Increasing Membrane Cholesterol Level Increases the Amyloidogenic Peptide by Enhancing the Expression of Phospholipase C
Cerebral elevation of 42-residue amyloid β-peptide (Aβ42) triggers neuronal dysfunction in Alzheimer's disease (AD). Even though a number of cholesterol modulating agents have been shown to affect Aβ generation, the role of cholesterol in the pathogenesis of AD is not clear yet. Recently, we have shown that increased membrane cholesterol levels downregulates phosphatidylinositol 4,5-bisphosphate (PIP2) via activation of phospholipase C (PLC). In this study, we tested whether membrane cholesterol levels may affect the Aβ42 production via changing PIP2 levels. Increasing membrane cholesterol levels decreased PIP2 and increased secreted Aβ42. Supplying PIP2, by using a PIP2-carrier system, blocked the effect of cholesterol on Aβ42. We also found that cholesterol increased the expressions of β1 and β3 PLC isoforms (PLCβ1, PLCβ3). Silencing the expression of PLCβ1 prevented the effects of cholesterol on PIP2 levels as well as on Aβ42 production, suggesting that increased membrane cholesterol levels increased secreted Aβ42 by downregulating PIP2 via enhancing the expression of PLCβ1. Thus, cholesterol metabolism may be linked to Aβ42 levels via PLCβ1 expression and subsequent changes in PIP2 metabolism.
AD is a progressive and irreversible neurodegenerative disorder leading to cognitive, memory, and behavioral impairments. Cerebral elevation and accumulation of Aβ are necessary steps in the pathogenesis of AD [1–3]. Sequential proteolytic cleavages of amyloid precursor protein (APP) by membrane-bound β-secretase and γ-secretase produce two major isoforms of Aβ, Aβ40, and Aβ42. Therefore, this pathway is called amyloidogenic pathway. More amyloidogenic Aβ42 is considered as a pathogenic agent [4, 5]. Alternatively, APP can be sequentially processed by α-secretase, and γ-secretase, precluding Aβ production (nonamyloidogenic pathway). Even though advanced age serves as a major risk factor, approximately 5% of AD cases are familial (FAD), and some of them are attributable to autosomal dominant mutations in presenilin (PS) genes, PS1 and PS2. PS1 and PS2 function as catalytic subunits of γ-secretase, and FAD mutations in PSs affect APP processing increasing the ratio of Aβ42 to Aβ40 [6–8].
Growing evidence indicates that dysregulation of lipid pathways have regulatory consequences for APP processing and Aβ generation . Especially, cholesterol has been suggested to participate in the etiology of AD by increasing the generation of Aβ . Cholesterol can directly regulate the activities of β-secretase or γ-secretase to alter amyloidogenesis [11–13]. Alternatively, changes in cholesterol level may affect the lipid environment for APP processing and Aβ generation. APP is located either within or outside of lipid rafts. Since BACE1 (β-secretase) is predominantly located in lipid rafts, APP processing occurring within lipid rafts is amyloidogenic, whereas APP processing occurring outside lipid rafts is considered nonamyloidogenic . When cholesterol is depleted, the association of BACE1 with lipid rafts is decreased, producing less Aβ [15–17]. In contrast, increasing cholesterol induces the co-clustering of APP and BACE1, producing more Aβ . From these results it could be hypothesized that high cholesterol levels may be responsible for initiating the pathogenesis of AD. However, it was recently demonstrated that lowering cholesterol levels results in increased β-amyloid production in neurons . Abad-Rodriguez et al. reported that lowering the intracellular levels of cholesterol could increase the rate of amyloidogenic processing of APP by placing the hydrolyzing enzyme (BACE1) and APP in close proximity within the same intracellular compartments. Therefore, more experiments will be needed to clarify the conflicting results about the role of cholesterol in pathogenesis of AD.
PIP2 is known as one of phospholipid component of cell membrane , playing important regulatory roles in a variety of cell functions, such as rearrangement of the cytoskeleton and membrane trafficking . We have reported that FAD-linked PS mutants down-regulate PIP2 levels, and that PIP2 levels are inversely correlated to the production of Aβ42 . We also demonstrated that increased membrane cholesterol level decreases the level of PIP2 via the activation of PLC . Therefore, there exists a crosstalk between two plasma membrane-enriched lipids, cholesterol and PIP2. Considering the close relationship between PIP2 levels and the production of Aβ42, we suspected that increased membrane cholesterol levels affect the Aβ42 production via down-regulating PIP2 levels. In this study, we found that membrane cholesterol decreased PIP2 levels and increased secreted Aβ42. Supplying PIP2 by using a PIP2-carrier system blocked the effect of cholesterol, which might indicate that the effect of cholesterol on Aβ42 was by downregulation of PIP2 levels. Enriching membrane with cholesterol increased the expression of some PLC isoforms, such as PLCβ1 and PLCβ3. Blocking the new protein synthesis prevented the effect of cholesterol on PIP2 levels as well as on Aβ42 production. We found that the expression of PLCβ1 was specifically linked to Aβ42 production. These results suggest that increased membrane cholesterol levels, and FAD-linked PS mutations may share the same molecular mechanism, that is, the downregulation of PIP2, which may serve as the molecule linking cholesterol metabolism to the pathogenesis of AD.
2. Materials and Methods
2.1. Cell Culture
HeLa cells stably transfected with APP751 carrying the Swedish mutation (APPsw) were cultured at 37°C, 5% CO2, in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum containing 100 units/mL penicillin, 100 μg/mL streptomycin, 260 μg/mL Zeocin, and 400 μg/mL G418. Human neuroblastoma SH-SY5Y cells were cultured in DMEM containing 10% heat-inactivated fetal bovine serum, 100 units/mL penicillin, and 100 μg/mL streptomycin.
To enrich the cells with cholesterol, cells were exposed to DMEM culture medium containing methyl-β-cyclodextrin (MβCD, Sigma, USA) saturated with cholesterol (water-soluble cholesterol). During the incubation, cells were maintained in a humidified CO2 incubator at 37°C. In some experiments, to avoid the use of MβCD, cells were incubated with cholesterol which was solubilized by sonication. For this purpose, cholesterol in methanol/chloroform mixture (1 : 1 v/v) was dried under nitrogen gas and sonicated for 2 min in phosphate-buffered saline before use.
In some experiments, cells were pretreated with 10 μM Actinomycin-D (Sigma) or 50 μg/mL cyclohexamide (Sigma) for 0.5 h before 75 μM water-soluble cholesterol was added. PIP2 were delivered into the cells using the PIP2-carrier system (Echelon Bioscience Inc., USA). Carrier-PIP2 complex was incubated with APP-transfected HeLa cells for 4 h in the absence or presence of 75 μM water-soluble cholesterol.
2.3. Antisense Oligonucleotides Treatments
The antisense oligonucleotides (IDT, USA) targeted at PLCβ1 and PLCβ3 were designed to be complementary to the 5′ sequences and were phosphorothionated at all positions to minimize intracellular cleavage by enzymes and to enhance their stability (5′-actccgggttgagccccggc-3′ for PLCβ1 and 5′-tccaactgcagcgcgtggac-3′ for PLCβ3). Antisense oligonucleotide (5′-gccccgtatgaccgcgccgg-3′) having no target was used as a control in all of experiments. The APP-transfected HeLa cells were plated at a density of 2 × 106 cells per 60 mm dish and incubated overnight and then treated with the 10 μM antisense oligonucleotides for 4 h in DMEM culture medium without serum. After treatment, the medium was replaced by a new medium containing 10 μM antisense oligonucleotides with or without water-soluble cholesterol for 2 h. Media were collected to measure levels of Aβ, and cells were homogenized to confirm PLC expression levels and PIP2 levels.
2.4. Cholesterol Assay
Filipin staining of cells (0.05%, DMSO 1%) was performed for 1 h at room temperature after cholesterol enrichment to confirm the changes of free cholesterol levels at the plasma membrane. Fluorescence images were obtained using a LSM 710 confocal microscope (Zeiss) using laser emitting at 351 nm. Images were quantified to obtain the mean fluorescence density values of plasma membrane from the edge of the cell to 500 nm inside using the ImageJ program.
2.5. sAPPα, sAPPβ, and Aβ Peptide Assay
Levels of Aβ peptides were assayed by using the Invitrogen. Aβ ELISA kits (USA) or Wako β-amyloid ELISA kits (high-sensitive; Japan). For sAPPα, sAPPβ detection, samples were analyzed by ELISA kit from IBL (USA). APP-transfected HeLa cells at 80% confluence in a 35 mm dish were cultured for 8 h with water-soluble cholesterol in DMEM culture medium without serum. Control cells were treated similarly and incubated with serum-free DMEM solution without any cholesterol. After exposure to cholesterol, supernatants were collected to measure levels of Aβ, sAPPα, or sAPPβ. To detect Aβ from SH-SY5Y cells, supernatants were desalted using PD-10 desalting column (GE Healthcare, USA), dried, and reconstituted in water. The samples were analyzed by ELISA kits according to the supplier’s instructions.
2.6. Protein Extraction
Cell fractionations were obtained by homogenizing with hypotonic buffer using a 23-gauge needle. The samples were then centrifuged at 1,000 for 10 min at 4°C to remove nuclei and debris. Supernatants were separated by centrifugation at 100,000 for 1 h at 4°C into membrane (pellet) and cytosol (supernatant) fractions. Whole cell lysates were prepared by homogenizing with lysis buffer (10 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, 0.25% Nonidet P-40, 2 mM EDTA, pH7.4) using a cell scraper. The lysed cells were centrifuged at 12,000 for 10 min at 4°C. The protein in the supernatant was determined by Bradford assay (Bio-rad, USA).
2.7. Western Blot Analysis
Proteins were resolved on SDS-PAGE and transferred to nitorcelluose membrane. Membranes were blocked with 5% nonfat milk powder in Tris-buffered saline/Tween 20 (TBST) for 1 h at room temperature, then incubated with rabbit polyclonal anti-PLCβ1 (SC-9050), PLCβ2 (SC-206), PLCβ3 (SC-13958), PLCβ4 (SC-20760), PLCγ2 (SC-9015), mouse monoclonal anti-PLCγ1 (SC-7290) antibodies (Santa Cruz Biotechnology, USA), anti-APP antibody (LN27, Zymed), anti β-actin (A5441, Sigma), and rabbit anti β-tubulin (T2200, Sigma) for overnight at 4°C. Dilutions were 1 : 500 for PLC isozymes and 1 : 4000 for β-tubulin, β-actin, and APP. After washing, membranes were incubated for 1 h at room temperature with horseradish peroxidase-conjugated goat anti-rabbit IgG or goat anti-mouse IgG antibodies (1 : 2000 dilution; Zymed, USA) and washed. Peroxidase activity was visualized with enhanced chemilluminescence. Blots were quantified with the Multi Gauge software using a LAS-3000 system (FugiFilm, Japan).
2.8. PIP2 Assay
The amount of PIP2 extracted from APP-transfected HeLa cells were measured by using PIP2 Mass ELISA kit (Echelon Biosciences Inc., USA). PIP2 was extracted from the control cells or cells treated with water-soluble cholesterol according to the supplier’s instructions. Cellular PIP2 quantities were estimated by comparing the values from the standard curve, which showed linear relationship at the range from 0.5 to 1000 pM concentrations.
2.9. Statistical Analysis
Data was expressed as mean ± SEM. Statistical comparisons between the controls and treated experimental groups were performed using the Student’s t-test. was considered statistically significant.
3.1. Increasing Membrane Cholesterol Levels Downregulates PIP2 and Increases Secreted Aβ42
MβCD, a water-soluble cyclic oligosaccharide, has hydrophobic cavity that is able to encapsulate insoluble compounds, thus enhances the solubility of cholesterol. MβCD saturated with cholesterol (water-soluble cholesterol) has been used to increase membrane cholesterol level, since it acts as a cholesterol donor [24–26]. We incubated APP-transfected HeLa cells with 15, 75, or 150 μM water-soluble cholesterol for 8 h, and filipin staining was performed for 1 h at room temperature to monitor the membrane cholesterol level. Typical confocal image in Figure 1(a) shows that the membrane cholesterol level increased by 75 μM water-soluble cholesterol. The changes of cholesterol level were confirmed by quantifying the filipin fluorescent intensities from plasma membranes. By incubating cells with 15 and 75 μM water-soluble cholesterol, the fluorescent intensities were increased by % and % (), respectively (Figure 1(b)). We also tested the time-dependent accumulation of cholesterol in the membrane by incubating cells with 75 μM water-soluble cholesterol. Cholesterol levels increased after 0.5 h, and it steadily increased further after 1.5 h or 5 h (Supplementary Figures 1(a) and 1(b), see Supplementary Material available online at http://dx.doi.org/10.1155/2013/407903). From these results we concluded that the direct administration of the water-soluble cholesterol leads to the increases in the membrane cholesterol levels.
Recently, we have reported that augmentation of membrane cholesterol levels downregulates PIP2 level . To validate this observation in the current system, APP-transfected HeLa cells were incubated with 15, 75, or 150 μM water-soluble cholesterol for 8 h, and the steady state levels of PIP2 were measured using a PIP2 ELISA. PIP2 levels in 75 and 150 μM cholesterol-treated cells were downregulated by % and % (), respectively (Figure 1(c)). We also tested the time-dependent effect of increased membrane cholesterol on the levels of PIP2 by incubating cells with 75 μM water-soluble cholesterol. As shown in Figure 1(d), the steady state levels of PIP2 after 1.5 h and 5 h incubation time were downregulated by % and % (), respectively.
Since we have reported that cellular PIP2 levels are closely correlated with the Aβ42 levels , we tested the effect of increased membrane cholesterol levels on secreted Aβ. For this purpose, APP-transfected HeLa cells were incubated for 8 h with 15, 75, or 150 μM water-soluble cholesterol, and Aβ levels were measured from the conditioned media by using an ELISA kits specific for Aβ40 or Aβ42. The secreted Aβ40 levels were not changed by increased membrane cholesterol levels (open bars in Figure 1(e)). However, Aβ42 levels were increased by %, and % () when cells were incubated with 75, and 150 μM water-soluble cholesterol, respectively (closed bars in Figure 1(e)). We also tested the effect of increased membrane cholesterol levels on the levels of secreted Aβ from neuroblastoma SH-SY5Y cells. The level of endogenous Aβ42 increased significantly (closed bars in Figure 1(f)), while the endogenous Aβ40 level was not changed (open bars in Figure 1(f)). Thus, these results suggest that the effect of cholesterol enrichment is specific to Aβ42 levels, and is not cell-type specific.
Aβ is produced by the sequential cleavages of APP by β-secretase followed by γ-secretase. Alternatively, APP can be cleaved sequentially by α-secretase followed by γ-secretase precluding the production of Aβ. Thus, the effect of increased membrane cholesterol levels on Aβ levels can occur in any of those processes. To begin to investigate the effects of membrane cholesterol on APP processing, we first examined the levels of full-length APP. Increased membrane cholesterol levels led to a moderate increase in the full-length APP levels (Figure 2(a)). However, the increase was less than 10% (). Then, we tested the effects of increased membrane cholesterol levels on the activities of α-secretase and β-secretase. For this purpose, we measured the levels of sAPPα and sAPPβ from the conditioned media using specific ELISA kits, since they are produced via the activities of α-secretase and β-secretase, respectively. In this experiment, we used sAPPβ ELISA kit for Swedish mutant. As shown in Figure 2(b), the levels of both sAPPα and sAPPβ were also slightly increased by increased membrane cholesterol levels, which might be due to the increased level of their precursor, APP. However, the amount of increased sAPPβ level was not robust to explain the Aβ42-selective changes associated with increased membrane cholesterol levels. Since membrane cholesterol levels affect Aβ42 but not Aβ40, it is conceivable that the effects of cholesterol may influence the specificity of γ-secretase-mediated cleavage of amyloidogenic APP C-terminal fragments (e.g., C99).
3.2. Intracellular Delivery of PIP2 Prevents the Effect of Increased Membrane Cholesterol Levels on Secreted Aβ42
In order to elucidate the role of PIP2 for the effect of increased membrane cholesterol levels on secreted Aβ42, we used a PIP2-carrier system for the intracellular delivery of PIP2. Because carrier compounds are “charge-neutralization” species, it could deliver the anionic PIP2 into the cells . After the carriers were added at a one-to-one molar ratio with PIP2 at room temperature, the complex was diluted to the desired final concentration. Then, the carrier-PIP2 complex was incubated with cells for 4 h before Aβ levels were measured from the conditioned media.
The presence of 2 μM and 5 μM carrier-PIP2 complex decreased secreted Aβ42 levels by % and % (), respectively (open bars in Figure 2(c)). This result is consistent with our previous result showing the close correlation between PIP2 levels and Aβ42 production . In the absence of carrier-PIP2 complex, incubating cells with 75 μM water-soluble cholesterol for 4 h increased Aβ42 levels by % (), which is consistent with the result in Figure 1(e). However, the presence of either 2 μM or 5 μM carrier-PIP2 complex completely prevented the effect of water-soluble cholesterol on Aβ42 levels (closed bars in Figure 2(c)). These results suggest that the relative levels of cholesterol and PIP2 correlate closely with secreted Aβ42 levels in a positive or negative manner, respectively. Unlike Aβ42, the Aβ40 levels were not affected by the presence of carrier-PIP2 complex (open bars in Figure 2(d)). Also, the Aβ40 levels were not affected by increased membrane cholesterol levels in the presence of carrier-PIP2 complex (closed bars in Figure 2(d)), which was consistent with the specific effect of cholesterol on Aβ42 level.
3.3. Increasing Membrane Cholesterol Level Increases PLCβ1 and PLCβ3 Expressions
The major catabolic pathway for PIP2 is the hydrolysis by membranous PLC. Since we suspected that the effect of increased membrane cholesterol levels on the secreted Aβ42 levels is due to the downregulation of PIP2, we tested whether increased membrane cholesterol levels affect the activity of PLC. We first examined the expression levels of PLC isoforms by monitoring them using Western blot analysis from cytosol and membrane fractions in APP-transfected HeLa cells. The expression levels of PLCβ1 were significantly increased by cholesterol only in membrane fractions as shown in Figure 3(a). Densitometry analysis of the bands corresponding to PLCβ1 clearly supports this conclusion (Figure 3(b); ). After 0.5 h incubation time with 75 μM water-soluble cholesterol, expression levels of PLCβ1 were increased by two folds in membrane fractions. The effect of cholesterol on PLCβ1 expression lasted as long as 5 h. The expression level of PLCβ3 was also increased in membrane fractions by cholesterol (Figures 3(a) and 3(c)). However, the effect was significant only after 5 h incubation time.
We used MβCD to increase cholesterol levels. However, the use of MβCD may cause nonspecific effects in addition to enrichment of membrane cholesterol levels . To avoid the use of MβCD, free cholesterol was solubilized in phosphate-buffered saline using sonication. When cells were incubated with 75 μM solubilized cholesterol for 1 h, the expression of PLCβ1 in the membrane fraction was increased (Supplementary Figure 2). At 1 h incubation time, the expression of PLCβ3 was not changed. These results indicated that the effect of cholesterol on PLCβ1 expression was not due to the nonspecific effect of MβCD.
Cholesterol-induced increases in the levels of PLCβ1 and PLCβ3 were observed almost exclusively in the membrane fraction. Also, increased membrane cholesterol levels did not change the expression levels of PLCγ2 as shown in Figure 3(a) from a typical experiment. The expression levels of other PLC isoforms (PLCβ2, PLCβ4, and PLCγ1) were not changed either (Supplementary Figure 3(a)). We also observed specific increase of PLCβ1 and PLCβ3 expressions from SH-SY5Y cells by enriching membrane cholesterol levels (Supplementary Figure 3(b)).
To determine whether the effect of cholesterol on PLC expression was due to increased transcription, cells were preincubated for 10 min with the transcription inhibitor, actinomycin-D (Act-D, 10 μM), or with the translation inhibitor, cyclohexamide (CHX, 50 μg/mL). Then, cells were incubated further for 0.5 h with 75 μM water-soluble cholesterol. Representative Western blots for PLCβ1 and PLCβ3 from membrane fractions are shown in Figure 4(a). The effect of cholesterol on PLCβ1 and PLCβ3 expressions was completely prevented either by Act-D or by CHX, indicating that the effect of cholesterol on PLCβ1 and PLCβ3 expression was via the up-regulation of transcription. Similar results were obtained from 4 different experiments. Then, we tested the effect of Act-D on the steady state level of PIP2. As we expected, PIP2 levels were downregulated by % () when cells were treated with 75 μM water-soluble cholesterol for 1 h in the absence of Act-D (Figure 4(b)). However, the effect of cholesterol on PIP2 levels was completely prevented when cells were pretreated with Act-D (). These results confirmed that increased membrane cholesterol levels increases the expression of PLCβ1 and PLCβ3, leading to the downregulation of PIP2 levels. Next, we tested the effect of Act-D on the level of secreted Aβ42. For this purpose, cells were incubated for 4 h with 75 μM water-soluble cholesterol with or without Act-D. As expected, cholesterol increased the levels of secreted Aβ42 by % () in the absence of Act-D (Figure 4(c)). However, the presence of Act-D completely prevented the effect of increased membrane cholesterol levels on Aβ42 (). These results suggest that the increase of PLC transcription by cholesterol induced the downregulation of PIP2 levels, which increased secreted Aβ42 levels.
3.4. Inhibition of PLCβ1 Expression Prevents the Effect of Increased Membrane Cholesterol Levels on Secreted Aβ42
We tested whether PLCβ1 or PLCβ3 is required for the observed effects of cholesterol on Aβ42. To avoid any further transfection, cells were pretreated with 10 μM antisense oligonucleotides against PLC isoforms for 4 h. Then the media were replaced by fresh media containing the same antisense oligonucleotides with or without 75 μM water-soluble cholesterol, followed by additional incubation for 2 h. Antisense oligonucleotides having no specific target were used for controls in all of experiments. A typical western blot for PLCβ1 is shown in Figure 5(a), and the band densities are expressed relative to β-tubulin density in Figure 5(c) (). As we expected, cholesterol increased PLCβ1 expression. However, the presence of antisense oligonucleotides against PLCβ1 (anti β1) blocked the effect of cholesterol. Similarly, an increase of PLCβ3 expression by cholesterol was blocked by antisense oligonucleotides against PLCβ3 (anti β3; Figures 5(b) and 5(d)). These results demonstrated that these antisense oligonucleotides were very effective to block the effect of increased membrane cholesterol levels on the expression of PLCβ1 and PLCβ3.
Then, we tested the effect of these antisense oligonucleotides on the PIP2 levels. In the presence of control antisense oligonucleotides (−anti β1), cholesterol decreased PIP2 levels by % (Figure 6(a); ). In the presence of antisense oligonucleotide against PLCβ1 (+anti β1), cholesterol decreased PIP2 levels only by % (). In contrast, cholesterol decreased PIP2 levels by % () even in the presence of antisense oligonucleotide against PLCβ3 (+anti β3; Figure 6(b)). Thus, inhibiting PLCβ1 expression, but not inhibiting PLCβ3 expression, prevented the effect of increased membrane cholesterol levels on PIP2 levels. These results may suggest that the downregulation of PIP2 levels by cholesterol enrichment is specifically via the increased expression of PLCβ1.
Next, we tested the effect of increased membrane cholesterol levels on the levels of secreted Aβ42 in the presence of these antisense oligonucleotides. As we expected, cholesterol increased the levels of secreted Aβ42 by % () in the presence of control antisense oligonucleotides (Figure 6(c)). In the presence of antisense oligonucleotide against PLCβ1, cholesterol increased the secreted Aβ42 level only by % (). Thus, PLCβ1 antisense oligonucleotide almost completely prevented the effect of cholesterol on the secreted Aβ42 levels. In contrast, the presence of PLCβ3 antisense oligonucleotides failed to prevent the effect of cholesterol. The Aβ42 production was still increased by % (Figure 6(d); ). Together, these results strongly suggest that increased membrane cholesterol levels increased the levels of secreted Aβ42 by down-regulating PIP2 levels via specific enhancement of PLCβ1 expression.
We have previously reported that FAD-linked PS mutants down-regulate PIP2 levels, which are closely related to increased Aβ42 level . In addition, up, or downregulation of PIP2 levels by pharmacological means decreases or increases the production of Aβ42, respectively. In this study, we showed that cholesterol enrichment increases the secreted Aβ42 levels by down-regulating PIP2 levels. Thus, there exists a close relationship between PIP2 levels and Aβ42 levels, consistent with our previous results. Recently, we demonstrated that enrichment of cholesterol decreases the levels of PIP2 via the activation PLC . Therefore, it was suggested that there exists a crosstalk between two plasma membrane-enriched lipids, cholesterol, and PIP2. In this paper, we confirmed that cholesterol decreases the level of PIP2 via the activation PLC. In addition, we showed that cholesterol specifically increases the expression levels of PLCβ1 and PLCβ3. Consistent with this conclusion, inhibiting transcription prevented the effects of cholesterol not only on PIP2 levels but also on the production of Aβ42. Also, the inhibition of PLCβ1 expression, but not that of PLCβ3, prevented the effects of cholesterol, indicating a close link between PLCβ1 and regulation of PIP2 levels.
Although PIP2 is a minor component in the plasma membrane, it plays important regulatory roles in a variety of cell functions, such as rearrangement of the cytoskeleton and membrane trafficking . The concept of spatially confined PIP2 pools was proposed to explain the multiple roles of PIP2 . Cholesterol- and sphingolipid-rich rafts may serve to confine PIP2 within the plasma membrane, allowing PIP2 hydrolysis to occur locally and restrict signaling mechanisms to the site of activation [30, 31]. In addition to this confined regulation of PIP2, it is possible that the steady-state level of PIP2 is dynamically determined by the concerted action of phosphoinositide kinases and phosphatases. In this study, we showed another way of regulating PIP2 levels within specific microdomains: cholesterol content in a specific microdomain may regulate PIP2 levels via PLC activity. Interestingly, PLCβ1 is shown to localize in detergent-resistant membrane microdomains prepared from the synaptic plasma membrane fraction of rat brain . For this reason, increase of PLCβ1 expression by cholesterol enrichment may directly induce downregulation of PIP2 in the confined microdomain.
The effect of cholesterol enrichment on Aβ secretion was recently demonstrated . It was shown that cholesterol increases clathrin-dependent APP endocytosis, and that it is likely the direct cause of the increased Aβ42 secretion. Since PIP2 is a key regulator for the rearrangement of the cytoskeleton and membrane trafficking , it is possible that the downregulation of PIP2 may be the underlying mechanism for the increased clathrin-dependent APP endocytosis by cholesterol enrichment. Alternatively, the downregulation of PIP2 may directly activate γ-secretase since PIP2 is shown to inhibit γ-secretase activity by suppressing its association with the substrate . It is also possible that PIP2 induces changes in γ-secretase conformation to alter the generation of Aβ42. Recent reports show that the changes in PS1 conformation by various manipulations of PS1 itself, Pen2, Aph1, APP, and pharmacological agents known as γ-secretase modulators can allosterically modify γ-secretase catalytic specificity, leading to increase of Aβ42/Aβ40 ratio [34, 35]. Further studies will be needed to clarify the role of PIP2 in the production of Aβ.
The physiological relevance of our finding is not clear since a large increase of cholesterol in the membrane was required to induce meaningful downregulation of PIP2 and up-regulation of Aβ42 in HeLa cells (Figures 1(c) and 1(e)). In SH-SY5Y cells, however, Aβ42 level was significantly increased even by 15 μM water-soluble cholesterol (Figure 1(f)). We also observed a significant increase of PLCβ1 expression and downregulation of PIP2 level by 15 μM water-soluble cholesterol when we used HEK cells (data not shown). Thus, it seems that the effective amount of cholesterol to induce changes in PIP2 and Aβ42 levels is cell type dependent. However, even the minimal increase of Aβ42 by the mild cholesterol enrichment will have profound cytotoxic effects since the cytotoxicity of Aβ42 is caused by its small molecular aggregates, such as dimer form of Aβ42 , and the produced Aβ42 will accumulate. It has been demonstrated that cholesterol levels in the brains of AD patients are increased [36, 37]. The levels of cholesterol increase in the brain even during normal aging . In an AD brain, cholesterol homeostasis is impaired, and cholesterol retention likely enhances Aβ production . These results may suggest that high cholesterol levels in the brain participate in the etiology of AD by increasing the generation of Aβ. However, further work is needed to understand how cholesterol is implicated in AD pathogenesis. In this study, we showed that membrane cholesterol levels may share the same molecular mechanism with FAD PS mutations, that is, downregulation of PIP2, for the increased generation of Aβ42. Therefore, PIP2 may serve as the molecule linking cholesterol metabolism to the pathogenesis of AD.
Conflict of Interests
The authors of the paper do not have a direct financial relation with the commercial identity mentioned in the paper that might lead to a conflict of interests for any of the authors.
This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) and funded by the Ministry of Education, Science and Technology (2009-0072220), Samsung Biomedical Research Institute (B-B1-003) to S. Chung, and by NIH Grant (NS074536) to T.-W. Kim.
Supplementary Figure 1: Water-soluble cholesterol increases membrane cholesterol levels.
Supplementary Figure 2: The expression level of PLCβ1 was increased by cholesterol augmentation.
Supplementary Figure 3: Augmentation of membrane cholesterol levels did not increase the expression levels of PLCβ2, PLCβ4, and PLCγ1 from APP-transfected HeLa cells.
J. A. Hardy and G. A. Higgins, “Alzheimer's disease: the amyloid cascade hypothesis,” Science, vol. 256, no. 5054, pp. 184–185, 1992.View at: Google Scholar
D. Scheuner, C. Eckman, M. Jensen et al., “Secreted amyloid β-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease,” Nature Medicine, vol. 2, no. 8, pp. 864–870, 1996.View at: Publisher Site | Google Scholar
N. Landman, S. Y. Jeong, S. Y. Shin et al., “Presenilin mutations linked to familial Alzheimer's disease cause an imbalance in phosphatidylinositol 4,5-bisphosphate metabolism,” Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 51, pp. 19524–19529, 2006.View at: Publisher Site | Google Scholar
A. E. Christian, M. P. Haynes, M. C. Phillips, and G. H. Rothblat, “Use of cyclodextrins for manipulating cellular cholesterol content,” Journal of Lipid Research, vol. 38, no. 11, pp. 2264–2272, 1997.View at: Google Scholar
V. G. Romanenko, G. H. Rothblat, and I. Levitan, “Modulation of endothelial inward-rectifier K+ current by optical isomers of cholesterol,” Biophysical Journal, vol. 83, no. 6, pp. 3211–3222, 2002.View at: Google Scholar
S. Ozaki, D. B. DeWald, J. C. Shope, J. Chen, and G. D. Prestwich, “Intracellular delivery of phosphoinositides and inositol phosphates using polyamine carriers,” Proceedings of the National Academy of Sciences of the United States of America, vol. 97, no. 21, pp. 11286–11291, 2000.View at: Google Scholar
K. Taguchi, H. Kumanogoh, S. Nakamura, and S. Maekawa, “Localization of phospholipase Cβ1 on the detergent-resistant membrane microdomain prepared from the synaptic plasma membrane fraction of rat brain,” Journal of Neuroscience Research, vol. 85, no. 6, pp. 1364–1371, 2007.View at: Publisher Site | Google Scholar
R. G. Cutler, J. Kelly, K. Storie et al., “Involvement of oxidative stress-induced abnormalities in ceramide and cholesterol metabolism in brain aging and Alzheimer's disease,” Proceedings of the National Academy of Sciences of the United States of America, vol. 101, no. 7, pp. 2070–2075, 2004.View at: Publisher Site | Google Scholar