Further details can be found in [257]Supporting
   Information.

Dual Luciferase Gene Reporter Assay

   HEK293 cells were retrieved from liquid nitrogen, quickly thawed in a
   37°C water bath, and centrifuged at 1300 rpm for 3 min. The cell
   suspension was inoculated into a 6 cm dish containing 2 mL of complete
   medium, gently shaken, and placed in a 37°C, 5% CO[2] incubator. Cells
   in the logarithmic growth phase were prepared as suspensions and
   inoculated into 24‐well plates (approximately 10^5 cells per well,
   depending on cell morphology), cultured until reaching approximately
   60% confluence. Plasmid (1 µg) and Lipo2000 (2 µL) were dissolved in
   100 µL of Opti‐MEM per well, mixed, and incubated at room temperature
   for 20 min before being added to the cells. Luciferase activity was
   measured 48 h later using a luciferase kit (E1910). For detailed
   protocols, refer to the Dual‐Luciferase Reporter Assay System
   Description: Promega Technical Manual. The luciferase plasmid used was
   SHANK1 (Rat 78957, NM031751 promoter). Vector information included:
   MCS‐SV40‐firefly Luciferase‐PolyA‐Tk‐Renilla Luciferase‐PolyA.

cAMP Assay

   cAMP levels were measured using an ELISA kit according to the
   manufacturer's instructions. Freshly collected hippocampal CA1 tissues
   were immediately frozen and thawed, then tissues were homogenized on
   ice with a homogenizer, followed by centrifugation at room temperature
   at 12 000 rpm for 5 min. Experimental procedures included adding 50 µL
   of Neutralizing Buffer to each well, followed by 50 µL of 0.1 M HCl to
   the NSB well (Non‐Specific Binding), and adding 50 µL of standard
   solution at each concentration to corresponding wells. Sample solution
   (50 µL) was added to corresponding wells, followed by the Assay Buffer
   working solution and cAMP‐HRP Conjugate solution (50 µL each).
   Anti‐cAMP monoclonal antibody (50 µL) was then added to each well.
   Following a 10‐minute incubation at room temperature with TMB Substrate
   (150 µL), the reaction was stopped with Stop Solution (50 µL), and the
   optical density was measured at 450 nm using a plate reader to
   calculate cAMP levels.

PKA Kinase Activity Assay

   PKA activity was measured using the PKA kinase activity kit, a
   non‐radioactive assay designed for safe, rapid, and reliable
   measurements. The positive control (active PKA) was prepared, and
   samples were diluted in Kinase Dilution Buffer in wells. ATP was then
   added to initiate the kinase reaction, which was incubated for 90 min
   at 30°C. Phospho‐specific antibody was added to each well and incubated
   for 60 min at room temperature. After adding the HRP‐conjugated
   secondary antibody and incubating for another 30 min at room
   temperature, the wells were washed. TMB solution was added and
   incubated for 40 min at room temperature, followed by Stop Solution.
   Optical density was measured at 450 nm, and PKA activity was
   calculated.

Fluorescence Imaging and Confocal Microscopy

   Brain slices or primary hippocampal neurons were fixed in 4%
   paraformaldehyde and permeabilized at room temperature in PBS
   containing 0.5% Triton X‐100 for 30 min. Samples were incubated for 60
   min in a 5% BSA solution, followed by overnight incubation at 4°C with
   target primary antibodies. Afterward, samples were incubated at 37°C
   for 1 h with Invitrogen fluorescent secondary antibodies (Alexa Fluor
   488 or 594), and nuclei were stained with DAPI (0.1 µg/mL).
   Fluorescence images were captured using a Zeiss LSM 710 laser‐scanning
   confocal microscope (Zeiss, Jena, Germany) equipped with Zen software.
   Sholl analysis was conducted to measure dendritic complexity, and
   dendritic length was assessed using Imaris software (Media Cybernetics,
   California, USA).

Golgi Staining

   The Golgi solution was prepared as follows: Solution A: 5% potassium
   dichromate (10 g dissolved in 200 mL ddH₂O); Solution B: 5% mercury
   chloride (10 g in 200 mL ddH₂O); Solution C: 5% potassium chromate (8 g
   dissolved in 160 mL ddH₂O). Solutions A and B were mixed to form
   solution AB; this was then mixed with 400 mL ddH₂O containing solution
   C. The final volume ratio of potassium dichromate: mercury chloride:
   potassium chromate: water was 5:5:4:10. The solution was stored in the
   dark for approximately one week. Sampling involved anesthetizing rats
   with isoflurane, injecting approximately 300 mL of normal saline into
   the left ventricle to clear the brain of invading vesicles, and
   immersing brain tissues in dark AC solution for 30 days, changing the
   Golgi solution every 2 days. Brain tissue was sectioned using a
   vibrating microtome (Leica, Wetzlar, Germany) at a thickness of 100 µm
   and dried on slides at room temperature. Staining involved alkalizing
   samples with ammonia, dehydrating, and soaking in CXA solution for 15
   min (1:1:1 mixture of 1000 mL chloroform, 1000 mL xylene, and 1000 mL
   anhydrous ethanol), then sealing with neutral gum and allowing to dry.
   Image‐Pro Plus software was used for image analysis.^[ [258]^57 ^]

Acute Slice Preparation and LTP Recording

   Rats were decapitated under isoflurane anesthesia, and brains were
   rapidly excised and placed in ice‐cold, oxygenated artificial
   cerebrospinal fluid (aCSF) composed of: 119 mm NaCl, 2.5 mm KCl, 26.2
   mm NaHCO₃, 1 mm NaH₂PO₄, 11 mm glucose, 1.3 mm MgSO₄, and 2.5 mm CaCl₂
   (pH 7.4). Coronal brain slices of 350 µm thickness were prepared using
   a vibratome (Leica, VT1000S, Germany) and subsequently transferred to a
   recovery chamber filled with oxygenated aCSF for a minimum of 1.5 h at
   room temperature prior to electrophysiological recordings. For
   long‐term potentiation recordings, acute brain slices were submerged in
   aCSF within a recording chamber. The slices were positioned on an 8 × 8
   microelectrode array (Parker Technology, Beijing, China), featuring 50
   × 50 µm electrodes with a 150 µm interelectrode distance. Electrical
   signals were recorded using the MED64 system (Alpha MED Sciences,
   Panasonic), targeting field excitatory postsynaptic potentials (fEPSP)
   in CA1 neurons by stimulating CA3 neurons. LTP induction was
   accomplished via three trains of high‐frequency stimulation (HFS; 100
   Hz, 1 s duration). The magnitude of LTP was quantified as the
   percentage change in the fEPSP slope, measured between 60–90 min
   post‐induction.^[ [259]^57 , [260]^58 , [261]^59 ^]

Recording of mEPSC

   Neurons in the CA1 hippocampal slice were visualized using a
   DIC‐infrared upright microscope and recorded using whole‐cell patch
   clamp procedures at ‐70 mV. The pipette resistance was in the range of
   4–6MΩ. The intracellular solution contained (in mm): 140 K‐gluconate, 2
   MgCl[2], 10 HEPES, 10 BAPTA, 2 Mg‐ATP, 0.5 CaCl[2]‐H[2]O, 0.5 Li‐GTP,
   pH 7.2‐7.4, 280–290 mOsm. 5mm QX‐314 was added to block voltage‐gated
   Na+ channels and GABAB receptors. The mEPSCs were recorded in the
   presence of 1 µm tetrodotoxin (TTX), and 100 µm picrotoxin (PTX).

Recording of mIPSCs

   The pipette electrodes, with a resistance of 3–5MΩ, were filled with
   the internal solution containing (in mm): 140 CsCl, 10 HEPES, 2
   MgCl[2], 0.5 EGTA, 2 MgATP, 0.5 Na[3]GTP, 12 phosphocreatine and 30
   NMG, pH 7.2‐7.4, 280–290 mOsm. 5mm QX‐314 was added to block
   voltage‐gated Na+ channels and GABABRs. After the neuron was
   voltage‐clamped at −70 mV, mIPSCs were recorded in the presence of 5 µm
   TTX and 50 µM kynurenic acid (KYN).

Recording of Paired Pulse Ratio

   Paired pulse ratios were calculated as a ratio of EPSC2 to EPSC1
   separated by inter‐stimulus intervals of 50, 100, 150, and 200 ms for
   EPSCs. To measure the input‐output curves for pyramidal neurons, a
   bipolar tungsten electrode was placed in the stratum radiatum with a
   bipolar tungsten electrode ≈50 um from the recording electrode in order
   to test potential presynaptic effects.

Recording of AP

   Recordings were obtained using borosilicate glass micropipettes (3–8
   MΩ). The internal pipette solution consisted of the following (in mm):
   145 K‐gluconate, 10 HEPES, 1.0 EGTA, 2.0 Na[2]ATP, and 0.4 NaGTP (300
   mOsmol/kgH[2]O, pH 7.2). Neurons were slightly depolarized with current
   injection (500 ms, current clamp) to generate a regular spiking
   activity (range, ‐50 to 230 pA).

Behavior Test–Open Field Test

   The open field test was employed to assess anxiety and exploratory
   behaviors. The apparatus consisted of a 100 × 100 cm^2 PVC square arena
   with 70 cm high walls. Individual rats were placed in the arena for a 5
   min session. Anxiety levels were quantified by the number of entries
   into the center of the arena, while exploratory activity was measured
   by tracking the total distance covered by the rats. Data collection
   utilized tracking files recorded with LabState version 1.0 software
   (AniLab Software and Instruments Co., LTD., China).

Elevated Plus Maze

   This test was designed to evaluate the conflict between exploration of
   a new environment and fear of the elevated open arms, reflecting
   anxiety in rats. The elevated plus maze featured two open arms (50 cm
   long × 10 cm wide), two closed arms (50 cm long × 10 cm wide × 40 cm
   high), and a central platform (10 cm × 10 cm), elevated 70 cm above the
   ground. Rats were acclimated in the testing room for 24 h prior to the
   experiment. During the test, each rat was positioned in the maze's
   center and observed for 5 min, The tines and duration of entries into
   the open and closed arms were recorded. The open arm duration ratio =
   Duration in open arm / (Duration in open arm + Duration in closed arm)
   and the open arm entry ratio where open arm entry ratio = (open arm
   entries) / (open arm entries + closed arm entries).

Social Interaction Test

   The experimental setup consisted of three boxes, each measuring 19 cm ×
   45 cm, separated by transparent plexiglass with a central channel for
   connectivity. A uniform metal cage, sized appropriately for a rat, was
   placed centrally in the left and right boxes. Prior to testing, rats
   were acclimated in the behavioral testing room for 30 min. A novel rat
   was positioned in the metal cage on one side, while the opposite cage
   remained empty. Initially, the three boxes were separated by a
   transparent glass resin plate, and the test rat was placed in the third
   box for 5 min. Following this, the glass resin plate was removed,
   permitting the test rat to explore the three boxes freely for an
   additional 5 min. The time spent interacting with familiar and
   unfamiliar rats was meticulously recorded using AniLab Software, which
   tracked and quantified interaction durations based on predefined
   proximity criteria. The track plots were generated by the software to
   visualize these interactions.

Novel Objective Recognition Test

   Rats were placed in the novel object recognition room 24 h prior to the
   test and subsequently transferred to a 100 cm × 100 cm × 100 cm plastic
   container for a 5 min familiarization period with objects A and B.
   After each familiarization phase, the arena and objects were cleaned
   with 70% ethanol. 2 h following the initial familiarization, object B
   was replaced with object C, and rats were allowed 5 min to explore both
   objects. After 24 h, object C was swapped for object D, and rats were
   again given 5 min to explore. The duration of exploration for objects C
   and D was recorded during both the 2 and 24 h intervals. The
   recognition index (RI) was calculated as: RI (2h) = time at object C
   [(time at object C) / (time at objects A and C)], RI(24h) = time at
   object D [(time at object D)/(time at objects A and D)].

Morris Water Maze Test

   The Morris water maze was utilized to assess spatial learning and
   memory. Rats were acclimated to the water maze room for 24 h, where the
   walls were adorned with distinct patterns to facilitate environmental
   orientation. During the training phase, rats were tasked with locating
   a hidden platform over five to six consecutive days, with four trials
   conducted each day at intervals of 40 min between 8:00 AM and 2:00 PM.
   Each trial commenced in one of four quadrants, with the rat facing the
   pool wall, concluding when the rat successfully climbed onto the
   platform. If a rat fails to locate the platform within 1 minute, the
   experimenter would manually guide it to the platform, where it remained
   for 20 s. A camera recorded the swim paths and time taken to find the
   platform. Spatial memory was evaluated following the training phase; on
   day 6 or 7, the platform was removed, and the duration spent in the
   target quadrant and the number of platform crossings was recorded.

Fear Condition

   Fear conditioning was employed to evaluate the effect of contextual
   training on memory consolidation. On the initial day, rats were
   acclimated to the conditioning chamber for a duration of 3 min, after
   which an auditory stimulus (70 dB, 30 s) was presented, followed by a
   foot shock (0.5 mA, 2 s). This sequence was replicated twice, with a
   2‐minute interval between each trial. Two h post‐conditioning, the rats
   were reintroduced to the chamber for 3 min in the absence of any
   stimuli, during which freezing behavior was recorded as a measure of
   memory retention. The following day, rats were placed back in the
   chamber for an additional 3 min without any stimuli to evaluate
   contextual recall by quantifying the duration of freezing behavior.

RNA Sequencing

   Tissue samples were obtained from the rat hippocampal CA1 region.
   Following appropriate pretreatment, samples were sent to Novogene
   Sequencing Company (Tianjin, China) for sequencing. RNA was extracted
   from tissues using standard extraction methods, followed by stringent
   quality control to ensure an accurate assessment of RNA integrity.
   After library construction, initial quantification was performed using
   a Qubit 2.0 Fluorometer, and the library was diluted to 1.5 ng µL^−1.
   The effective concentration of the library was accurately quantified
   using qRT‐PCR to ensure quality. Upon qualification of the library
   test, Illumina sequencing was conducted following the pooling of
   different libraries.

Data Sets Collection and Processing

   In this study, RNA‐seq was conducted to identify marker genes
   associated with HS diet rats in the hippocampus. The datasets
   [262]GSE36980^[ [263]^60 ^] and [264]GSE84422^[ [265]^61 ^] were
   obtained from the Gene Expression Omnibus (GEO) database, and
   corresponding mRNA expression matrices were aligned for comprehensive
   analysis. In the [266]GSE36980 dataset, analysis was performed on
   postmortem human brain tissues donated for the Hisayama study to
   identify molecular pathological changes in AD brains. The control group
   consisted of hippocampal samples [267]GSM907861‐[268]GSM907870, which
   served as references for bioinformatics analysis. The experimental