Inhibition of Cholesterol Biosynthesis Reduces γ -Secretase Activity and Amyloid-β Generation

Abstract

Amyloid-β (Aβ) is one of major molecules contributing to the pathogenesis of Alzheimer’s disease (AD). Aβ is derived from amyloid-β protein precursor (AβPP) through sequential cleavages by β- and γ-secretases. Regulation of these components is thought to be an important factor in Aβ generation during the pathogenesis of AD. AβPP, β-secretase, and γ-secretase reside in lipid rafts, where cholesterol regulates the integrity and flexibility of membrane proteins and Aβ is generated. However, the relationship between cholesterol and Aβ generation is controversial. In this study, we aimed to elucidate the direct effects of cholesterol depletion on AβPP processing using AY9944, which blocks the last step of cholesterol biosynthesis and thus minimizes the unknown side effects of upstream inhibitors, such as HMG-CoA reductase inhibitors. Treatment with AY9944 decreased γ-secretase activity and Aβ generation. These results suggested that changes in membrane composition by lowering cholesterol with AY9944 affected γ-secretase activity and Aβ generation, which is associated with AD pathogenesis.

Keywords

Alzheimer’s disease amyloid-β AY9944 cholesterol γ-secretase

INTRODUCTION

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that is defined by the presence of intracellular neurofibrillary tangles and senile plaques in the brain. Amyloid-β (Aβ), the major component of senile plaques, is generated through sequential endoproteolysis of amyloid-β protein precursor (AβPP) [1, 2]. The processing of AβPP through α- and γ-secretase results in the secretion of AβPP (sAβPPα) and p3 peptide, which does not aggregate into plaques. Alternatively, AβPP can be cleaved by β-secretase at the N-terminus of Aβ, resulting in the release of soluble AβPP (sAβPPβ) and AβPP carboxyl terminal fragment (AβPP-CTF; C99). AβPP-CTF is subsequently cleaved by γ-secretase in the transmembrane domain, liberating Aβ into the extracellular space [3, 4]. The senile plaque is composed of these secreted Aβ. Although various genetic mutations have been shown to affect AβPP processing, the formation of senile plaques and the development of dementia are also influenced by some nongenetic factors [5, 6]. Cholesterol homeostasis is a key factor in the genetic and nongenetic risk for AD [7]. In the central nervous system, the ɛ4 allele of the apolipoprotein E (APOE) gene encodes a lipid-binding protein that is critical for cholesterol transport. Expression of the APOE ɛ4 gene is a well-known risk factor for AD and correlates with high serum cholesterol [8, 9]. The nongenetic aspects of cholesterol’s effects on AD development involve AβPP processing in lipid rafts [10, 11]. Lipid rafts are small platforms composed of cholesterol, sphingolipids, phospholipids, and proteins and they play a central role in many cellular processes, including signal transduction, membrane trafficking, and protein processing. In this domain, cholesterol is fixed within the raft structure and assembles the functional raft-associated components [12 –14]. Previous studies have shown that β- and γ-secretase subunits are predominantly localized within lipid rafts [15 –17], whereas α-secretase resides in the non-lipid raft domain [18]. Therefore, production of Aβ occurs exclusively in lipid rafts, and changes in the components of rafts can alter amyloidogenic activity and Aβ generation [19 –21]. In particular, an increased cholesterol level triggers clustering of AβPP with β-secretase in lipid rafts, whereas cholesterol depletion inhibits β-cleavage and Aβ formation in neurons and other cells, simultaneously promoting α-cleavage [22, 23]. Also, cholesterol directly binds to AβPP-CTF and may play a role in Aβ generation through substrate recognition or catalysis [24, 25]. These factors indicate that cholesterol is a relevant in AβPP processing and Aβ generation in AD pathogenesis.

Based on these results, cholesterol-lowering drug treatments, particularly using HMG-CoA reductase inhibitors (statin drugs), have been studied for their potential to mitigate the risk of AD [26 –29]. Clinical trials and prospective cohort studies have produced inconsistent and mixed results; therefore, the effects of these inhibitors remain controversial [30 –32]. In addition, because statins elicit their pharmacological activity far upstream in the cholesterol biosynthesis pathway, they may cause many undesired physiological side effects [33 –35]. In this study, we aimed to identify the effects of cholesterol depletion on AβPP processing using an alternative cholesterol lowering drug, trans-1,4-bis(2-dichlorobenzylamino-ethyl)cyclohexane dihydrochloride, also known as AY9944, which inhibits Δ7-dehydrocholesterol (7-DHC) reductase and reduces cholesterol esterification, thereby blocking cholesterol synthesis [36, 37].

The results of the present study showed that AY9944 decreased Aβ production and γ-secretase activity. Interestingly, we found that the distribution of the membrane composition was altered as the cholesterol content altered, leading to the dissociation of AβPP protein from the rafts after AY9944 treatment. From these results, we concluded that cholesterol biosynthesis by AY9944 disrupted the colocalization of AβPP and γ-secretase in the membrane. These changes consequently rendered these proteins nonfunctional, blocking the generation of Aβ. Our data provide a better understanding of the connections among cholesterol and AβPP processing in AD pathogenesis.

MATERIALS AND METHODS

Cell culture and drug treatment

Chinese hamster ovary (CHO), SH-SY5Y human neuroblastoma, and 7w-PSML cells (stably overexpressing AβPP and PS1 M146L; [38]) were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Hyclone, Irvine, CA) containing 10% fetal bovine serum (Hyclone, Salt Lake City, UT) and 1% penicillin and streptomycin (Sigma, St. Louis, MO). Cells were treated with 1–2μg/mL AY9944 (Santa Cruz, Dallas, TX) for 12 h, 1μM staurosporine (Sigma) for 12 h, and 100μM etoposide (Sigma) for 24 h.

Filipin staining

Cells were seeded on 4-well glass plates and treated with 2μg/mL AY9944 for 12 h. The cells were subsequently rinsed with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde (PFA) for 15 min at room temperature (RT). The cells were then incubated with 50μg/mL filipin solution (Sigma) for 2 h at 37°C and washed with PBS. The distribution of cholesterol in cells was visualized by fluorescence microscopy (Olympus DPS, Center Valley, PA). The fluorescence intensity was measured using the ImageJ sofeware (US National Institutes of Health, Bethesda, MD).

Methylthiazol-2-yl-2,5-diphenyl tetrazolium bromide (MTT) assay

Cells were seeded in 96-well plates and treated with 2μg/mL AY9944 for 12 h. MTT was dissolved in sterile PBS and filtered through a 0.2μm filter. MTT solution was added to the medium followed by incubation for 3 h under dark conditions. The resulting formazan crystals were dissolved in DMSO. The degree of cell survival was determined based on the absorbance at OD 570 mm using a multiplate reader (PowerWave XS, BioTek, Winooski, VT).

TdT-mediated dUTP nick-end labeling (TUNEL) assay

Cells were seeded in poly-L-lysine-coated 6-well plates, and treated with 2μg/mL AY9944. After the 12 h drug treatment, cells were fixed with 4% PFA for 20 min at RT and permeabilized using 0.2% Triton X-100 solution. Detection of apoptotic cells was performed using a DeadEnd Colorimetric TUNEL assay kit (Promega, Madison, WI) with a light microscope (Olympus DPS) following the manufacturer’s protocol.

Luciferase reporter gene assay

Cells were seeded in plates and transfected with the AβPP-C99-GAL4, Notch-ΔE-GVP and UAS-Luc DNA constructs using Lipofectamine LTX (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. After 24 h, drugs were applied as described above for 12 h. The cells were lysed with a 1×passive lysis solution (Promega). When γ-secretase cleaves the AβPP-C99 or Notch-ΔE, the cleaved product AICD-GAL4 or NICD-GVP translocates into the nucleus and activates luciferase gene. Luciferase substrates (Promega) were mixed with the proteins, and luciferase activity was then measured with a luminometer (Infinite M200; TECAN, Switzerland).

Western blotting

Cells were seeded in culture plates, and drugs were applied as described above. After 12 h drug treatment, cells were lysed in RIPA buffer supplemented with a protease and phosphatase inhibitor cocktail (Sigma). For whole cell lysates, cells were sonicated and centrifuged for 20 min at 17,950×g at 4°C and the supernatant was collected. For membrane and cytosol fractionation, washed cells were resuspended in hypotonic buffer (10 mM Tris-HCl, pH 7.4, 1 mM EDTA) and homogenized using Wheaton^® Dounce Tissue Grinders (Fisher scientific, Hampton, NH). After centrifugation at 700 g for 20 min, the post-nuclear supernatant was recentrifuged at 20,000 g for 20 min and the supernatant was collected as the cytosolic extract. Pellets were resuspended in RIPA buffer and centrifuged at 20,000 g for 25 min. The supernatant was collected as a membrane fraction. The resulting cell lysates were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to polyvinylidene difluoride (PVDF) membranes. Primary antibodies were applied in 3% bovine serum albumin (BSA) and incubated overnight at 4°C. Then signal was enhanced using enhanced chemiluminescence (ECL; GE Healthcare Biosciences, Pittsburgh, PA) followed by image analysis with a bioimaging analyzer (LAS-3000; Fuji Film, Inc., Japan) and Multi-Gauge software (Fuji). The following primary antibodies were used: Anti-flotillin (BD Biosciences, San Jose, CA), anti-actin (Sigma), and anti-calnexin (abcam, Cambridge, UK); anti-PS1 loop (Chemicon, Temecula, CA), anti-nicastrin (Abcam and BD Biosciences), anti-APH-1aL (Covance, Berkeley, CA), and anti-PEN-2 (Oncogene, San Diego, CA) to detect the components of the γ-secretase complex; anti-beta amyloid 1–16 (6E10) monoclonal antibodies (SIGNET, Dedham, MA) to detect full-length AβPP; and AB5232 antibody (Chemicon), which recognizes the 762–770 amino acid portion of human AβPP 770 to detect the AβPP- CTF.

Immunocytochemistry

Cells were seeded on 4-well glass plates and treated with 2μg/mL AY9944 for 12 h. The cells were then washed with PBS and fixed with MeOH for 10 min at –20°C. The cells were then incubated with primary antibodies against flotillin (BD), 4G8 (SIGNET), BACE1 (R&D) and nicastrin (Abcam) overnight at 4°C. The cells were then incubated with Alexa 488 and Alexa 594 (Invitrogen) secondary antibodies at RT for 1 h. Finally, the cover slips were mounted on slides, and fluorescence was visualized using a confocal fluorescence microscope (Olympus, FluoView FV10i). FV10-ASW sofrware (Olympus) was used to analyze the obtained images.

Lipid raft isolation

Cells were seeded in 100 mm Culture Dishes and treated with 2μg/mL AY9944 for 12 h. The cells were then resuspended in cell suspension medium (10 mM Tricine, 0.85 % NaCl, 1% Triton X-100, pH 7.4 supplemented with a cocktail of protease inhibitors) and leave on ice. After 30 min, cells were homogenized by Wheaton^® Dounce Tissue Grinders (Fisher scientific) and 22 gauge syringe needle. The homogenate was then mix gently with 2 volume of OptiPreptrademark (Axis– shield, Oslo, Norway) and placed at the bottom of an ultracentrifuge tube (344059, Beckman Instruments, Palo Alto, CA). Density gradient solutions (35%, 30%, 25%, 20%, and 5%) were prepared by dilution of the 40% iodixanol working solution (0.85% NaCl, 30 mM Tricine, pH 7.4. Mix 2 volume of OptiPreptrademark with 1 volume of diluent) with cell suspension medium and carefully layered over samples. The gradients were centrifuged for at 160,000 g for 4 h (SW-41Ti Rotor, Swinging Bucket, Beckman Instruments) and divided into 12 fractions of 1 ml from the top of gradients. Equal volumes of each fraction were subjected to SDS-PAGE, as previously described. Western blotting was performed to check the protein distribution in each fraction.

Aβ enzyme-linked immunosorbent assay (ELISA)

7w-PSML cells were seeded on culture plates and treated with 2μg/mL AY9944 for 24 h. Measurement of Aβ₄₀ and Aβ₄₂ in the culture medium was carried out using a human amyloid immunoassay kit (Biosource, Carlsbad, CA) according to the manufacturer’s instructions.

DNA plasmids and constructs

pECFP-N1 and pEYFP-N1 (BD Clontech) were used as the backbone for fluorescence resonance energy transfer (FRET) analysis. CFP and YFP were mutagenized (Ley221 to Lys221) to generate monomeric CFP and YFP mutants. The C99 site of AβPP L720 was replaced with Pro720 using a QuickChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA). C99 L720P was conjugated with mCFP, and NCT was conjugated with mYFP. AβPP- C99-L720P mutant form, which is resistant to γ-secretase cleavage [39].

Dynamic FRET analysis

Dynamic FRET was performed using an IX81 inverted epifluorescence microscope (Olympus). For dual-emission imaging studies, we used an excitation filter-XF1071 (440AF21), diachronic mirror-XF2034 (455DRLP), and two emission filters (CFP-XF3075 [480AF30] and YFP-XF3079 [535AF26]; Omega Optical, Brattleboro, VT). A Cool SNAP-HQ (Roper Scientific, Canada) cooled CCD camera was used. A 75-W xenon lamp (Olympus) was used as the light source for fluorescent imaging, and a halogen lamp (Olympus) was used for DIC image acquisition. We used 35-mm glass-bottom dishes coated with PDL to culture cells and transfected C99-CFP and NCT-YFP using Lipofectamine LTX (Invitrogen) according to the manufacturer’s protocol. Growth medium was changed to Leibovitz’s L-15 medium (Invitrogen) before images were taken. CFP/YFP and DIC images were recorded every 30 s for 45 min using the dual-image mode. The images were separated into CFP and YFP signals as stack files. MetaMorph software (MetaMorph Inc., Nashville, TN) was used to create ratio images from the raw images of YFP and CFP. All images for each type of measurement (intensity or anisotropy) were pseudo-colored using the same pseudocolor look-up table (LUT). The calibration bar for the LUT is provided on the right of the image. AutoQuant sofrware (Media Cybernetics, Inc., Rockville, MD) was used to analyze the FRET ratio.

Statistical analysis

For statistical analysis, an unpaired t-test or one-way analysis of variance (ANOVA) with a post-hoc Tukey test was performed using Graphpad Instat 5.1 (GraphPad Software Inc., San Diego, CA).

RESULTS

AY9944 inhibited cholesterol biosynthesis without cytotoxicity

To investigate the mechanisms through which AY9944 affected cholesterol biosynthesis, cells were stained with filipin, a fluorescent polyene antibiotic that forms complexes with unesterified sterol [40]. AY9944-treated cells exhibited more filipin-stained areas than vehicle-treated cells (Fig. 1A, B). A filipin-positive signal indicated that AY9944 inhibited cholesterol biosynthesis and caused accumulation of unesterified 7-DHC, a cholesterol precursor. Next, we examined the cytotoxicity of AY9944 using TUNEL assays (Fig. 1C) and MTT assays (Fig. 1D). AY9944 did not induce cell death and did not cause any apparent morphological changes in cells. Staurosporine was used as a positive control for cell death. These data suggested that cholesterol biosynthesis was effectively blocked by AY9944, with no effects on cellular viability.

AY9944 induced AβPP processing and decreased Aβ generation

Next, we analyzed the pathway of AβPP processing in 7w-PSML cells, which stably overexpress AβPP and PS1 M146L. Cells did not exhibit changes in full-length AβPP levels following AY9944 treatment. AβPP-CTF is a substrate for cleavage by γ-secretase to generate Aβ and was found here to accumulate in total cell lysates and membrane fractions (Fig. 2A, B). We measured both Aβ₄₀ and Aβ₄₂ levels using a human β-amyloid (1-40 or 1-42) ELISA kit. AY9944 treatment significantly decreased both Aβ₄₀ and Aβ₄₂ levels in cell culture media (Fig. 2C, D). These data indicated that AY9944 treatment affect AβPP processing.

Changes in the distributions of AβPP and nicastrin following AY9944 treatment

First, we tested the expression of flotillin, a lipid raft marker protein, in AY9944-treated cells; the expression of this target has been shown to decrease in low-cholesterol states [41, 42]. Interestingly, there were no differences in flotillin expression level between treated and untreated cells (Supplementary Figure 1A). To determine whether the decrease in Aβ expression occurred in response to AY9944 treatment, we evaluated the localization of lipid raft, AβPP and nicastrin, a member of the γ-secretase complex, in the cell membrane using immunocytochemistry. We found that the distribution of flotillin was altered after AY9944 treatment compared with that in vehicle-treated cells, not likely flotillin expression. As shown in Figure 3, when AY9944 was treated, scattered lipid raft signals were re-distributed to a single location. AβPP did not localize with the lipid rafts, whereas nicastrin was still associated with the rafts in the presence of AY9944 (Fig. 3A, B). The distribution of the fluorescent intensity is shown in Supplementary Figure 1B and C. We again evaluated the distributions of AβPP-CTF and nicastrin by co-immunocytochemistry. As shown in Figure 3C, these proteins were colocalized in vehicle-treated cells. However, after AY9944 treatment, AβPP-CTF and nicastrin were no longer colocalized (Fig. 3C, D). The distribution of the fluorescent intensity is shown in Figure 3D. We additionally evaluated the cellular localization of AβPP and nicastrin using lipid fractionation assay (Fig. 3E). After AY9944 treatment, distribution of AβPP was changed from fraction 5 (lipid raft fraction as indicated by flottilin) to other fractions, while nicastrin was stayed in the same fractions, indicating that AβPP does not localize with the lipid rafts in the presence of AY9944. We also explored the spatial distribution of BACE1 in the AY9944-treated cells. As previously shown, BACE1 is partially localized in lipid rafts similar to nicastrin. However, in contrast to nicastrin, BACE1 did not undergo significant changes in subcellular distribution following AY9944 treatment (Supplementary Figure 2A, B). Next, FRET was performed to confirm the localization of AβPP and γ-secretase in the presence or absence of AY9944. FRET is powerful approach that detects whether two membrane components are spatially close (<10 nm) [43]. We used AβPP-C99-CFP and nicastrin-YFP as donor and acceptor probes, respectively. We generated a CHO stable cell line overexpressing nicastrin-YFP protein and transfected these cells with AβPP-C99 or AβPP-C99-L720P substitution mutant cDNA. When wild-type AβPP-C99 is sequentially cleaved by γ-secretase, the FRET signal is no longer detected because the C-terminus of the AβPP construct contains the CFP signal. To obtain a continuous CFP signal, despite any interaction between C99 and the γ-secretase, we used the AβPP- C99-L720P mutant form. In contrast with AβPP- C99, γ-secretase did not cleave AβPP-C99-L720P protein. When the cells were stimulated with etoposide (Eto), a DNA damage-inducing agent that activates γ-secretase [44], cells overexpressing AβPP-C99 produced AβPP intracellular domain (AICD), a γ-secretase product from AβPP-C99, but cells expressing AβPP-C99-L720P did not, owing to elimination of the γ-secretase cleavage site (Supplementary Figure 3). As shown in Figure 4A, when AβPP-CTF and nicastrin were in close proximity, a red signal could be observed. After AY9944 treatment, fluorescence values at the membrane were decreased, while those in vehicle cells were unchanged. We quantitatively analyzed the images and confirmed that the distance between the two fluorophores became increasingly larger in AY9944-treated cells (Fig. 4B).

The activity of γ-secretase was downregulated by AY9944

We investigated whether AY9944 affected γ-secretase activity using cells transfected with a GAL4-VP16-tagged AβPP-C99 fragment. In these cells, γ-secretase cleaves AβPP-C99, and the cleaved product, AICD-GAL4-VP16, translocates into the nucleus, binds to the UAS promoter, and activates the downstream luciferase gene. Cells were treated with AY9944 or vehicle for 12 h, and endogenous γ-secretase activity was subsequently measured by luciferase reporter gene assays. Our data showed that γ-secretase activity was decreased by AY9944 (Fig. 4C and Supplementary Figure 4). To examine whether the decrease of γ-secretase activity by AY9944 is not AβPP specific, Notch1, another well known γ-secretase substrate, was used for γ-secretase activity assay. It showed that γ-secretase activity was reduced by AY9944 as AβPP-CTF substrate did (Fig. 4D). The reduction of γ-secretase activity could be explained by several possible mechanisms. The simplest explanation is related to increased enzyme expression. To test this possibility, we examined the expression level of γ-secretase components. However, this analysis showed that the protein levels of APH1, PS1, PEN2, and NCT were not changed by treatment with AY9944 (Fig. 4E).

DISCUSSION

Lipid rafts are plasma membrane microdomains that are enriched with cholesterol, and they act as a molecular platform for proteins involved in Aβ generation [23]. For this reason, several hypotheses regarding the relationship between cholesterol and AD are currently under investigation. In this study, we showed that Aβ generation was downregulated by the Δ7-dehydrocholesterol reductase inhibitor AY9944, which blocks the final step of cholesterol biosynthesis. Interestingly, we found that the dissociation between γ-secretase and AβPP, a substrate of γ-secretase, was observed in AY9944-treated cells. AβPP and nicastrin were associated with lipid rafts under normal conditions, but that AY9944 treatment resulted in changes in the distribution of lipid rafts, with only nicastrin remaining in the rafts. In addition, we used FRET technology to determine the effects of AY9944 on the distance between AβPP-C99 and γ-secretase. Our observations were consistent with previous findings, which showed that AβPP-CTF and γ-secretase are preferentially colocalized in the lipid rafts. After AY9944 treatment, the lipid raft composition was changed, and AβPP-C99 and γ-secretase can no longer interact because of the increase in the physical distance between them. Therefore, we suggest that AY9944 reduces the opportunities for interaction between AβPP-CTF and γ-secretase for proteolysis, leading to decreased levels of Aβ generation.

Our results also showed that γ-secretase activity was downregulated by AY9944. Wahrle and colleagues demonstrated that MβCD, a compound that binds to cholesterol and disrupts microdomains, decreases γ-secretase activity in vitro in buoyant fractions; γ-secretase activity can be restored when cholesterol is replenished [16, 45]. The addition of soluble cholesterol to purified γ-secretase in model membranes significantly increases the cleavage rate of AβPP and enhances the cleavage of C99 in lipid vesicles [46]. In addition, Wada and colleagues characterized the γ-secretase activity of low-density membrane domains in a cell-free system and showed that Aβ production in low-density membrane fractions is inhibited by specific γ-secretase inhibitors [47]. Although these reports have shown a possible relationship between cholesterol and γ-secretase activity, the specific mechanism needed to be studied in a whole-cell system because these studies used cell-free systems. Our study confirmed the relationship between cholesterol and γ-secretase in intact cells. Although AY9944 decreased γ-secretase activity, there were no significant changes in the protein levels of γ-secretase complex components, indicating that AY9944 modulated the enzymatic activity of the γ-secretase complex, rather than altering the protein expression of each component. The possible mechanisms underlying the modulation of γ-secretase activity by AY9944 were further examined.

Numerous studies have shown a relationship between cholesterol levels and the risk of AD [21, 48]. This relationship has prompted researchers to analyze the effects of cholesterol-lowering drugs, such as statins, on AD. Statins are well-tolerated drugs [49] and could be considered as a preventive treatment for AD [32, 50]. The biosynthesis of cholesterol is a complex process involving more than 20 steps through various intermediates [51]. Statins block the first step in the cholesterol biosynthesis pathway. Thus, all intermediates below HMG-CoA, including mevalonate, squalene, lanosterol, and ultimately cholesterol, are affected. These intermediates also serve as various precursors for the synthesis of vitamin D, coenzyme Q10, heme A, farnesyl and geranylgeranyl moieties, dolichol, ubiquinone, oxysterols, and sterols [52 –55]. Therefore, an imbalance in the intermediates of cholesterol biosynthesis will have severe consequences for protein modification, lipid metabolism, steroid hormones formation, spermatogenesis, and immune function [56, 57]. Accordingly, in 2012, the United States Food and Drug Administration (FDA) issued a warning regarding the potential adverse effects of statins [58], including liver injury, memory loss, diabetes, and muscle damage [34, 35]. Thus, cholesterol biosynthesis should be targeted further downstream in the reaction sequence in order to prevent unknown side effects. Because AY9944 inhibits the last step of cholesterol biosynthesis, thus minimizing the negative effects of the intermediates of cholesterol biosynthesis.

Although statins are effective cholesterol-lowering drugs and Aβ generation is decreased by cholesterol lowering, Aβ generation could be affected by other pathways possibly causing unknown final outcomes. Because of the potentially compounding effects of statin treatment, interpretation of the statin effect is unclear. Reduction of Aβ generation by statins is not only due to blockage of cholesterol biosynthesis but also includes isoprenylation and dysregulation of other signaling pathways that are known to involve AβPP processing [59]. In this study, we identified the acute effects of cholesterol on Aβ generation using AY9944 and found a significant decrease in Aβ production. This result has important implications in the relationships among cholesterol, AβPP processing, and AD pathogenesis. However, AY9944 has also been shown to elicit major adverse effects in vivo, such as seizures and Smith-Lemli-Opitz syndrome [60]. Therefore, AY9944 must be used appropriately for clinical treatment or a modified version of a Δ7-dehydrocholesterol reductase inhibitor is needed.

In conclusion, we found that accumulation of immature cholesterol fallowing AY9944 treatment affected the physical properties of lipid rafts, leading to the dissociation of AβPP-derived proteolytic fragments from the raft domain. This dissociation may inhibit lateral interactions between AβPP-CTF and γ-secretase. We also found that the AY9944-induced increase in immature cholesterol did not enhance γ-secretase activity. Taken together, our work provides the first evidence to explain how AY9944 may decrease Aβ generation through physical and enzymatic mechanisms. Therefore, compounds that prevent complex formation between cholesterol and AβPP-CTF or γ-secretase may represent potential therapeutic agents for prevention or treatment of AD.

Footnotes

ACKNOWLEDGMENTS

This work was supported by grants from the NRF (2015R1A2A1A05001794, 2014M3C7A1046047, 2015M3C7A1028790) and MRC (2012R1A5A2A 44671346) for I.M.-J.

Authors’ disclosurse available online ().

Appendix

References

Spires-Jones

, Hyman

(2014) The intersection of amyloid beta and tau at synapses in Alzheimer’s disease. Neuron 82, 756–771.

Bloom

(2014) Amyloid-beta and tau: The trigger and bullet in Alzheimer disease pathogenesis. JAMA Neurol 71, 505–508.

O’Brien

, Wong

(2011) Amyloid precursor protein processing and Alzheimer’s disease. Annu Rev Neurosci 34, 185–204.

Mucke

, Selkoe

(2012) Neurotoxicity of amyloid beta-protein: Synatic and network dysfunction. Cold Spring Harb Perspect Med 2, a006338.

Karch

, Cruchaga

, Goate

(2014) Alzheimer’s disease genetics: From the bench to the clinic. Neuron 83, 11–26.

Guerreiro

, Hardy

(2014) Genetics of Alzheimer’s disease. Neurotherapeutics 11, 732–737.

Gamba

, Testa

, Sottero

, Gargiulo

, Poli

, Leonarduzzi

(2012) The link between altered cholesterol metabolism and Alzheimer’s disease. Ann N Y Acad Sci 1259, 54–64.

Notkola

, Sulkava

, Pekkanen

, Erkinjuntti

, Ehnholm

, Kivinen

, Tuomilehto

, Nissinen

(1998) Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer’s disease. Neuroepidemiology 17, 14–20.

Refolo

, Malester

, LaFrancois

, Bryant-Thomas

, Wang

, Tint

, Sambamurti

, Duff

, Pappolla

(2000) Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model. Neurobiol Dis 7, 321–331.

10.

Ehehalt

, Keller

, Haass

, Thiele

, Simons

(2003) Amyloidogenic processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts. J Cell Biol 160, 113–123.

11.

Maulik

, Westaway

, Jhamandas

, Kar

(2013) Role of cholesterol in APP metabolism and its significance in Alzheimer’s disease pathogenesis. Mol Neurobiol 47, 37–63.

12.

Simons

, Ikonen

(1997) Functional rafts in cell membranes. Nature 387, 569–572.

13.

Nicolson

(2014) The Fluid-Mosaic Model of Membrane Structure: Still relevant to understanding the structure, function and dynamics of biological membranes after more than 40 years. Biochim Biophys Acta 1838, 1451–1466.

14.

Sonnino

, Prinetti

(2013) Membrane domains and the “lipid raft” concept. Curr Med Chem 20, 4–21.

15.

Hattori

, Asai

, Onishi

, Sasagawa

, Hashimoto

, Saido

, Maruyama

, Mizutani

, Ishiura

(2006) BACE1 interacts with lipid raft proteins. J Neurosci Res 84, 912–917.

16.

Wahrle

, Das

, Nyborg

, McLendon

, Shoji

, Kawarabayashi

, Younkin

, Golde

(2002) Cholesterol-dependent gamma-secretase activity in buoyant cholesterol-rich membrane microdomains. Neurobiol Dis 9, 11–23.

17.

Hur

, Welander

, Behbahani

, Aoki

, Franberg

, Winblad

, Frykman

, Tjernberg

(2008) Active gamma-secretase is localized to detergent-resistant membranes in human brain. FEBS J 275, 1174–1187.

18.

Harris

, Pereira

, Parkin

(2009) Targeting ADAM10 to lipid rafts in neuroblastoma SH-SY5Y cells impairs amyloidogenic processing of the amyloid precursor protein. Brain Res 1296, 203–215.

19.

Lee

, Liyanage

, Bickel

, Xia

, Lansbury

Jr , Kosik

(1998) A detergent-insoluble membrane compartment contains A beta in vivo. Nat Med 4, 730–734.

20.

Vetrivel

, Thinakaran

(2010) Membrane rafts in Alzheimer’s disease beta-amyloid production. Biochim Biophys Acta 1801, 860–867.

21.

Di Paolo

, Kim

(2011) Linking lipids to Alzheimer’s disease: Cholesterol and beyond. Nat Rev Neurosci 12, 284–296.

22.

Marquer

, Devauges

, Cossec

, Liot

, Lecart

, Saudou

, Duyckaerts

, Leveque-Fort

, Potier

(2011) Local cholesterol increase triggers amyloid precursor protein-Bace1 clustering in lipid rafts and rapid endocytosis. FASEB J 25, 1295–1305.

23.

Simons

, Ehehalt

(2002) Cholesterol, lipid rafts, and disease. J Clin Invest 110, 597–603.

24.

Beel

, Sakakura

, Barrett

, Sanders

(2010) Direct binding of cholesterol to the amyloid precursor protein: An imortant interaction in lipid-Alzheimer’s disease relationships? Biochim Biophys Acta 1801, 975–982.

25.

Barrett

, Song

, Van Horn

, Hustedt

, Schafer

, Hadziselimovic

, Beel

, Sanders

(2012) The amyloid precursor protein has a flexible transmembrane domain and binds cholesterol. Science 336, 1168–1171.

26.

Fassbender

, Simons

, Bergmann

, Stroick

, Lutjohann

, Keller

, Runz

, Kuhl

, Bertsch

, von Bergmann

, Hennerici

, Beyreuther

, Hartmann

(2001) Simvastatin strongly reduces levels of Alzheimer’s disease beta -amyloid peptides Abeta 42 and Abeta 40 in vitro and in vivo. Proc Natl Acad Sci U S A 98, 5856–5861.

27.

Jick

, Zornberg

, Jick

, Seshadri

, Drachman

(2000) Statins and the risk of dementia. Lancet 356, 1627–1631.

28.

Wolozin

, Kellman

, Ruosseau

, Celesia

, Siegel

(2000) Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors. Arch Neurol 57, 1439–1443.

29.

Silva

, Teixeira

, Remiao

, Borges

(2013) Alzheimer’s disease, cholesterol, and statins: The junctions of important metabolic pathways. Angew Chem Int Ed Engl 52, 1110–1121.

30.

Sanchez-Ferro

, Benito-Leon

, Mitchell

, Bermejo-Pareja

(2013) A review of the potential therapeutic role of statins in the treatment of Alzheimer’s disease: Current research and opinion. Neuropsychiatr Dis Treat 9, 55–63.

31.

Fonseca

, Resende

, Oliveira

, Pereira

(2010) Cholesterol and statins in Alzheimer’s disease: Current controversies. Exp Neurol 223, 282–293.

32.

Wong

, Lin

, Boudreau

, Devine

(2013) Statins in the prevention of dementia and Alzheimer’s disease: A meta-analysis of observational studies and an assessment of confounding. Pharmacoepidemiol Drug Saf 22, 345–358.

33.

Golomb

, Evans

(2008) Statin adverse effects: A review of the literature and evidence for a mitochondrial mechanism. Am J Cardiovasc Drugs 8, 373–418.

34.

Fitchett

, Hegele

, Verma

(2015) Cardiology patient page. Statin intolerance. Circulation 131, e389–e391.

35.

Dimmitt

, Stampfer

, Warren

(2015) Common statin side effects explain poor compliance. Br J Clin Pharmacol 80, 170–171.

36.

Kraml

, Bagli

, Dvornik

(1964) Inhibition of the conversion of 7-dehydrocholesterol to cholesterol by AY-9944. Biochem Biophys Res Commun 15, 455–457.

37.

Horlick

(1966) Effect of a new inhibitor of cholesterol biosynthesis (AY 9944) on serum and tissue sterols in the rat. J Lipid Res 7, 116–121.

38.

Xia

, Zhang

, Kholodenko

, Citron

, Podlisny

, Teplow

, Haass

, Seubert

, Koo

, Selkoe

(1997) Enhanced production and oligomerization of the 42-residue amyloid beta-protein by Chinese hamster ovary cells stably expressing mutant presenilins. J Biol Chem 272, 7977–7982.

39.

Tesco

, Ginestroni

, Hiltunen

, Kim

, Dolios

, Hyman

, Wang

, Berezovska

, Tanzi

(2005) APP substitutions V715F and L720P alter PS1 conformation and differentially affect Abeta and AICD generation. J Neurochem 95, 446–456.

40.

Yoshikawa

(1991) Effects of drugs on cholesterol esterification in normal and Niemann-Pick type C fibroblasts: AY-9944, other cationic amphiphilic drugs and DMSO. Brain Dev 13, 115–120.

41.

Browman

, Hoegg

, Robbins

(2007) The SPFH domain-containing proteins: More than lipid raft markers. Trends Cell Biol 17, 394–402.

42.

Kirsch

, Eckert

, Mueller

(2003) Statin effects on cholesterol micro-domains in brain plasma membranes. Biochem Pharmacol 65, 843–856.

43.

Jares-Erijman

, Jovin

(2003) FRET imaging. Nat Biotechnol 21, 1387–1395.

44.

Jin

, Cho

, Jung

, Shim

, Mook-Jung

(2008) DNA damage-inducing agents elicit gamma-secretase activation mediated by oxidative stress. Cell Death Differ 15, 1375–1384.

45.

Yancey

, Rodrigueza

, Kilsdonk

, Stoudt

, Johnson

, Phillips

, Rothblat

(1996) Cellular cholesterol efflux mediated by cyclodextrins. Demonstration Of kinetic pools and mechanism of efflux. J Biol Chem 271, 16026–16034.

46.

Osenkowski

, Ye

, Wang

, Wolfe

, Selkoe

(2008) Direct and potent regulation of gamma-secretase by its lipid microenvironment. J Biol Chem 283, 22529–22540.

47.

Wada

, Morishima-Kawashima

, Qi

, Misono

, Shimada

, Ohno-Iwashita

, Ihara

(2003) Gamma-secretase activity is present in rafts but is not cholesterol-dependent. Biochemistry 42, 13977–13986.

48.

Wood

, Igbavboa

, Eckert

, Johnson-Anuna

, Muller

(2005) Is hypercholesterolemia a risk factor for Alzheimer’s disease? Mol Neurobiol 31, 185–192.

49.

Davidson

(2001) Safety profiles for the HMG-CoA reductase inhibitors: Treatment and trust. Drugs 61, 197–206.

50.

Petanceska

, DeRosa

, Olm

, Diaz

, Sharma

, Thomas-Bryant

, Duff

, Pappolla

, Refolo

(2002) Statin therapy for Alzheimer’s disease: Will it work? J Mol Neurosci 19, 155–161.

51.

Kandutsch

, Russell

(1960) Preputial gland tumor sterols. 3. A metabolic pathway from lanosterol to cholesterol. J Biol Chem 235, 2256–2261.

52.

Ikonen

(2006) Mechanisms for cellular cholesterol transport: Defects and human disease. Physiol Rev 86, 1237–1261.

53.

Miller

(1987) Structure of genes encoding steroidogenic enzymes. J Steroid Biochem 27, 759–766.

54.

Goldstein

, Brown

(1990) Regulation of the mevalonate pathway. Nature 343, 425–430.

55.

Waterham

(2006) Defects of cholesterol biosynthesis. FEBS Lett 580, 5442–5449.

56.

Holstein

, Hohl

(2004) Isoprenoids: Remarkable diversity of form and function. Lipids 39, 293–309.

57.

Spann

, Garmire

, McDonald

, Myers

, Milne

, Shibata

, Reichart

, Fox

, Shaked

, Heudobler

, Raetz

, Wang

, Kelly

, Sullards

, Murphy

, Merrill

Jr , Brown

, Dennis

, Li

, Ley

, Tsimikas

, Fahy

, Subramaniam

, Quehenberger

, Russell

, Glass

(2012) Regulated accumulation of desmosterol integrates macrophage lipid metabolism and inflammatory responses. Cell 151, 138–152.

58.

Food

, Drug, Administration (2012) FDA Expands Advice on Statin Risks. Consumer Updates. http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm293330.htm updated January 31, 2014, Accessed December, 13, 2015.

59.

Ostrowski

, Wilkinson

, Golde

, Landreth

(2007) Statins reduce amyloid-beta production through inhibition of protein isoprenylation. J Biol Chem 282, 26832–26844.

60.

Irons

, Elias

, Salen

, Tint

, Batta

(1993) Defective cholesterol biosynthesis in Smith-Lemli-Opitz syndrome. Lancet 341, 1414.