• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of fasebjClick here to go to the full-featured web site.Click here to learn how to submit an article to The FASEB Journal.Never miss an issue. Click here to subscribe to The FASEB Journal s full content.Click here for the latest news from The FASEB Journal.Find out what s being published every day. Click here to subscribe to RSS feeds.Click here to visit The FASEB Journal's full-featured site.
FASEB J. Nov 2009; 23(11): 3819–3828.
PMCID: PMC2775008

Inhibition of acyl-coenzyme A: cholesterol acyl transferase modulates amyloid precursor protein trafficking in the early secretory pathway


Amyloid β-peptide (Aβ) has a central role in the pathogenesis of Alzheimer’s disease (AD). Cellular cholesterol homeostasis regulates endoproteolytic generation of Aβ from the amyloid precursor protein (APP). Previous studies have identified acyl-coenzyme A: cholesterol acyltransferase (ACAT), an enzyme that regulates subcellular cholesterol distribution, as a potential therapeutic target for AD. Inhibition of ACAT activity decreases Aβ generation in cell- and animal-based models of AD through an unknown mechanism. Here we show that ACAT inhibition retains a fraction of APP molecules in the early secretory pathway, limiting the availability of APP for secretase-mediated proteolytic processing. ACAT inhibitors delayed the trafficking of immature APP molecules from the endoplasmic reticulum (ER) as shown by metabolic labeling and live-cell imaging. This resulted in partial ER retention of APP and enhanced ER-associated degradation of APP by the proteasome, without activation of the unfolded protein response pathway. The ratio of mature APP to immature APP was reduced in brains of mice treated with ACAT inhibitors, and strongly correlated with reduced brain APP-C99 and cerebrospinal fluid Aβ levels in individual animals. Our results identify a novel ACAT-dependent mechanism that regulates secretory trafficking of APP, likely contributing to decreased Aβ generation in vivo.—Huttunen, H. J., Peach, C., Bhattacharyya, R., Barren, C., Pettingell, W., Hutter-Paier, B., Windisch, M., Berezovska, O., Kovacs, D. M. Inhibition of acyl-coenzyme A: cholesterol acyl transferase modulates amyloid precursor protein trafficking in the early secretory pathway.

Keywords: Alzheimer’s disease, cholesterol, endoplasmic reticulum, lipid, protein maturation

Amyloid precursor protein (APP) is a ubiquitously expressed type 1 glycoprotein involved in cell adhesion, migration, and synaptogenesis (1,2,3). Even modest increases in APP expression levels, due to APP gene duplication or promoter mutations, have profound effects on Alzheimer’s disease (AD) pathogenesis (4,5,6,7). Like most plasma membrane proteins, APP is synthesized and N-glycosylated in the endoplasmic reticulum (ER), from where the folded, immature APP moves to the Golgi complex for maturation (O-glycosylation) before transport to the cell surface. The two forms of APP holoprotein, immature N-glycosylated APP (APP-im) and mature APP (APP-m) carrying terminally modified N- and O-linked glycans, can be distinguished by their different electrophoretic mobilities (8) (see also (Fig. 1A). For Aβ biogenesis, APP is first cleaved by β-site APP cleaving enzyme 1 (BACE1; β-cleavage), producing a 99-aa C-terminal fragment (APP-C99). APP can also be cleaved in the middle of the Aβ sequence (nonamyloidogenic α-cleavage by cell-surface metalloproteases ADAM10 and ADAM17), generating an 83-aa C-terminal fragment of APP (APP-C83). Both APP-C99 and APP-83 can be cleaved by presenilin-dependent γ-secretase, releasing the APP intracellular domain and either Aβ peptide (from C99) or p3 peptide (from C83) (9, 10). Most Aβ generation is thought to occur in post-Golgi compartments of the cell, on the plasma membrane, or within the endocytic machinery (11,12,13,14). Consequently, altered trafficking of APP along the secretory pathway directly affects the availability of mature APP for Aβ generation (15,16,17,18).

Figure 1.
ACAT inhibition slows maturation of APP holoprotein. A) CHO cells expressing human APP751 (CHO/APP) were treated with increasing concentrations of ACAT inhibitor CI-1011 for 4 d. Cell extracts were resolved on SDS-PAGE gel and probed with C-terminal APP ...

The role of cholesterol as a risk factor for AD is supported by a long line of studies (reviewed in refs. 19,20,21). Our previous studies have shown that inhibition of ACAT activity potently reduces Aβ generation while reducing the levels of both α- and β-secretase-generated C-terminal fragments, APP-C83 and APP-C99 (22,23,24). Notably, ACAT inhibitor CP-113,818 strongly protected from development of amyloid pathology correlating with improved cognitive capacity in a transgenic mouse model of AD (22). Moreover, an ACAT inhibitor suitable for human use, CI-1011, was able to reduce diffuse amyloid pathology also in the brains of aged mice that displayed strong AD-like pathology before the treatment was started (unpublished results). Because ACAT inhibitors comprise a promising therapeutic modality for lowering brain Aβ levels, understanding the molecular events linking ACAT inhibition to reduced Aβ generation is essential for further clinical development of ACAT inhibitors for AD. Furthermore, mechanistic insights into the complex relationship of cholesterol, APP, and Aβ (19,20,21, 25) may eventually provide entirely novel therapeutic approaches for treatment of AD.

Here we present evidence suggesting that inhibition of ACAT activity targets nascent APP molecules in the early secretory pathway. This leads to delayed maturation of newly synthesized APP, limiting the availability of APP for Aβ generation at or near the cell surface. Consistent with our cell-based data, we found correlations between reduced maturation of APP and β-cleavage products of APP in the brains and CSF of ACAT-inhibitor-treated hAPP transgenic mice. We conclude that ACAT activity regulates APP trafficking in the early secretory pathway and consequently the availability of APP for Aβ generation.


Cell culture and transfection

Parental CHO, H4, and mouse embryonic fibroblasts (MEFs) were grown in standard conditions (26). Parental CHO cells were transfected with a construct encoding APP695-paGFP (18) using Effectene transfection reagent (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. Stable cell lines (CHO/APP751 and CHO/APP695-paGFP) were selected and maintained in G418 (Calbiochem, San Diego, CA, USA).

Antibodies and Western blots

The following antibodies were used: APP (A8717, C-terminal) and β-tubulin were from Sigma (St Louis, MO, USA); V5 from Invitrogen (Carlsbad, CA, USA); BiP/GRP78, transferrin receptor, and CHOP/GADD153 from Affinity BioReagents (Golden, CO, USA); Nicastrin and GAPDH from Chemicon/Millipore (Temecula, CA, USA); and GM130 from BD Biosciences (San Jose, CA, USA). Cell extraction and Western blotting was performed as described previously (26). Band intensities on Western blot images were quantitated by using Quantity One software (Bio-Rad, Hercules, CA, USA) and normalized as indicated in figure legends.

Pulse-chase and cycloheximide (CHX) block assay

Semiconfluent cells treated with 10 μM CP-113,818 for 4 d on 100-mm plates were first preincubated in methionine/cysteine-free medium for 1 h. Then 100 μCi of [35S]-methionine/cysteine (MP Biomedicals, Solon, OH, USA) was added per plate for 15 min (pulse). Cells were incubated in the presence of excess amounts of cold methionine/cysteine (MP Biomedicals) and harvested at 40-min intervals (chase). The cells were then washed with cold PBS and lysed in Triton-Nonidet P-40 buffer. For immunoprecipitation, 250–500 μg of total protein was immunoprecipitated with excess amounts of either APP (A8717) or transferrin receptor antibodies. Immunocomplexes were captured with protein G agarose beads (Pierce, Rockford, IL, USA), washed 4 times with the extraction buffer, and heated for 10 min at 70°C in 1× LDS gel-loading buffer (Invitrogen) containing β-mercaptoethanol. Next, 4–12% gradient Bis-Tris gels (Novex/Invitrogen) used to resolve the samples were fixed, dried, and exposed to a phosphorimaging screen (Bio-Rad). Images were read and quantitated using a Personal Molecular Imager FX and Quantity One software (Bio-Rad).

For CHX block assay, CHO/APP cells were pretreated for 4 d with 10 μM CP-113,818. At the end of the treatment, 10 μg/ml CHX was added to block protein synthesis and allow degradation of APP in the cells (APP t1/2 is <60 min in most cells). After 6 h, CHX was removed to allow synthesis of new proteins. CP-113,818 or vehicle was present at all times. Cells were harvested at 15-min intervals to analyze APP content. Immature and mature forms of APP were detected in Western blots with A8717 antibody (Sigma).

Subcellular fractionation and retrotranslocation assay

Postnuclear supernatants (1500 g) of parental CHO cells treated with 10 μM CP-113,818 for 4 d were fractionated on a 7.5–30% continuous iodixanol gradients (OptiPrep; Axis-Shield, Norton, MA, USA) as described previously (24, 26).

Retrotranslocation of proteins to the cytosol was analyzed as described previously (26, 27). Where indicated, cells were pretreated with 1 μM epoxomicin (Biomol, Plymouth Meeting, PA, USA) for 5 h.

Live-cell imaging

Stably transfected CHO/APP-paGFP cells plated on 35-mm glass-bottom dishes (MatTek Cultureware; Mat Tek Corp., Ashland, MA, USA) were pretreated with 10 μM CP-113,818 or vehicle for 3 d. On the third day, the cells were transfected with a plasmid encoding DsRed2-ER (Clontech, Palo Alto, CA, USA), a red fluorescent protein DsRed2 fused to a signal peptide at the N terminus and an ER retention signal KDEL at the C-terminus. The cells were incubated in the presence of CP-113,818 or vehicle for an additional 20 h. Before imaging, cells were washed once and transferred to Opti-MEM medium lacking Phenol Red (Life Technologies/Invitrogen) with CP-113,818 or vehicle. Live-cell imaging studies were performed in a heated 37°C, 5% CO2 live-cell chamber attached to a Zeiss LSM510 inverted confocal microscope equipped with a krypton-argon laser (488 nm, 514 nm, and 543 nm lines) for confocal microscopy (Carl Zeiss, Oberkochen, Germany), and a 720- to 950-nm tunable, pulsed Ti:sapphire laser (Chameleon; Coherent Inc., Santa Clara, CA, USA) for 2-photon live-cell imaging. ER within transfected cells was identified by DsRed2-ER signal, and selected ER regions were photoactivated by a 750-nm laser pulse (5% transmission) for 5 s. Stacks of z sections of paGFP (excitation 488 nm, emission 500–550 nm BP) and DsRed2 (excitation 543 nm, emission 560 nm LP) were sequentially acquired before photoactivation and at 2-min intervals after photoactivation for up to 20 min. Projections of z stacks were analyzed for mean fluorescence intensity within the selected, photoactivated regions using the histogram and analysis tools of Adobe Photoshop CS (Adobe Systems, San Jose, CA, USA). Signal in the paGFP channel before photoactivation was used as a background and subtracted from final values. Sixteen cells for both CP-113,818 and vehicle groups were imaged and quantitated in 4 independent experiments.

Mice, treatments, and tissue sampling

hAPP transgenic mice overexpress human APP751 with the London (V717I) and Swedish (K670M/N671L) mutations under the regulatory control of the neuron-specific murine Thy-1 promoter (mThy-1-hAPP751; heterozygous with respect to the transgene, on a C57BL/6 F3 background) (28). Mice were handled and treated as described previously (ref. 22 and unpublished results). CSF was obtained from anesthesized mice during and after the treatment. Animals were sacrificed on d 56 of treatment. Brains were collected and divided along the sagittal plane, and hemispheres were frozen in liquid nitrogen.

Aβ determinations

CSF Aβ1-40 and Aβ1-42 were analyzed using commercially available ELISA kits (Genetics Company, Zurich, Switzerland). Total CSF Aβ values (the sum of Aβ1-40 and Aβ1-42) were used in correlation analysis.

Statistical analysis

Statistical analyses were performed using Student’s t test. For correlation analysis, Pearson coefficients were calculated. All statistical analyses were performed using Microsoft Excel (Microsoft, Redmond, WA, USA) and GraphPad Prism software (GraphPad, San Diego, CA, USA). Significance was placed at P < 0.05.


Effects of ACAT inhibition on maturation of APP holoprotein

To begin characterization of the molecular mechanisms by which ACAT inhibition modulates APP metabolism, we have treated various cell lines expressing human APP with two structurally different ACAT inhibitors, CP-113,818 and CI-1011 (29). Cells were treated with micromolar concentrations of ACAT inhibitors for up to 4 d to allow subcellular cholesterol distribution to reach an equilibrium (24). As shown in Fig. 1A, treatment of CHO/APP cells with the ACAT inhibitor CI-1011 dose-dependently reduced the steady-state level of mature APP (APP-m) while reducing the levels of APP-C99 and APP-C83. Because the immature APP (APP-im) levels were affected to a much lesser extent, the ratio of APP-m/APP-im was reduced by up to 53% in this APP overexpressing cell line. We also tested naive cell lines to study the effect of ACAT inhibition on proteolytic processing of endogenous APP. In human H4 neuroglioma cells and MEFs, similar effects were observed, suggesting that ACAT inhibitors modulate the maturation process of both endogenous and overexpressed APP holoprotein in various cell types (Fig. 1B).

APP holoprotein has a short turnover rate in most cell types (t1/2<60 min) (8). Next, we used a CHX block-release assay to study the effect of ACAT inhibitor CP-113,818 on APP maturation kinetics. First, we treated CHO/APP cells with CP-113,818 for 4 d, and for the last 6 h with CHX to block the synthesis of new proteins, allowing nearly complete degradation of short-lived proteins such as APP. After 6 h, CHX was removed, and maturation of APP was followed at 15-min intervals for up to 60 min in the presence of vehicle or CP-113,818. As shown in Fig. 1C, appearance of mature APP is delayed in CP-113,818-treated cells (e.g., compare APP-m levels between vehicle- and CP-113,818-treated cells at 45- and 60-min time points in Fig. 1C). Quantitation of these data showed that at 60 min after CHX removal, there is 30% (n=3, P=0.0103) less mature APP present in ACAT inhibitor-treated cells. These data suggest that ACAT inhibition may target newly synthesized APP molecules early in the secretory pathway, likely before they get access to the medial-Golgi complex, where enzymes responsible for O-glycosylation are located (30). This is consistent with the observed reduction of both α- and β-cleavage products in ACAT-inhibitor treated cells, because most α-secretase cleavage occurs in the trans-Golgi network and plasma membrane, whereas β-secretase cleavage occurs mostly in the endosomes (31,32,33,34,35).

ACAT inhibition slows trafficking of APP in the early secretory pathway

To characterize the effects of ACAT inhibition on early secretory pathway trafficking of APP in more detail, we used metabolic labeling and live-cell imaging. CHO/APP cells treated with CP-113,818 were pulse-labeled with [35S]-methionine/cysteine for 15 min, then chased at 40-min intervals for up to 160 min. Labeled APP was then immunoprecipitated from cell lysates and detected by fluorography. In vehicle-treated cells, the level of mature APP was highest at 40 min, whereas in CP-113,818-treated cells the appearance of APP-m was delayed, peaking at 80 min (Fig. 2A). The total amount of APP-im converted to APP-m throughout the 160-min chase period was also reduced by CP-113,818 treatment, consistent with decreased steady-state levels of APP-m (Fig. 1). Notably, when transferrin receptor (TfR), a cell surface protein with a single transmembrane domain, was isolated in a secondary round of immunoprecipitation, CP-113,818 had little effect on TfR levels or maturation (Fig. 2A). The effect on APP maturation is quantitated in a plot showing the rate and amount of APP maturation in vehicle- and CP-113,818-treated cells as percentage change of 0-min chase control samples (Fig. 2B). Interestingly, the delayed and decreased maturation of APP was accompanied by stabilization of APP-im, suggesting retention of APP-im in the early secretory pathway (Fig. 2A). Indeed, determination of the half-life of APP-im in vehicle- and CP-113,818-treated cells by logarithmic fitting showed that ACAT inhibition prolonged t1/2 of APP-im from 44 to 77 min (Fig. 2C).

Figure 2.
APP trafficking is altered in the early secretory pathway by ACAT inhibition. A) APP maturation was analyzed by metabolic labeling in CHO/APP cells. Cells were pretreated with CP-113,818 or vehicle for 4 d, followed by a 15-min pulse labeling with [ ...

The prolonged half-life of immature APP and the delayed appearance of mature APP in ACAT inhibitor-treated cells suggests that APP may be retained in the early compartments of the secretory pathway. Next, we investigated the subcellular distribution of APP using OptiPrep fractionation. Naive CHO cells were treated with CP-113,818 for 4 d and then subjected fractionation in 7.5–30% OptiPrep gradients. Mature APP codistributed with GM130 to fractions 2–4, suggesting that most of the Golgi-complex proteins are found in these fractions (Fig. 3A). Immature APP was localized to fractions 5–10 together with GRP78/BiP, a typical ER protein. CP-113,818 had no effect on the distribution of mature APP (Fig. 3A, B). On the contrary, the distribution of immature APP was significantly altered in CP-113,818-treated cells (Fig. 3A, B). In particular, fraction 5 accumulated a high amount of immature APP. Interestingly, the GRP78/BiP-containing ER membranes from CP-113,818-treated cells distributed in a more compact manner across the gradient and had a lighter buoyant density as compared to control cells (Fig. 3A). These data suggest that ACAT inhibition specifically targets the immature APP molecules in the early secretory pathway.

Figure 3.
Effect of ACAT inhibition of the distribution of APP in OptiPrep gradients. A) Naive CHO cells were treated for 4 d with 10 μM CP-113,818. Postnuclear homogenates were fractionated in 7.5–30% OptiPrep gradients. Fractions were analyzed ...

To analyze the effect of ACAT inhibition on the dynamics of early secretory pathway trafficking of APP, we used a live-cell-imaging technique utilizing a photoactivatable variant of green fluorescent protein (paGFP) (36). CHO cells stably expressing a photoactivatable APP-paGFP fusion protein (18) were first treated with CP-113,818 for 4 d. Twenty hours before imaging, the cells were transiently transfected with an ER marker, DsRed2-ER, a constitutively fluorescent protein localized to the ER. Using the DsRed2-ER signal, ER was localized within the cells, followed by photoactivation of selected areas of the ER. Trafficking of photoactivated APP-paGFP out of the ER was then followed for up to 20 min (Fig. 4A). Because the activated state of paGFP remains stable for days (36), reduced paGFP fluorescence within the activation area directly reflects trafficking of APP-paGFP out of this area. In vehicle-treated cells, the overall signal from photoactivated APP-paGFP within the activation area was reduced by 46% (n=16, P<0.0001 vs. 0 min) after 20 min. In contrast, the APP-paGFP signal was reduced by only 24% (n=16, P<0.0001 vs. 0 min; P<0.0001 vs. vehicle at 20 min) in cells treated with CP-113,818 (Fig. 4A, B). Intracellular retention half-times can be used to describe kinetic aspects of protein trafficking (30). We used normalized fluorescence values to calculate ER retention half-time for APP-paGFP. In vehicle-treated cells, APP-paGFP exited the ER with a retention time of 24 min, whereas in CP-113,818-treated cells, the ER retention time was significantly longer, 55 min (Fig. 4C). The change caused by CP-113,818 treatment is comparable to the turnover rate of APP-im determined by metabolic labeling (Fig. 2C). These data also suggest that, under normal conditions, roughly half (55%) of the time (24 min, Fig. 4C) it takes to convert a newly synthesized, immature APP molecule to a mature APP molecule (44 min, Fig. 2C) is spent in the ER. In comparison, CP-113,818 treatment increases the time APP spends in the ER to ~71% of the total half-life of APP-im. The time between ER exit and conversion of APP-im to APP-m was not significantly different in vehicle- vs. CP-113,818-treated cells (19 vs. 22 min). Thus, together with the metabolic labeling data, these results suggest that inhibition of ACAT activity in cells promotes retention of APP in the ER.

Figure 4.
ACAT inhibition slows ER exit of APP in living cells. A) CHO cells expressing photoactivatable APP-paGFP fusion protein and DsRed2-ER (fluorescent ER marker protein) were treated with CP-113,818 for 4 d. APP-paGFP was activated by a short laser pulse ...

ACAT inhibition enhances retrotranslocation of APP without inducing ER stress

Several studies have suggested that the ubiquitin-proteasome system (UPS) is involved in the turnover of APP holoprotein, especially in the ER-associated degradation pathway of the nascent, immature APP molecules (26, 37). Cytosolic chaperones Hsc73 and the C-terminal Hsp70 interacting protein (CHIP) regulate proteasomal degradation of APP holoprotein (38, 39). Also, a signaling adaptor protein known as MOCA, also known as DOCK3, reduces Aβ generation and APP maturation by directing nascent APP molecules to proteasomal degradation (16). Thus, it is possible that the ER-associated degradation pathway (ERAD; reviewed in ref. 40) could participate in disposal of ER-retained APP molecules in ACAT inhibitor-treated cells.

To examine the accumulation of APP holoprotein in the cytosol, we used a previously described retrotranslocation assay (26, 27). In this assay, cells are treated with proteasome inhibitor, such as epoxomicin, to allow accumulation of UPS substrates, followed by isolation of cytosolic fraction from cells semipermeabilized with 0.04% digitonin. To test whether ER retention of APP is coupled to retrotranslocation and degradation by the cytosolic UPS, we used CHO cells expressing an APP construct carrying a dilysine ER retention signal at the C terminus (APP751-V5-KKAA). C-terminal dilysine motif has been previously shown to effectively retain APP in pre-Golgi compartments, resulting in reduced Aβ secretion (41, 42). As shown in Fig. 5A, ER retention of APP751-V5-KKAA results in enhanced retrotranslocation to the cytosol as compared to cells expressing wild-type APP751-V5. Thus, ER retention of APP is coupled to its dislocation from the ER membrane to the cytosol followed by degradation by the proteasome.

Figure 5.
ACAT inhibition enhances ERAD of APP but does not induce ER stress. A) CHO cells expressing either wild-type APP751–V5 or an ER-retention signal containing APP751-V5-KKAA were treated for 4 d with 10 μM CP-113,818 or vehicle. For the last ...

Next, we treated CHO cells and B104 neuroblastoma cells expressing wild-type human APP with CP-113,818 and, for the last 5 h of the treatment, with epoxomicin. CP-113,818 treatment increased the amount of cytosolic APP forms in both cell lines in a proteasome inhibitor-dependent manner (Fig. 5B). We have previously reported the presence of two forms of APP in the cytosol of proteasome inhibitor-treated cells: APP-im and an N-terminally truncated form of APP-im (26). Both forms seem to be increased in CP-113,818-treated CHO/APP and B104/APP cells. To confirm that the APP species seen in the cytosols of proteasome inhibitor-treated cells are newly synthesized, we combined metabolic labeling with the retrotranslocation assay. CHO/APP cells were treated with for 4 d with CP-113,818, and for the last 5 h, 1 μM epoxomicin was added. The cells were labeled with [35S]-methionine/cysteine for 15 min, then chased for 60 min (CP-113,818 and epoxomicin remained present throughout these steps). After isolation of the cytosolic fractions from semipermeabilized cells, APP was immunoprecipitated from the fractions. Total detergent lysates were used as controls (Fig. 5C). In the total lysates at the 60-min time point, the mature APP is already decreasing in control-treated cells but peaking in the CP-113,818-treated cells. As shown in Fig. 5C, epoxomicin treatment reveals labeled APP accumulating in the cytosol after 60 min chase. The major species of labeled APP migrates slightly below the immature APP (as compared to APP in the total lysates) at ~95–97 kDa, similarly to that shown in Fig. 5B. Notably, CP-113,818 treatment increased the level of cytosolic APP by 58.3 ± 17.6% (P=0.0146; Fig. 5D), suggesting that ACAT inhibition enhances retrotranslocation of APP from the ER for degradation by the cytosolic proteasomes.

Quality control mechanisms of the secretory pathway are essential for cellular homeostasis (reviewed in refs. 43, 44). Specific stress-sensor systems will activate the unfolded protein response (UPR) if the folding machinery becomes overloaded or misfolded proteins begin to accumulate in the ER (reviewed in ref. 45). The cellular response to ER stress is to transcriptionally activate genes that encode protein chaperones, such as BiP/GRP78, as well as various transcription factors, such as CHOP [CCAAT/enhancer-binding protein (C/EBP)]-homologous protein). Increased BiP/GRP78 expression appears to be an early response to ER stressors, whereas CHOP is a UPR-induced cell-death mediator linking prolonged ER stress to apoptosis (45). To evaluate whether ACAT inhibition induces UPR or chronic ER stress, we treated H4 neuroglioma cells overexpressing APP for 4 d with CP-113,818. As a positive control, we used a prototypical ER stress inducer, tunicamycin (for 16 h). As shown in Fig. 5E, CP-113,818-treated cells did not contain detectable levels of CHOP, whereas in tunicamycin-treated cells, a prominent induction of CHOP expression was detected. Moreover, BiP/GRP78 levels in CP-113,818-treated cells were comparable to vehicle-treated control cells, whereas tunicamycin-treated cells displayed a strong up-regulation of BiP/GRP78 (Fig. 5E). Altogether, these data show that inhibition of ACAT activity in cells results in enhanced ERAD of nascent APP molecules, and that this occurs independently of UPR activation.

Correlation of reduced brain levels of mature APP holoprotein with APP-C99 and Aβ in ACAT inhibitor-treated mice

Finally, we tested whether ACAT inhibition also reduces APP maturation in animal models. Our previous studies have shown that ACAT inhibition efficiently reduces Aβ generation and amyloid pathology in a transgenic mouse model of AD (ref. 22 and unpublished results). A 2-mo treatment of hAPP mice with CP-113,818 resulted in markedly reduced brain levels of Aβ and APP-CTFs without affecting other γ-secretase substrates, ApoE, BACE1, or γ-secretase component levels. Moreover, CP-113,818 had no effect on BACE1 or γ-secretase activity or on Aβ aggregation in vitro, suggesting that α- and β-secretase-mediated proteolytic processing of APP is specifically modulated by CP-113,818. We homogenized and extracted brain hemispheres from wild-type mice treated for 2 mo with 7.2 mg/kg/d CP-113,818 (from 4.5 to 6.5 mo of age) (22) and analyzed the levels of endogenous APP holoprotein in brain extracts. As compared to placebo-treated control mice, animals that received CP-113,818 showed a significant reduction in mature endogenous APP (Fig. 6A). In fact, the ratio of mature APP to immature APP was reduced by 50.9% (n=5, P=0.0063) in CP-113,818-treated mice (Fig. 6B). Nicastrin, one of the proteins comprising the γ-secretase complex, is a heavily glycosylated type 1 transmembrane protein that undergoes similar maturation processes along the secretory pathway as APP (9, 46). The ratio of mature to immature nicastrin was not significantly altered in CP-113,818-treated mice as compared to the placebo cohort (Fig. 6C), suggesting that CP-113,818 does not alter the maturation of all secretory proteins.

Figure 6.
Treatment with ACAT inhibitor CP-113,818 modulates maturation of endogenous APP in brains of wild-type mice. A) Wild-type mice (4.5 mo old; C57BL/6) were treated with 7.2.mg/kg/d CP-113,818 for 57 d. Dissected brain hemispheres were extracted in 1% Triton ...

We have recently tested the efficacy of CI-1011, a clinically relevant ACAT inhibitor, in an hAPP transgenic mouse model of AD. Here hAPP mice overexpressing human APP751 carrying the Swedish and London mutations (that increase Aβ production) were treated for 2 mo with 14.4 mg/kg/d CI-1011, starting at 4.5 mo of age. Triton X-100 extracts of brain hemispheres were analyzed on Western blots for APP holoprotein content. As shown in Fig. 7A, B, the ratio of mature APP to immature APP was reduced by 24.6% (n=9, P=0.0014) in CI-1011-treated animals. The milder effect of CI-1011 as compared to CP-113,818 may be explained by weaker efficacy of CI-1011 on ACAT and/or increased presence of APP in the transgenic mice as opposed to wild-type littermates in the CP-113,818 study. Notably, when a regression analysis was performed to evaluate relationship between mature APP and APP-C99 levels in individual animals, we noticed a strong positive correlation between these two forms of APP (Fig. 7C). Lower mature APP levels were associated with lower APP-C99 levels. Also, both treatment groups separated in different parts of the scatter plot with only partial overlap. Finally, we performed regression analysis between mature APP and CSF Aβ levels in individual animals. As shown in Fig. 7D, mature APP levels displayed a significant correlation with the amount of Aβ in CSF. These data strongly suggest that ACAT inhibitors limit the availability of mature APP molecules to Aβ-generating enzymes, BACE1 and γ-secretase, by modulating trafficking of APP in the early secretory pathway.

Figure 7.
Reduced maturation of APP correlates with decreased APP-C99 and cerebrospinal fluid Aβ levels in hAPP mice treated with CI-1011. A) hAPP transgenic mice (4.5 mo old) were treated with 14.4 mg/kg/d CI-1011 for 56 d. Dissected brain hemispheres ...


We have previously identified ACAT as a target molecule that modulates Aβ generation in cell- and animal-based models of AD (refs. 22,23,24). Here we have identified a molecular event connecting ACAT inhibition with altered APP metabolism. ACAT itself is localized to the ER, and reduced ACAT activity slightly increases the level of free cholesterol in the ER membrane while reducing cholesteryl ester production (23, 24). Our present results suggest that ACAT inhibition modulates APP trafficking in the early secretory pathway, likely at the ER-Golgi interface. Consequently, this delays and reduces maturation of APP, limiting the availability of APP holoprotein for Aβ-generating machinery, mostly localized to the post-Golgi compartments of the cell (31, 33,34,35). Retention of APP in the early secretory pathway was recently shown to reduce Aβ generation in cells (17). Notably, we found a reduced ratio of mature APP to immature APP in the brains of ACAT inhibitor-treated mice, correlating with decreased brain APP-C99 and CSF Aβ levels. Here we have presented evidence showing that APP trafficking is regulated in a sterol-dependent manner. Although ACAT inhibition may also directly regulate APP processing, our results suggest that decreased maturation of APP contributes to reduced Aβ generation.

What is the mechanism connecting reduced ACAT activity to enhanced ERAD of APP? One possibility is that ACAT activity, by contributing to formation of lipid droplets, could participate in extraction of misfolded proteins from the ER membrane. Cytoplasmic lipid droplets are composed mostly of neutral lipids such as cholesteryl esters that are generated by ACAT and form by budding from the ER membrane. It was recently hypothesized that cells could utilize lipid droplet formation in the ER membrane as an escape hatch to facilitate dislocation of ER proteins for cytosolic degradation (47). In another study, proteasomal and autophagic degradation pathways were suggested to converge on lipid droplet surfaces, particularly for disposal of poorly lipidated ApoB (48). Thus, ACAT activity could hypothetically be directly linked to the proteasomal and autophagic ERAD pathways. However, we have been unable to detect colocalization of APP with lipid droplet surface markers (unpublished results). Our data also show that ACAT inhibitors do not activate the UPR in cells with normal cholesterol levels. A recent report suggested that when macrophages were overloaded with acetylated LDL, ACAT inhibition induced CHOP expression and ER stress, leading to ER dysfunction and apoptosis (49). However, neither BiP/GRP78, an early UPR-induced gene, nor CHOP, a chronic ER stress marker, was induced in our cells treated with ACAT inhibitors for 4 d. Moreover, we have not observed any indications of ER stress, such as increased CHOP induction, in brain tissue from mice treated with ACAT inhibitors for 2 mo (unpublished results).

One explanation for APP retention could be altered APP protein interactions in the early secretory pathway. In fact, we have recently identified several ER-associated proteins, such as the chaperone GRP94, whose binding to APP is enhanced on ACAT inhibition (unpublished results). Although the causal relationship between the binding of these proteins to APP and ER retention of APP needs to be established in future studies, increased chaperone binding of immature APP on ACAT inhibition may indicate APP misfolding. Thus, based on our current data, we hypothesize that increased ERAD of APP is rather a consequence of the ER retention than an actual mechanism responsible for reduced ER exit and maturation of APP. Regardless of the exact mechanism responsible for sterol-responsive ER retention and ERAD of APP, our current data add another layer of complexity to the elaborate, bidirectional relationship between APP/Aβ and cellular lipid homeostasis.

At a subcellular level, we found that in cells with reduced ACAT activity additional mechanisms affect APP metabolism, possibly later in the life cycle of APP. Indeed, in AC29 cells, a cholesterol mutant cell line defective in ACAT activity, an alternative endosomal proteolytic processing event replaces normal α-, β-, and γ-secretase-mediated processing of APP, leading to a dramatic reduction in Aβ secretion (50). However, it is plausible that this alternative cleavage occurs only when total cellular cholesterol levels are increased (as seen in AC29 cells due to a mutation in the SREBP pathway) because we have been unable to detect these events in normocholesterolemic cells.

Treatments aiming at lowering brain Aβ content are expected to be the most effective disease-modifying therapies for AD. Moreover, understanding molecular mechanisms is an important factor in the overall success of clinical drug development, particularly with complex diseases such as AD. Our current data suggest a mechanism that may be responsible for the antiamyloidogenic effects of ACAT inhibitors and, thus, pave the way for clinical testing of ACAT inhibitors for prevention and/or treatment of AD.

Supplementary Material

Supplemental Data:


This study was supported by grants from the Cure Alzheimer’s Fund (D.M.K.), the U.S. National Institutes of Health (R01 NS45860; D.M.K.), the Helsingin Sanomat Centennial Foundation (H.J.H.), and the Maud Kuistila Foundation (H.J.H.). CI-1011 and CP-113,818 were kind gifts from Lit-Fui Lau and James Harwood (Pfizer, Groton, CT, USA), respectively.


  • Wolfe M S, Guenette S Y. APP at a glance. J Cell Sci. 2007;120:3157–3161. [PubMed]
  • Anliker B, Muller U. The functions of mammalian amyloid precursor protein and related amyloid precursor-like proteins. Neurodegener Dis. 2006;3:239–246. [PubMed]
  • Senechal Y, Larmet Y, Dev K K. Unraveling in vivo functions of amyloid precursor protein: insights from knockout and knockdown studies. Neurodegener Dis. 2006;3:134–147. [PubMed]
  • Theuns J, Brouwers N, Engelborghs S, Sleegers K, Bogaerts V, Corsmit E, De Pooter T, van Duijn C M, De Deyn P P, Van Broeckhoven C. Promoter mutations that increase amyloid precursor-protein expression are associated with Alzheimer disease. Am J Hum Genet. 2006;78:936–946. [PMC free article] [PubMed]
  • Rovelet-Lecrux A, Hannequin D, Raux G, Le Meur N, Laquerriere A, Vital A, Dumanchin C, Feuillette S, Brice A, Vercelletto M, Dubas F, Frebourg T, Campion D. APP locus duplication causes autosomal dominant early-onset Alzheimer disease with cerebral amyloid angiopathy. Nat Genet. 2006;38:24–26. [PubMed]
  • Lott I T, Head E. Alzheimer disease and Down syndrome: factors in pathogenesis. Neurobiol Aging. 2005;26:383–389. [PubMed]
  • Matsui T, Ingelsson M, Fukumoto H, Ramasamy K, Kowa H, Frosch M P, Irizarry M C, Hyman B T. Expression of APP pathway mRNAs and proteins in Alzheimer’s disease. Brain Res. 2007;1161:116–123. [PubMed]
  • Weidemann A, Konig G, Bunke D, Fischer P, Salbaum J M, Masters C L, Beyreuther K. Identification, biogenesis, and localization of precursors of Alzheimer’s disease A4 amyloid protein. Cell. 1989;57:115–126. [PubMed]
  • Haass C. Take five-BACE and the gamma-secretase quartet conduct Alzheimer’s amyloid beta-peptide generation. EMBO J. 2004;23:483–488. [PMC free article] [PubMed]
  • Gandy S. The role of cerebral amyloid beta accumulation in common forms of Alzheimer disease. J Clin Invest. 2005;115:1121–1129. [PMC free article] [PubMed]
  • Cataldo A M, Barnett J L, Pieroni C, Nixon R A. Increased neuronal endocytosis and protease delivery to early endosomes in sporadic Alzheimer’s disease: neuropathologic evidence for a mechanism of increased beta-amyloidogenesis. J Neurosci. 1997;17:6142–6151. [PubMed]
  • Vassar R, Bennett B D, Babu-Khan S, Kahn S, Mendiaz E A, Denis P, Teplow D B, Ross S, Amarante P, Loeloff R, Luo Y, Fisher S, Fuller J, Edenson S, Lile J, Jarosinski M A, Biere A L, Curran E, Burgess T, Louis J C, Collins F, Treanor J, Rogers G, Citron M. Beta-secretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane aspartic protease BACE. Science. 1999;286:735–741. [PubMed]
  • Chyung J H, Raper D M, Selkoe D J. Gamma-secretase exists on the plasma membrane as an intact complex that accepts substrates and effects intramembrane cleavage. J Biol Chem. 2005;280:4383–4392. [PubMed]
  • Kaether C, Schmitt S, Willem M, Haass C. Amyloid precursor protein and Notch intracellular domains are generated after transport of their precursors to the cell surface. Traffic. 2006;7:408–415. [PubMed]
  • Yang Y, Turner R S, Gaut J R. The chaperone BiP/GRP78 binds to amyloid precursor protein and decreases Abeta40 and Abeta42 secretion. J Biol Chem. 1998;273:25552–25555. [PubMed]
  • Chen Q, Kimura H, Schubert D. A novel mechanism for the regulation of amyloid precursor protein metabolism. J Cell Biol. 2002;158:79–89. [PMC free article] [PubMed]
  • Paganetti P, Calanca V, Galli C, Stefani M, Molinari M. Beta-site specific intrabodies to decrease and prevent generation of Alzheimer’s Abeta peptide. J Cell Biol. 2005;168:863–868. [PMC free article] [PubMed]
  • Schmidt V, Sporbert A, Rohe M, Reimer T, Rehm A, Andersen O M, Willnow T E. SorLA/LR11 regulates processing of amyloid precursor protein via interaction with adaptors GGA and PACS-1. J Biol Chem. 2007;282:32956–32964. [PubMed]
  • Puglielli L, Tanzi R E, Kovacs D M. Alzheimer’s disease: the cholesterol connection. Nat Neurosci. 2003;6:345–351. [PubMed]
  • Wolozin B. Cholesterol and the biology of Alzheimer’s disease. Neuron. 2004;41:7–10. [PubMed]
  • Hirsch-Reinshagen V, Wellington C L. Cholesterol metabolism, apolipoprotein E, adenosine triphosphate-binding cassette transporters, and Alzheimer’s disease. Curr Opin Lipidol. 2007;18:325–332. [PubMed]
  • Hutter-Paier B, Huttunen H J, Puglielli L, Eckman C B, Kim D Y, Hofmeister A, Moir R D, Domnitz S B, Frosch M P, Windisch M, Kovacs D M. The ACAT inhibitor CP-113,818 markedly reduces amyloid pathology in a mouse model of Alzheimer’s disease. Neuron. 2004;44:227–238. [PubMed]
  • Huttunen H J, Greco C, Kovacs D M. Knockdown of ACAT-1 reduces amyloidogenic processing of APP. FEBS Lett. 2007;581:1688–1692. [PMC free article] [PubMed]
  • Puglielli L, Konopka G, Pack-Chung E, Ingano L A, Berezovska O, Hyman B T, Chang T Y, Tanzi R E, Kovacs D M. Acyl-coenzyme A: cholesterol acyltransferase modulates the generation of the amyloid beta-peptide. Nat Cell Biol. 2001;3:905–912. [PubMed]
  • Grimm M O, Grimm H S, Hartmann T. Amyloid beta as a regulator of lipid homeostasis. Trends Mol Med. 2007;13:337–344. [PubMed]
  • Huttunen H J, Guenette S Y, Peach C, Greco C, Xia W, Kim D Y, Barren C, Tanzi R E, Kovacs D M. HtrA2 regulates beta-amyloid precursor protein (APP) metabolism through endoplasmic reticulum-associated degradation. J Biol Chem. 2007;282:28285–28295. [PubMed]
  • Forster M L, Sivick K, Park Y N, Arvan P, Lencer W I, Tsai B. Protein disulfide isomerase-like proteins play opposing roles during retrotranslocation. J Cell Biol. 2006;173:853–859. [PMC free article] [PubMed]
  • Rockenstein E, Mallory M, Mante M, Sisk A, Masliaha E. Early formation of mature amyloid-beta protein deposits in a mutant APP transgenic model depends on levels of Abeta(1-42) J Neurosci Res. 2001;66:573–582. [PubMed]
  • Alegret M, Llaverias G, Silvestre J S. Acyl coenzyme A: cholesterol acyltransferase inhibitors as hypolipidemic and antiatherosclerotic drugs. Methods Find Exp Clin Pharmacol. 2004;26:563–586. [PubMed]
  • Yeo K T, Parent J B, Yeo T K, Olden K. Variability in transport rates of secretory glycoproteins through the endoplasmic reticulum and Golgi in human hepatoma cells. J Biol Chem. 1985;260:7896–7902. [PubMed]
  • Haass C, Lemere C A, Capell A, Citron M, Seubert P, Schenk D, Lannfelt L, Selkoe D J. The Swedish mutation causes early-onset Alzheimer’s disease by beta-secretase cleavage within the secretory pathway. Nat Med. 1995;1:1291–1296. [PubMed]
  • Parvathy S, Hussain I, Karran E H, Turner A J, Hooper N M. Cleavage of Alzheimer’s amyloid precursor protein by alpha-secretase occurs at the surface of neuronal cells. Biochemistry. 1999;38:9728–9734. [PubMed]
  • Ehehalt R, Keller P, Haass C, Thiele C, Simons K. Amyloidogenic processing of the Alzheimer beta-amyloid precursor protein depends on lipid rafts. J Cell Biol. 2003;160:113–123. [PMC free article] [PubMed]
  • Grbovic O M, Mathews P M, Jiang Y, Schmidt S D, Dinakar R, Summers-Terio N B, Ceresa B P, Nixon R A, Cataldo A M. Rab5-stimulated up-regulation of the endocytic pathway increases intracellular beta-cleaved amyloid precursor protein carboxyl-terminal fragment levels and Abeta production. J Biol Chem. 2003;278:31261–31268. [PubMed]
  • Rajendran L, Honsho M, Zahn T R, Keller P, Geiger K D, Verkade P, Simons K. Alzheimer’s disease beta-amyloid peptides are released in association with exosomes. Proc Natl Acad Sci U S A. 2006;103:11172–11177. [PMC free article] [PubMed]
  • Patterson G H, Lippincott-Schwartz J. A photoactivatable GFP for selective photolabeling of proteins and cells. Science. 2002;297:1873–1877. [PubMed]
  • Hare J F. Protease inhibitors divert amyloid precursor protein to the secretory pathway. Biochem Biophys Res Commun. 2001;281:1298–1303. [PubMed]
  • Kouchi Z, Sorimachi H, Suzuki K, Ishiura S. Proteasome inhibitors induce the association of Alzheimer’s amyloid precursor protein with Hsc73. Biochem Biophys Res Commun. 1999;254:804–810. [PubMed]
  • Kumar P, Ambasta R K, Veereshwarayya V, Rosen K M, Kosik K S, Band H, Mestril R, Patterson C, Querfurth H W. CHIP and HSPs interact with beta-APP in a proteasome-dependent manner and influence Abeta metabolism. Hum Mol Genet. 2007;16:848–864. [PubMed]
  • Meusser B, Hirsch C, Jarosch E, Sommer T. ERAD: the long road to destruction. Nat Cell Biol. 2005;7:766–772. [PubMed]
  • Maltese W A, Wilson S, Tan Y, Suomensaari S, Sinha S, Barbour R, McConlogue L. Retention of the Alzheimer’s amyloid precursor fragment C99 in the endoplasmic reticulum prevents formation of amyloid beta-peptide. J Biol Chem. 2001;276:20267–20279. [PubMed]
  • Cupers P, Bentahir M, Craessaerts K, Orlans I, Vanderstichele H, Saftig P, De Strooper B, Annaert W. The discrepancy between presenilin subcellular localization and gamma-secretase processing of amyloid precursor protein. J Cell Biol. 2001;154:731–740. [PMC free article] [PubMed]
  • Bukau B, Weissman J, Horwich A. Molecular chaperones and protein quality control. Cell. 2006;125:443–451. [PubMed]
  • Anelli T, Sitia R. Protein quality control in the early secretory pathway. EMBO J. 2008;27:315–327. [PMC free article] [PubMed]
  • Ron D, Walter P. Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol. 2007;8:519–529. [PubMed]
  • Shah S, Lee S F, Tabuchi K, Hao Y H, Yu C, LaPlant Q, Ball H, Dann C E R, Sudhof T, Yu G. Nicastrin functions as a gamma-secretase-substrate receptor. Cell. 2005;122:435–447. [PubMed]
  • Ploegh H L. A lipid-based model for the creation of an escape hatch from the endoplasmic reticulum. Nature. 2007;448:435–438. [PubMed]
  • Fujimoto T, Ohsaki Y. Proteasomal and autophagic pathways converge on lipid droplets. Autophagy. 2006;2:299–301. [PubMed]
  • Feng B, Yao P M, Li Y, Devlin C M, Zhang D, Harding H P, Sweeney M, Rong J X, Kuriakose G, Fisher E A, Marks A R, Ron D, Tabas I. The endoplasmic reticulum is the site of cholesterol-induced cytotoxicity in macrophages. Nat Cell Biol. 2003;5:781–792. [PubMed]
  • Huttunen H J, Puglielli L, Ellis B C, Mackenzie Ingano L A, Kovacs D M. Novel N-terminal cleavage of APP precludes Abeta generation in ACAT-defective AC29 cells. J Mol Neurosci. 2009;37:6–15. [PMC free article] [PubMed]

Articles from The FASEB Journal are provided here courtesy of The Federation of American Societies for Experimental Biology
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...