Cellular Cholesterol Loss Impairs Synaptic Vesicle Mobility via the CAMK2/Synapsin-1 Signaling Pathway

Abstract

Background:

Neuronal cholesterol deficiency may contribute to the synaptopathy observed in Alzheimer’s disease (AD). However, the underlying mechanisms remain poorly understood. Intact synaptic vesicle (SV) mobility is crucial for normal synaptic function, whereas disrupted SV mobility can trigger the synaptopathy associated with AD. In this study, we investigated whether cellular cholesterol deficiency affects SV mobility, with the aim of identifying the mechanism that links cellular cholesterol loss to synaptopathy in AD.

Methods:

Lentiviruses carrying 3β-hydroxysteroid-Δ24 reductase-complementary DNA (DHCR24-cDNA), DHCR24-short hairpin RNA (DHCR24- shRNA) or empty lentiviral vectors were transfected into SHSY-5Y cells in order to construct DHCR24 knock-down and knock-in models, along with corresponding controls. Filipin III cholesterol staining was employed to visualize membrane and intracellular cholesterol in the different cell models, and fluorescence intensity was assessed using confocal microscopy. Additionally, we performed immunoblotting to quantify the expression of DHCR24, total calmodulin-dependent protein kinase 2 (CAMK-2), p-CAMK2 (T286), caveolin-1, total synapsin-1, phosphorylated synapsin-1 (p-synapsin-1; S605), and synaptophysin in each experimental group.

Results:

In DHCR24-silenced cells, the loss of cellular cholesterol caused by knock-down of DCHR24 resulted in a significant decrease in the levels of phosphorylated CAMK2 (p-CAMK2) and phosphorylated synapsin-1 (p-synapsin-1) compared to control cells. The reduction in p-CAMK2 and p-synapsin-1 could disrupt SV mobility, thereby reducing replenishment of the readily releasable pool (RRP) from the reserve pool (RP). Furthermore, cells with DHCR24 knock-down showed downregulation of caveolin-1, a crucial lipid raft marker, compared to control cells. Conversely, elevated cellular cholesterol levels caused by knock-in of DHCR24 reversed the effects of cholesterol deficiency, suggesting that CAMK2-mediated synapsin-1 phosphorylation may be regulated in a lipid raft-associated manner. Additionally, we found that cellular cholesterol loss could significantly downregulate the expression of synaptophysin protein, which is vital for SV biogenesis and synaptic plasticity.

Conclusion:

These results suggest that depletion of cellular cholesterol following knock-down of DHCR24 can decrease synaptophysin protein expression and impair SV mobility by regulating the CAMK2-meditated synapsin-1 phosphorylation pathway, potentially via a lipid raft-associated mechanism. Our study indicates a critical role for cellular cholesterol deficiency in AD-related synaptopathy, thus highlighting the potential for targeting cellular cholesterol metabolism in therapeutic strategies.

Graphical abstract

Keywords

Alzheimer’s disease / DHCR24 / cholesterol / CAMK2 / synapsin-1 / synaptic vesicle / synaptopathy

1. Introduction

Synaptic dysfunction is a characteristic feature of several neurodegenerative diseases, including Alzheimer’s disease (AD). This condition is closely linked to cognition and memory [1]. Mounting evidence suggests that defective memory and cognition in AD are not only a direct consequence of synaptic dysfunction, but also an indirect result of other major pathological changes, such as amyloid beta accumulation, phosphorylated tau, neuroinflammation, and programmed cell death. These can also initiate or compromise memory and cognition by inducing synaptic disorder [2, 3, 4].

Cholesterol is critical in the proper structural and functional development of the brain through its influence on neurogenesis, synaptic plasticity, neurotransmitter release, and signal transduction [5]. Since the 1990s, numerous epidemiological, genetic and biological studies have implicated altered cholesterol metabolism in the pathogenesis of AD [6, 7, 8, 9]. Mutations or polymorphisms in genes can reduce cholesterol transport to the brain, thus impairing neuronal cholesterol uptake and trafficking. Such genes include apolipoprotein E4 (ApoE4), ATP binding cassette transporter, the low-density lipoprotein receptor family, and Niemann-Pick C disease 1 or 2, resulting in cellular cholesterol deficiency in the brain tissue of patients and animal models with inherited AD [9, 10, 11]. Various studies involving inherited AD animal models, particularly those with the ApoE4 genotype, strongly suggest that loss of cellular cholesterol in the brain contributes to diverse AD-related pathological processes [9, 10, 11].

Methyl-\beta-cyclodextrin (M\betaCD) is used to study the effects of cholesterol deficiency on various pathological changes in previous research [12, 13]. However, M\betaCD can also remove other lipids, leading to non-specific effects on cellular function [12, 13]. Moreover, M\betaCD causes acute cholesterol depletion [12], which may not accurately reflect the chronic cholesterol deficiency seen in AD and other neurodegenerative diseases [14]. In contrast, the modulation of cholesterol synthetase, such as 3\beta-hydroxysteroid-\mathrm{\Delta }24 reductase (DHCR24), can model chronic cholesterol deficiency and allow specific control over the synthesis and level of cellular cholesterol [15]. DHCR24 is a critical enzyme in the brain that catalyzes cholesterol synthesis from desmosterol. Moreover, evidence from in vitro cell models shows that a reduction in cellular cholesterol induced by knock-down of DHCR24 can lead to AD-related pathological impairments, including increases in A\beta generation and tau hyperphosphorylation [9, 16, 17]. Collectively, these findings indicate that cellular cholesterol deficiency in the brain may exacerbate AD-related pathology. Furthermore, DHCR24 could be a potential target because of its crucial role in cellular cholesterol metabolism and in the molecular mechanism that underlies AD pathogenesis.

The precise role of cholesterol metabolism in AD-associated synaptopathy is still unclear. Synaptic cholesterol deficiency could diminish neurotransmitter release by reducing calcium (Ca²⁺) influx and Ca²⁺ channel activity [18, 19]. Notably, the Ca²⁺ channel is localized in the lipid raft [20], and Ca²⁺ influx can directly activate calcium/calmodulin-dependent protein kinase 2 (CAMK2). This serine/threonine protein kinase is crucial for synaptic vesicle (SV) mobility [21]. Cholesterol is an important component of membrane lipid rafts, which are essential for cell membrane signal transduction. Deficiency in cellular cholesterol disrupts lipid raft formation and is strongly implicated in AD pathology [22, 23]. An excellent model used to investigate the biochemical mechanisms underlying SV mobility is the human neuroblastoma cell line SH-SY5Y. This expresses major SV markers and contains various types of vesicles, as well as the capability for calcium-dependent synaptic exocytosis [24, 25, 26, 27]. In the present study, SH-SY5Y cells were used to construct cell models with varying cholesterol levels by modulating the DHCR24 gene. We hypothesized that decreased cellular cholesterol may impair SV mobility in a lipid raft-associated manner, potentially contributing to AD synaptopathy.

2. Materials and Methods

2.1 Antibodies and Reagents

Rabbit anti-GAPDH (1:3000, 2118, Cell Signaling Technology, Danvers, MA, USA), anti-DHCR24/Seladin-1 (1:1000, 2033, Cell Signaling Technology, Danvers, MA, USA), anti-Phospho-Synapsin-1 (Ser605) (1:1000, 88246S, Cell Signaling Technology, Danvers, MA, USA), anti-Synapsin-1 (1:1000, 5297S, Cell Signaling Technology, Danvers, MA, USA), anti-CaMKII (pan) (1:1000, 4436S, Cell Signaling Technology, Danvers, MA, USA), anti-Phospho-CaMKII (Thr286) (1:1000, 12716T, Cell Signaling Technology, Danvers, MA, USA), anti-Caveolin-1 (1:1000, 3267T, Cell Signaling Technology, Danvers, MA, USA), anti-Synaptophysin (1:1000, 36406S, Cell Signaling Technology, Danvers, MA, USA), anti-rabbit IgG HRP-linked antibody (1:5000, 7074P2, Cell Signaling Technology, Danvers, MA, USA), anti-mouse IgG HRP-linked antibody (1:5000, 7076P2, Cell Signaling Technology, Danvers, MA, USA), Filipin III (0.1 mg/mL, abs42018484, Absin, Shanghai, China), propidium iodide (PI) (3.5 µg/mL, abs9105, Absin, Shanghai, China).

2.2 Cell Culture and Transfection

SH-SY5Y cells were purchased from Chinese Academy of Sciences (Shanghai, China) and grown in DMEM (Gibco, New York, NY, USA) containing 10% FBS (Biological Industries, Be’er Sheva, Israel) at 37 °C in a 5% CO₂ atmosphere and in 10 cm culture dishes (Chinese Academy of Sciences, Shanghai, China). The cells were then seeded into 6-well plates and grown to approximately 30–40% confluence. Lentivirus packed with DHCR24 cDNA, DHCR24 shRNA, or empty vector (Jima, Shanghai, China) were transfected into SH-SY5Y cells (Multiplicity of Infection (MOI) = 4). After 24 h, the medium was removed and fresh virus-free medium was added. Afterward, cells were treated with puromycin (8 µg/mL) for 48 h to kill uninfected cells and screen out successfully transfected cells. Infected cells that expressed green fluorescent protein (GFP) were observed under the fluorescent microscope. Cells were constantly cultured in puromycin-containing medium at the appropriate concentration for several passages. Short tandem repeat (STR) profiling was used to validate all cell lines, which also tested negative for mycoplasma. We confirmed that SH-SY5Y cells were well transfected with lentivirus and ready for subsequent experiments.

2.3 Western Blot Analysis

SDS lysis buffer (KeyGen Biotech, Nanjing, China) supplemented with protease and phosphatase inhibitors (KeyGen Biotech, China) was used to lyse cells. Protein samples were then boiled at 100 °C with sample buffer and fractioned by SDS-PAGE. Subsequently, proteins were transferred onto PVDF membranes, blocked with 5% skimmed milk, and incubated overnight at 4 °C with the corresponding primary antibody. Secondary antibody contained the Horseradish Peroxidase (HRP) label was then used to detect protein expression by chemiluminescence (Bio-Rad, Hercules, CA, USA).

2.4 Filipin III Staining

Cells were cultured in 24-plate cover glass dishes to reach confluence of 60–70% and then washed thrice with PBS after removal of the medium. The cells were subsequently incubated with Filipin III solution for 30 minutes at room temperature and then stained with PI for 10 minutes. Filipin III, PI and GFP were detected by confocal fluorescence microscopy (Leica, Wetzlar, Germany).

2.5 Statistical Analysis

Immunoblotting results and fluorescence intensity were analyzed by ImageJ2 software (National Institutes of Health (NIH), Bethesda, MD, USA). The p-value for comparison of two groups was calculated using the t-test and GraphPad Prism 9 (GraphPad Software, Inc., San Diego, CA, USA). Statistical data were presented as the mean \pm SD of three or more independent experiments. Statistically significant differences between groups were indicated by * (p 0.05), ** (p 0.01), *** (p 0.001) and **** (p 0.0001). No statistical difference was indicated as ns (p > 0.05).

3. Results

3.1 Cell Models were Successfully Transfected by DHCR24 Knock-Down/Knock-In

Our previous experiments demonstrated that cholesterol levels in neuronal cell lines could be modulated by DHCR24 knock-in or knock-down [28, 29]. In the present experiments, we performed DHCR24 knock-down to create a neuronal cell model of cellular cholesterol deficiency, and DHCR24 knock-in to create a neuronal cell model of elevated cellular cholesterol. Fluorescence microscopy demonstrated that lentivirus-meditated DHCR24-shRNA and DHCR24-cDNA were successfully transfected into the SH-SY5Y neuroblastoma cell line, as visualized by green staining with microscopy (Fig. 1A). Compared with the control group, DHCR24 protein expression was significantly decreased in DHCR24 silenced cells (Fig. 1B,C). In contrast, the expression of DHCR24 protein was markedly increased in cells overexpressing DHCR24 compared to the controls (Fig. 1D,E). Thus, we confirmed that our cell models were successfully transfected by DHCR24 knock-down or knock-in.

3.2 Cellular Cholesterol Level Analyzed by Filipin III Staining

Next, we performed Filipin III staining to evaluate the cholesterol level in transfected cell lines. After staining, the plasma membrane and intracellular cholesterol levels were examined by confocal laser scanning microscopy. In DHCR24-silenced cells, the fluorescence intensity was weaker than in control cells, indicating lower plasma membrane and intracellular cholesterol levels (Fig. 2A,B). In contrast, Filipin III staining showed markedly increased levels of plasma membrane and intracellular cholesterol in cells overexpressing DHCR24 compared to the controls (Fig. 2A,C). These findings were consistent with observations from previous studies [17, 23, 28]. Based on our earlier work and the present results, the above data demonstrate that DHCR24 knock-down or knock-in can be used to construct neuronal cell models of cholesterol deficiency, or elevated cholesterol, respectively.

3.3 Reduced Cellular Cholesterol Leads to Decreased Expression of the Lipid-Raft Proteins Caveolin-1 and Phospho-CAMK2

Cholesterol is a key lipid molecule within cell membranes, and the cellular cholesterol level positively affects expression of the cholesterol-binding protein caveolin-1 (Cav-1) [17, 18, 28]. Both cholesterol and Cav-1 are critical structural components of the caveolae/lipid-rafts involved in many cell signaling pathways [17, 28, 29, 30]. Thus, we further analyzed the effect of cellular cholesterol depletion on the expression of Cav-1 protein. Cav-1 expression was found to be markedly lower in SH-SY5Y cells following DHCR24 knock-down compared to control cells, indicating disruption of membrane caveolae/lipid-rafts (Fig. 3A,B). Conversely, Western blot analysis revealed that DHCR24 overexpression significantly upregulated the expression of Cav-1 in the caveolae fraction of SH-SY5Y cells with DHCR24 knock-in, suggesting an increase in membrane caveolae (Fig. 3C,D). Consistent with earlier reports [17, 22, 29], we also found that knock-down of DHCR24 simultaneously decreased the levels of plasma membrane cholesterol and Cav-1 in caveolae/lipid-rafts. Therefore, cellular cholesterol depletion following knock-down of DHCR24 can impair the structure and function of caveolae/lipid-rafts in the plasma membrane by downregulating Cav-1 expression.

Caveolae are cholesterol-rich lipid-raft microdomains in the plasma membrane that regulate the function of protein kinases such as calmodulin-dependent protein kinase II (CAMK2), which contribute to SV recycling [31, 32]. Here, we further evaluated the phosphorylation state of CAMK2 at the T286 site in SH-SY5Y cells. The level of phosphorylated CAMK2 (p-CAMK2) was found to be significantly lower in cells with DHCR24 knock-down compared to control cells, leading to reduced activity of CAMK2 kinase (Fig. 4A,C). In contrast, elevated cellular cholesterol in cells with DHCR24 overexpression increased the level of phosphorylated CAMK2 protein compared to control cells, suggesting increased CAMK2 kinase activity (Fig. 4G,I). However, the total CAMK2 protein level was similar between the different groups and the control (Fig. 4A,B,G,H). In summary, the above findings demonstrate that depletion of cellular cholesterol by DHCR24 knock-down can impair the lipid-raft organization of caveolae, thus inhibiting activation of CAMK2 signaling in a lipid raft-associated manner.

3.4 Depletion of Cellular Cholesterol Reduces the Level of p-Synapsin-1 Protein via CAMK2 Signaling

Synapsin-1 is an SV-associated phosphoprotein that modulates the mobility of SVs [33, 34]. Synapsin-1 is an important downstream substrate of CAMK2 signaling and its phosphorylation is positively regulated by Ca²⁺/CAMK2 [34]. We therefore quantified total synapsin-1 and phospho-synapsin-1 (S605) by Western blot analysis. The level of total synapsin-1 was similar between the different groups (Fig. 4D,E,J,K), whereas phospho-synapsin-1 (S605) was significantly lower in cells with DHCR24 knock-down compared to the controls (Fig. 4D,F). Conversely, phospho-synapsin-1 (S605) was significantly increased in cells that over-expressed DHCR24 compared with the control group (Fig. 4J,L). Interestingly, CAMK2 can phosphorylate two sites in the carboxy-terminal region of synapsin-1, causing it to dissociate from SVs [33, 34], while dephosphorylated synapsin-1 can reverse this process. Based on results from earlier reports and from the present study, we postulate that cellular cholesterol deficiency caused by DHCR24 knock-down can markedly decrease synapsin-1 phosphorylation via the activation of CAMK2.

3.5 Loss of Cellular Cholesterol Decreases the Expression of Synaptophysin

Previous reports suggest that synaptophysin is a major protein in SVs and that it contributes to synaptic plasticity and SV biogenesis [35, 36]. We therefore investigated whether the depletion of cellular cholesterol affects the expression of SV synaptophysin. We found that depletion of cellular cholesterol resulted in decreased expression of synaptophysin protein in DHCR24 knock-down cells compared to controls (Fig. 5A,B). However, no significant difference in synaptophysin protein expression was found between cells with DHCR24 overexpression and control cells (Fig. 5C,D). Hence, the above findings indicate that depletion of cellular cholesterol following DHCR24 knock-down downregulates the expression of SV synaptophysin.

4. Discussion

Although there is increasing evidence that cholesterol loss contributes to synaptic dysfunction in the brain of AD patients, the underlying molecular mechanisms are not yet fully understood [37, 38]. Interestingly, our research team and other scholars have demonstrated that cellular cholesterol deficiency caused by knock-down of DHCR24 results in the disruption of membrane lipid raft-dependent signals [17, 28, 29, 39]. The use of Filipin III staining in the current study again revealed that knock-down of DHCR24 induced a significant decrease in the cellular cholesterol level. Moreover, free cholesterol is an important component of lipid rafts and promotes their formation and polarization. Membrane signaling molecules are concentrated within lipid rafts, thereby regulating cell signaling activity. Additionally, a critical structural protein within caveolae/lipid rafts, Cav-1, is a key molecular organizer of various cellular functions, including endocytosis/exocytosis and signal transduction. Previous studies have shown that reduction of cellular cholesterol via knock-down of DHCR24 markedly decreases Cav-1 protein expression and disrupts lipid raft integrity [17, 22, 28, 29]. The results of our study confirm previous reports that cellular cholesterol deficiency can lead to disruption of membrane lipid rafts.

Our experiments showed that reduction in cellular cholesterol significantly decreased the phosphorylation of CAMK2 at the T286 site in SH-SY5Y cells with DHCR24 knock-down, suggesting inhibition of the CAMK2 signaling pathway. In contrast, the level of phosphorylated CAMK2 (p-CAMK2) was markedly increased in SH-SY5Y cells that overexpressed DHCR24, indicating activation of the CAMK2 signaling pathway. Notably, cholesterol-enriched lipid raft/caveolae are Ca²⁺-signaling microdomains that regulate Ca²⁺ influx and Ca²⁺ channel activity [31, 32]. A previous study had shown that decreased cellular cholesterol contributed to lower neurotransmitter release by decreasing Ca²⁺influx and Ca²⁺ channel activity at synapses [19]. Interestingly, CAMK2 is a crucial Ca²⁺-dependent kinase, and Ca²⁺ influx leads to CAMK2 activation by inducing autophosphorylation at the T286 site [40]. Therefore, by integrating our findings with previous data, we propose that loss of cellular cholesterol could inhibit CAMK2 phosphorylation activity, potentially mediated by caveolae/lipid raft-dependent calcium signaling.

Furthermore, accumulating evidence indicates a significant relationship between phosphorylation of CAMK2 at the T286 site and AD. Specifically, reduced phosphorylation of CAMK2 at this site was observed in AD patients, whereas increased phosphorylation was associated with enhanced memory formation [41, 42, 43, 44, 45]. Our results showed that loss of cellular cholesterol can reduce CAMK2 activation by inhibiting the phosphorylation of CAMK2 at the T286 site.

Furthermore, synapsin-1 is critical in the SV life cycle and serves as a direct substrate for calcium/CAMK2 [46, 47]. To clarify the downstream effects of decreased phosphorylation of CAMK2 at the T286 site, we also measured the level of synapsin-1 phosphorylation. Our findings showed that cellular cholesterol deficiency could significantly inhibit CAMK2 phosphorylation (T286) activity, and simultaneously reduce the synapsin-1 phosphorylation (S605) level (Fig. 4). Conversely, the increase in cellular cholesterol caused by knock-in of DHCR24 promoted CAMK2 phosphorylation (T286) and increased synapsin-1 phosphorylation (S605) (Fig. 4). Collectively, these findings suggest that decreased activity of synapsin-1 phosphorylation may be induced by inhibition of CAMK2 signaling. Therefore, our data revealed firstly that cellular cholesterol deficiency caused by the knock-down of DHCR24 can lead to decreased phosphorylation of synapsin-1, potentially via lipid-raft-dependent CAMK2 signaling.

Importantly, synapsins are pivotal regulators of SV dynamics within presynaptic terminals and are thought to participate in regulating the SV life cycle [48, 49]. In particular, synapsin-1 is phosphorylated by calcium/calmodulin, resulting in SV mobilization from the reserve pool (RP) during neurotransmission. Specifically, synapsin-1 induces trafficking of SVs from the RP to the readily reversible pool (RRP) after stimulation [50, 51, 52, 53, 54, 55]. Increased synapsin-1 phosphorylation (S605 site) by CAMK2 signaling could reduce SV clustering, diminish interaction with actin filaments, and enhance SV movement towards the active zone, thereby increasing RRP formation [50, 53, 56, 57, 58]. Conversely, reduced synapsin-1 phosphorylation at the S605 site strengthens the SV binding and synapsin-1-actin interaction, thereby increasing construction of the SV-synapsin-1-actin complex [33, 57, 58, 59]. In addition, the decrease in phosphorylated synapsin-1 (S605 site) could lead to reduced SV movement towards the active zone of synapses, resulting in decreased RRP formation [59, 60, 61]. Thus, our results strongly indicate that cellular cholesterol deficiency reduces replenishment of the RRP from the RP by decreasing synapsin-1 phosphorylation via CAMK2 signaling, potentially in a lipid raft-associated manner. Taken together, our data suggest a novel mechanism for the effect of cellular cholesterol deficiency on SV mobility.

Of note, our study also found that decreased cellular cholesterol caused by knock-down of DHCR24 significantly reduced synaptophysin protein expression, with no significant increase observed in cells overexpressing DHCR24 (Fig. 5A–D). Although the specific function of synaptophysin is not yet fully understood, accumulating evidence highlights the important role played by synaptophysin during SV biogenesis, exocytosis, and synaptic plasticity [35, 36, 62, 63]. We hypothesize that disruption in the formation of lipid raft microdomains caused by loss of cellular cholesterol leads to altered lipid raft-dependent signal transduction, which then downregulates synaptophysin expression. Furthermore, previous studies have indicated that synaptophysin can bind to cholesterol, with this interaction being crucial for SV biogenesis [36, 62, 64]. Given the role of synaptophysin as a cholesterol-binding protein, cellular cholesterol deficiency may impair SV biogenesis by disrupting the cholesterol-synaptophysin interaction [65]. Consequently, our results indicate that decreased cellular cholesterol induced by knock-down of DHCR24 could also impair SV biogenesis and synaptic plasticity by downregulating synaptophysin expression and interfering with the cholesterol-synaptophysin interaction. Although this study investigated the impact of cholesterol deficiency on SV mobility and proposed a possible molecular mechanism, it was primarily focused on the expression level of important proteins in synaptic biosynthesis and SV mobility. More in-depth mechanistic studies are therefore required. The validity of the findings could be enhanced by future research employing live-cell imaging and neurophysiological techniques in primary cells, and animal models with DHCR24 knockdown-induced cholesterol deficiency. This approach would facilitate further observation and investigation of the relationship between cholesterol deficiency and SV transport in live cells, as well as the release of neurotransmitters. Additionally, it would allow for a more detailed exploration of the mechanisms underlying these molecular events and their roles in the onset of AD.

5. Conclusion

In conclusion, a model of cellular cholesterol loss involving silencing of DHCR24 was used to demonstrate that cellular cholesterol deficiency could inhibit phosphorylation of Calcium/CAMK2. This results in the inhibition of synapsin-1 phosphorylation, disruption of SV mobility, and a subsequent decrease in the formation of RRP from RP, potentially in a lipid raft-associated manner (Fig. 6). The present study also demonstrated that cellular cholesterol deficiency downregulated the expression of synaptophysin protein, which can decrease SV biogenesis and synaptic plasticity. Importantly, our study underscores the importance of cellular cholesterol deficiency/loss in synaptopathy during AD, and provides valuable insights into the complex interplay between loss of cellular cholesterol and AD pathogenesis.

Availability of Data and Materials

All data supporting the conclusions of this article are available from the corresponding author upon reasonable request.

References

[1] Leng F, Edison P. Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here? Nature Reviews. Neurology. 2021; 17: 157–172. https://doi.org/10.1038/s41582-020-00435-y.

[2] John A, Reddy PH. Synaptic basis of Alzheimer’s disease: Focus on synaptic amyloid beta, P-tau and mitochondria. Ageing Research Reviews. 2021; 65: 101208. https://doi.org/10.1016/j.arr.2020.101208.

[3] Sanders OD. Virus-Like Cytosolic and Cell-Free Oxidatively Damaged Nucleic Acids Likely Drive Inflammation, Synapse Degeneration, and Neuron Death in Alzheimer’s Disease. Journal of Alzheimer’s Disease Reports. 2023; 7: 1–19. https://doi.org/10.3233/ADR-220047.

[4] Guan PP, Ge TQ, Wang P. As a Potential Therapeutic Target, C1q Induces Synapse Loss Via Inflammasome-activating Apoptotic and Mitochondria Impairment Mechanisms in Alzheimer’s Disease. Journal of Neuroimmune Pharmacology. 2023; 18: 267–284. https://doi.org/10.1007/s11481-023-10076-9.

[5] Jha SK, Jha NK, Kumar D, Sharma R, Shrivastava A, Ambasta RK, et al. Stress-Induced Synaptic Dysfunction and Neurotransmitter Release in Alzheimer’s Disease: Can Neurotransmitters and Neuromodulators be Potential Therapeutic Targets? Journal of Alzheimer’s Disease. 2017; 57: 1017–1039. https://doi.org/10.3233/JAD-160623.

[6] Castello MA, Soriano S. Rational heterodoxy: cholesterol reformation of the amyloid doctrine. Ageing Research Reviews. 2013; 12: 282–288. https://doi.org/10.1016/j.arr.2012.06.007.

[7] Martín MG, Pfrieger F, Dotti CG. Cholesterol in brain disease: sometimes determinant and frequently implicated. EMBO Reports. 2014; 15: 1036–1052. https://doi.org/10.15252/embr.201439225.

[8] Grimm MOW, Mett J, Grimm HS, Hartmann T. APP Function and Lipids: A Bidirectional Link. Frontiers in Molecular Neuroscience. 2017; 10: 63. https://doi.org/10.3389/fnmol.2017.00063.

[9] Bai X, Mai M, Yao K, Zhang M, Huang Y, Zhang W, et al. The role of DHCR24 in the pathogenesis of AD: re-cognition of the relationship between cholesterol and AD pathogenesis. Acta Neuropathologica Communications. 2022; 10: 35. https://doi.org/10.1186/s40478-022-01338-3.

[10] Borràs C, Mercer A, Sirisi S, Alcolea D, Escolà-Gil JC, Blanco-Vaca F, et al. HDL-like-Mediated Cell Cholesterol Trafficking in the Central Nervous System and Alzheimer’s Disease Pathogenesis. International Journal of Molecular Sciences. 2022; 23: 9356. https://doi.org/10.3390/ijms23169356.

[11] Fernández-Calle R, Konings SC, Frontiñán-Rubio J, García-Revilla J, Camprubí-Ferrer L, Svensson M, et al. APOE in the bullseye of neurodegenerative diseases: impact of the APOE genotype in Alzheimer’s disease pathology and brain diseases. Molecular Neurodegeneration. 2022; 17: 62. https://doi.org/10.1186/s13024-022-00566-4.

[12] Mahammad S, Parmryd I. Cholesterol depletion using methyl-β-cyclodextrin. Methods in Molecular Biology. 2015; 1232: 91–102. https://doi.org/10.1007/978-1-4939-1752-5_8.

[13] Ilangumaran S, Hoessli DC. Effects of cholesterol depletion by cyclodextrin on the sphingolipid microdomains of the plasma membrane. The Biochemical Journal. 1998; 335: 433–440. https://doi.org/10.1042/bj3350433.

[14] Fabelo N, Martín V, Marín R, Santpere G, Aso E, Ferrer I, et al. Evidence for premature lipid raft aging in APP/PS1 double-transgenic mice, a model of familial Alzheimer disease. Journal of Neuropathology and Experimental Neurology. 2012; 71: 868–881. https://doi.org/10.1097/NEN.0b013e31826be03c.

[15] Zerenturk EJ, Sharpe LJ, Ikonen E, Brown AJ. Desmosterol and DHCR24: unexpected new directions for a terminal step in cholesterol synthesis. Progress in Lipid Research. 2013; 52: 666–680. https://doi.org/10.1016/j.plipres.2013.09.002.

[16] Bai X, Wu J, Zhang M, Xu Y, Duan L, Yao K, et al. DHCR24 Knock-Down Induced Tau Hyperphosphorylation at Thr181, Ser199, Thr231, Ser262, Ser396 Epitopes and Inhibition of Autophagy by Overactivation of GSK3β/mTOR Signaling. Frontiers in Aging Neuroscience. 2021; 13: 513605. https://doi.org/10.3389/fnagi.2021.513605.

[17] Huang Y, Zhang W, Guo X, Zhang Y, Wu J, Zu H. Cellular cholesterol loss by DHCR24 knockdown leads to Aβ production by changing APP intracellular localization. Journal of Lipid Research. 2023; 64: 100367. https://doi.org/10.1016/j.jlr.2023.100367.

[18] Jiang L, Fernandes D, Mehta N, Bean JL, Michaelis ML, Zaidi A. Partitioning of the plasma membrane Ca2+-ATPase into lipid rafts in primary neurons: effects of cholesterol depletion. Journal of Neurochemistry. 2007; 102: 378–388. https://doi.org/10.1111/j.1471-4159.2007.04480.x.

[19] Weber-Boyvat M, Kroll J, Trimbuch T, Olkkonen VM, Rosenmund C. The lipid transporter ORP2 regulates synaptic neurotransmitter release via two distinct mechanisms. Cell Reports. 2022; 41: 111882. https://doi.org/10.1016/j.celrep.2022.111882.

[20] Robinson P, Etheridge S, Song L, Armenise P, Jones OT, Fitzgerald EM. Formation of N-type (Cav2.2) voltage-gated calcium channel membrane microdomains: Lipid raft association and clustering. Cell Calcium. 2010; 48: 183–194. https://doi.org/10.1016/j.ceca.2010.08.006.

[21] Tao W, Lee J, Chen X, Díaz-Alonso J, Zhou J, Pleasure S, et al. Synaptic memory requires CaMKII. eLife. 2021; 10: e60360. https://doi.org/10.7554/eLife.60360.

[22] Fielding CJ, Fielding PE. Cholesterol and caveolae: structural and functional relationships. Biochimica et Biophysica Acta. 2000; 1529: 210–222. https://doi.org/10.1016/s1388-1981(00)00150-5.

[23] Harvey RD, Calaghan SC. Caveolae create local signalling domains through their distinct protein content, lipid profile and morphology. Journal of Molecular and Cellular Cardiology. 2012; 52: 366–375. https://doi.org/10.1016/j.yjmcc.2011.07.007.

[24] Goodall AR, Danks K, Walker JH, Ball SG, Vaughan PF. Occurrence of two types of secretory vesicles in the human neuroblastoma SH-SY5Y. Journal of Neurochemistry. 1997; 68: 1542–1552. https://doi.org/10.1046/j.1471-4159.1997.68041542.x.

[25] Vaughan PF, Peers C, Walker JH. The use of the human neuroblastoma SH-SY5Y to study the effect of second messengers on noradrenaline release. General Pharmacology. 1995; 26: 1191–1201. https://doi.org/10.1016/0306-3623(94)00312-b.

[26] Wang C, Ma Z, Yan DY, Liu C, Deng Y, Liu W, et al. Alpha-Synuclein and Calpains Disrupt SNARE-Mediated Synaptic Vesicle Fusion During Manganese Exposure in SH-SY5Y Cells. Cells. 2018; 7: 258. https://doi.org/10.3390/cells7120258.

[27] Xi YD, Zhang DD, Ding J, Yu HL, Yuan LH, Ma WW, et al. Genistein Inhibits Aβ25-35-Induced Synaptic Toxicity and Regulates CaMKII/CREB Pathway in SH-SY5Y Cells. Cellular and Molecular Neurobiology. 2016; 36: 1151–1159. https://doi.org/10.1007/s10571-015-0311-6.

[28] Qi Z, Zhang Y, Yao K, Zhang M, Xu Y, Zhang J, et al. DHCR24 Knockdown Lead to Hyperphosphorylation of Tau at Thr181, Thr231, Ser262, Ser396, and Ser422 Sites by Membrane Lipid-Raft Dependent PP2A Signaling in SH-SY5Y Cells. Neurochemical Research. 2021; 46: 1627–1640. https://doi.org/10.1007/s11064-021-03273-6.

[29] Mai M, Guo X, Huang Y, Zhang W, Xu Y, Zhang Y, et al. DHCR24 Knockdown Induces Tau Hyperphosphorylation at Thr181, Ser199, Ser262, and Ser396 Sites via Activation of the Lipid Raft-Dependent Ras/MEK/ERK Signaling Pathway in C8D1A Astrocytes. Molecular Neurobiology. 2022; 59: 5856–5873. https://doi.org/10.1007/s12035-022-02945-w.

[30] Chini B, Parenti M. G-protein coupled receptors in lipid rafts and caveolae: how, when and why do they go there? Journal of Molecular Endocrinology. 2004; 32: 325–338. https://doi.org/10.1677/jme.0.0320325.

[31] Bodrikov V, Sytnyk V, Leshchyns’ka I, den Hertog J, Schachner M. NCAM induces CaMKIIalpha-mediated RPTPalpha phosphorylation to enhance its catalytic activity and neurite outgrowth. The Journal of Cell Biology. 2008; 182: 1185–1200. https://doi.org/10.1083/jcb.200803045.

[32] Pani B, Singh BB. Lipid rafts/caveolae as microdomains of calcium signaling. Cell Calcium. 2009; 45: 625–633. https://doi.org/10.1016/j.ceca.2009.02.009.

[33] Ceccaldi PE, Grohovaz F, Benfenati F, Chieregatti E, Greengard P, Valtorta F. Dephosphorylated synapsin I anchors synaptic vesicles to actin cytoskeleton: an analysis by videomicroscopy. The Journal of Cell Biology. 1995; 128: 905–912. https://doi.org/10.1083/jcb.128.5.905.

[34] Benfenati F, Valtorta F, Rubenstein JL, Gorelick FS, Greengard P, Czernik AJ. Synaptic vesicle-associated Ca2+/calmodulin-dependent protein kinase II is a binding protein for synapsin I. Nature. 1992; 359: 417–420. https://doi.org/10.1038/359417a0.

[35] Valtorta F, Pennuto M, Bonanomi D, Benfenati F. Synaptophysin: leading actor or walk-on role in synaptic vesicle exocytosis? BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology. 2004; 26: 445–453. https://doi.org/10.1002/bies.20012.

[36] Thiele C, Hannah MJ, Fahrenholz F, Huttner WB. Cholesterol binds to synaptophysin and is required for biogenesis of synaptic vesicles. Nature Cell Biology. 2000; 2: 42–49. https://doi.org/10.1038/71366.

[37] Koudinov AR, Koudinova NV. Cholesterol homeostasis failure as a unifying cause of synaptic degeneration. Journal of the Neurological Sciences. 2005; 229–230: 233–240. https://doi.org/10.1016/j.jns.2004.11.036.

[38] Koudinov AR, Koudinova NV. Cholesterol, synaptic function and Alzheimer’s disease. Pharmacopsychiatry. 2003; 36: S107–S112. https://doi.org/10.1055/s-2003-43055.

[39] Zhang WB, Huang Y, Guo XR, Zhang MQ, Yuan XS, Zu HB. DHCR24 reverses Alzheimer’s disease-related pathology and cognitive impairment via increasing hippocampal cholesterol levels in 5xFAD mice. Acta Neuropathologica Communications. 2023; 11: 102. https://doi.org/10.1186/s40478-023-01593-y.

[40] Sanhueza M, McIntyre CC, Lisman JE. Reversal of synaptic memory by Ca2+/calmodulin-dependent protein kinase II inhibitor. The Journal of Neuroscience. 2007; 27: 5190–5199. https://doi.org/10.1523/JNEUROSCI.5049-06.2007.

[41] Ghosh A, Giese KP. Calcium/calmodulin-dependent kinase II and Alzheimer’s disease. Molecular Brain. 2015; 8: 78. https://doi.org/10.1186/s13041-015-0166-2.

[42] Reese LC, Laezza F, Woltjer R, Taglialatela G. Dysregulated phosphorylation of Ca(2+) /calmodulin-dependent protein kinase II-α in the hippocampus of subjects with mild cognitive impairment and Alzheimer’s disease. Journal of Neurochemistry. 2011; 119: 791–804. https://doi.org/10.1111/j.1471-4159.2011.07447.x.

[43] Ly PTT, Song W. Loss of activated CaMKII at the synapse underlies Alzheimer’s disease memory loss. Journal of Neurochemistry. 2011; 119: 673–675. https://doi.org/10.1111/j.1471-4159.2011.07473.x.

[44] Wang YJ, Chen GH, Hu XY, Lu YP, Zhou JN, Liu RY. The expression of calcium/calmodulin-dependent protein kinase II-alpha in the hippocampus of patients with Alzheimer’s disease and its links with AD-related pathology. Brain Research. 2005; 1031: 101–108. https://doi.org/10.1016/j.brainres.2004.10.061.

[45] Popugaeva E, Pchitskaya E, Bezprozvanny I. Dysregulation of neuronal calcium homeostasis in Alzheimer’s disease - A therapeutic opportunity? Biochemical and Biophysical Research Communications. 2017; 483: 998–1004. https://doi.org/10.1016/j.bbrc.2016.09.053.

[46] Llinás R, McGuinness TL, Leonard CS, Sugimori M, Greengard P. Intraterminal injection of synapsin I or calcium/calmodulin-dependent protein kinase II alters neurotransmitter release at the squid giant synapse. Proceedings of the National Academy of Sciences of the United States of America. 1985; 82: 3035–3039. https://doi.org/10.1073/pnas.82.9.3035.

[47] McGuinness TL, Lai Y, Greengard P. Ca2+/calmodulin-dependent protein kinase II. Isozymic forms from rat forebrain and cerebellum. The Journal of Biological Chemistry. 1985; 260: 1696–1704.

[48] Bykhovskaia M. Synapsin regulation of vesicle organization and functional pools. Seminars in Cell & Developmental Biology. 2011; 22: 387–392. https://doi.org/10.1016/j.semcdb.2011.07.003.

[49] Vasileva M, Horstmann H, Geumann C, Gitler D, Kuner T. Synapsin-dependent reserve pool of synaptic vesicles supports replenishment of the readily releasable pool under intense synaptic transmission. The European Journal of Neuroscience. 2012; 36: 3005–3020. https://doi.org/10.1111/j.1460-9568.2012.08225.x.

[50] Zhang M, Augustine GJ. Synapsins and the Synaptic Vesicle Reserve Pool: Floats or Anchors? Cells. 2021; 10: 658. https://doi.org/10.3390/cells10030658.

[51] Rizzoli SO, Betz WJ. Synaptic vesicle pools. Nature Reviews. Neuroscience. 2005; 6: 57–69. https://doi.org/10.1038/nrn1583.

[52] Alabi AA, Tsien RW. Synaptic vesicle pools and dynamics. Cold Spring Harbor Perspectives in Biology. 2012; 4: a013680. https://doi.org/10.1101/cshperspect.a013680.

[53] Rosenmund C, Stevens CF. Definition of the readily releasable pool of vesicles at hippocampal synapses. Neuron. 1996; 16: 1197–1207. https://doi.org/10.1016/s0896-6273(00)80146-4.

[54] Pieribone VA, Shupliakov O, Brodin L, Hilfiker-Rothenfluh S, Czernik AJ, Greengard P. Distinct pools of synaptic vesicles in neurotransmitter release. Nature. 1995; 375: 493–497. https://doi.org/10.1038/375493a0.

[55] Rose T, Schoenenberger P, Jezek K, Oertner TG. Developmental refinement of vesicle cycling at Schaffer collateral synapses. Neuron. 2013; 77: 1109–1121. https://doi.org/10.1016/j.neuron.2013.01.021.

[56] Sansevrino R, Hoffmann C, Milovanovic D. Condensate biology of synaptic vesicle clusters. Trends in Neurosciences. 2023; 46: 293–306. https://doi.org/10.1016/j.tins.2023.01.001.

[57] Bähler M, Greengard P. Synapsin I bundles F-actin in a phosphorylation-dependent manner. Nature. 1987; 326: 704–707. https://doi.org/10.1038/326704a0.

[58] Maritzen T, Haucke V. Coupling of exocytosis and endocytosis at the presynaptic active zone. Neuroscience Research. 2018; 127: 45–52. https://doi.org/10.1016/j.neures.2017.09.013.

[59] Yan P, Liu H, Zhou T, Sun P, Wang Y, Wang X, et al. Crosstalk of Synapsin1 palmitoylation and phosphorylation controls the dynamicity of synaptic vesicles in neurons. Cell Death & Disease. 2022; 13: 786. https://doi.org/10.1038/s41419-022-05235-4.

[60] Shupliakov O, Haucke V, Pechstein A. How synapsin I may cluster synaptic vesicles. Seminars in Cell & Developmental Biology. 2011; 22: 393–399. https://doi.org/10.1016/j.semcdb.2011.07.006.

[61] Bloom O, Evergren E, Tomilin N, Kjaerulff O, Löw P, Brodin L, et al. Colocalization of synapsin and actin during synaptic vesicle recycling. The Journal of Cell Biology. 2003; 161: 737–747. https://doi.org/10.1083/jcb.200212140.

[62] Wiedenmann B, Franke WW. Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38,000 characteristic of presynaptic vesicles. Cell. 1985; 41: 1017–1028. https://doi.org/10.1016/s0092-8674(85)80082-9.

[63] White DN, Stowell MHB. Room for Two: The Synaptophysin/Synaptobrevin Complex. Frontiers in Synaptic Neuroscience. 2021; 13: 740318. https://doi.org/10.3389/fnsyn.2021.740318.

[64] Jang DJ, Park SW, Kaang BK. The role of lipid binding for the targeting of synaptic proteins into synaptic vesicles. BMB Reports. 2009; 42: 1–5. https://doi.org/10.5483/bmbrep.2009.42.1.001.

[65] Mitter D, Reisinger C, Hinz B, Hollmann S, Yelamanchili SV, Treiber-Held S, et al. The synaptophysin/synaptobrevin interaction critically depends on the cholesterol content. Journal of Neurochemistry. 2003; 84: 35–42. https://doi.org/10.1046/j.1471-4159.2003.01258.x.

Funding

Shanghai Municipal Health Bureau, Shanghai(201740209)

Shanghai Jinshan District Key Medical Foundation, Shanghai(JSZK2019A06)

Jinshan Hospital of Fudan University Development Foundation(HBXK-2022-2)