- Email: [email protected]
Secretion of apolipoprotein E by brain glia requires protein prenylation and is suppressed by statins
Brain Research 958 (2002) 100–111 www.elsevier.com / locate / brainres
Research report
Secretion of apolipoprotein E by brain glia requires protein prenylation and is suppressed by statins Asha Naidu, Qiang Xu 1 , Rosanne Catalano, Barbara Cordell* Scios Inc., 820 West Maude Avenue, Sunnyvale, CA 94085, USA Accepted 19 August 2002
Abstract Apolipoprotein E (ApoE) genotype modulates the risk of Alzheimer’s disease. ApoE has been shown essential for amyloid b-peptide fibrillogenesis and deposition, a defining pathological feature of this disease. Because astrocytes and microglia represent the major source of extracellular apoE in brain, we investigated apoE secretion by glia. We determined that protein prenylation is required for apoE release from a continuous microglial cell line, primary mixed glia, and from organotypic hippocampal cultures. Using selective protein prenylation inhibitors, apoE secretion was found to require protein geranylgeranylation. This prenylation involved a protein critical to apoE secretion, not apoE proper. ApoE secretion could also be suppressed by inhibiting synthesis of mevalonate, the precursor to both types of protein prenylation, using hydroxyl-3-methylglutaryl coenzyme A reductase inhibitors (statins). Recent reports have described the beneficial effects of statins on the risk of dementia. Our finding that protein geranylgeranylation is required for apoE secretion in the brain parenchyma provides another contributing mechanism to explain the effective properties of statins against the development of dementia. In this model, statin-mediated inhibition of mevalonate synthesis, an essential reaction in forming geranylgeranyl lipid, would lower extracellular levels of parenchymal apoE. Because apoE has been found necessary for plaque development in transgenic models of Alzheimer’s disease, suppressing apoE secretion by statins could reduce plaques and, in turn, improve cognitive function. 2002 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: Alzheimer’s Keywords: ApoE; Alzheimer; Glia; Prenylation
1. Introduction Multiple genetic and environmental factors interact to generate Alzheimer’s disease (AD). Genetic defects, including both causative and susceptibility genes, have been identified. One common susceptibility gene is that encoding the type 4 isoform of apolipoprotein E (APOE), a cholesterol transport protein [5,7,38]. Epidemiological studies suggest a relationship between cholesterol levels and AD. Individuals carrying the APOE-´4 allele have elevated plasma cholesterol levels [17,28] and manifest an *Corresponding author. Tel.: 11-650-493-2715; fax: 11-925-5603901. E-mail address: [email protected] (B. Cordell). 1 Current address: Osel Inc., 1800 Wyatt Drive, Santa Clara, CA 95054, USA.
increased risk of cardiovascular disease [8]. Correspondingly, cardiovascular disease has also been shown to predispose to AD, vascular dementia, and stroke [42,43]. This association between cholesterol and AD is further supported by animal studies. Sparks et al. found that rabbits fed a high fat cholesterol diet developed amyloid b-peptide (Ab) immunoreactive deposits in the frontal cortex and hippocampus, the brain regions displaying pathology in AD [44,45]. Transgenic studies have also demonstrated a direct correlation between elevated cholesterol levels and increased Ab plaque burden upon increased dietary cholesterol intake [36]. In addition to the role of apoE in cholesterol transport, this protein has also been implicated in AD pathogenesis by other mechanisms. The APOE-´4 allele not only increases the risk of developing AD but also the plaque burden in a gene dosage-dependent manner [13], as well as
0006-8993 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03480-7
A. Naidu et al. / Brain Research 958 (2002) 100–111
Ab production [33]. Neuropathological analyses have revealed extracellular apoE immunoreactivity associated with Ab amyloid plaques in the parenchyma and in the cerebrovasculature, although not with non-fibrillar Ab deposits [14,39,47]. Moreover, stable complexes between apoE and Ab have been isolated from AD brain [27,31,37]. These in vivo observations, plus the high avidity binding of apoE to Ab in vitro promoting fibrillogenesis [20,23,50], suggest that apoE promotes plaque formation through the conversion of Ab to a fibrillar structure. This concept is further supported by transgenic crosses which have shown the absence of fibrillar plaques in transgenic mice on an APOE null background, in contrast to abundant plaques in the context of APOE alleles [1,2]. Second to the liver, the brain produces the largest quantities of apoE. Astrocytes and microglia are the major sources of extracellular apoE [6,26,32,53]. Glia release apoE as free protein and as a component of lipoprotein particles [32,53]. Because apoE appears to play a critical role in AD pathogenesis by modulating cholesterol levels and Ab plaque formation, we sought to better define the molecular factors required for apoE secretion. Our initial approach was to assess the effects of hydroxyl-3methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) on apoE secretion from cultured glial systems. Statins are commonly prescribed to reduce plasma cholesterol levels and consequently the risk of cardiovascular disease. Our reasons for testing statins in this system are several. First, there is growing evidence for additional benefits of statins that cannot be completely explained by their lipid-lowering effects. Indeed, studies have indicated immunomodulatory actions of this drug class [19,51]. Clinical studies involving transplant recipients indicate possible immunosuppressive effects of statins. Additionally, anti-inflammatory actions of statins have been reported in which NF-kB activation and subsequent induction of nitric oxide synthase and cytokines in isolated macrophages and glia could be prevented by lovastatin [29]. In this context, it is noteworthy that we reported that pro-inflammatory agents, including Ab and interleukin-6, can promote the secretion of glial apoE via NF-kB activation [3], providing a rationale for testing statins effects on apoE secretion. Second, a lower prevalence of AD and a lower risk for dementia in general have been observed with individuals taking statins [18,52]. Whether this effect is due solely to a reduction in cholesterol levels or whether statins affect Ab plaque formation, or both, is unclear. Third, HMG-CoA reductase inhibitors have been shown capable of altering both gene transcription and signaling [9,30]. Here, we show that statins block apoE release from glial cell lines, primary mixed glia, and organotypic hippocampal cultures. The inhibition of glial apoE secretion by statins was determined to be due to prenylation of an unidentified protein required for apoE secretion, not apoE itself.
101
2. Materials and methods
2.1. Reagents Lovastatin, mevastatin, FTI-277, FTI-276, GGTI-287, GGTI-286, FPT-II, FPT-III, and FTase-I were purchased from CalBiochem. Mevalonate, geranylgeranyl pyrophosphate, and farnesyl pyrophosphate were purchased from Sigma. Alendronate (4-amino-1-hydroxybutylidene bisphosphonate) was a gift from Merck Research Laboratories.
2.2. Cell and tissue culture The growth and propagation of the BV-2 murine microglial cell line has been described previously [53]. Primary mixed glia were prepared from JU [34] or bAPP V717F mice [1,2] using cortices from 1-day to 3-day old pups. Dissected cortices, from which the meninges were removed, were washed twice with Hank’s balanced salt solution containing 20 mM Ca 21 and 20 mM Mg 21 (HBSS with Ca 21 Mg 21 ). Tissue was trypsinized with 0.05% trypsin in HBSS Ca 21 Mg 21 containing 50 mM EDTA, 100 units / ml penicillin and 100 mg / ml streptomycin for 20 min at room temperature, after which the trypsin solution was removed and the tissue rinsed twice with HBSS Ca 21 Mg 21 . The cortices were triturated in growth medium (Delbecco’s minimal essential medium with high glucose, 1 mM sodium pyruvate, 2 mM Lglutamine, 10% fetal bovine serum, 50 units / ml penicillin and 50 mg / ml streptomycin) then seeded into flasks at a density of six cortices per T75 flask. The cultures were incubated in 5% CO 2 at 37 8C. Approximate cell populations in the isolated primary glial cultures were assessed using immunohistochemistry. Staining for glial acidic fibrillary protein (antibody purchased from Sigma) to identify astrocytes indicated that this glial cell type represented 80–90% of the total culture over a 10-day period. Staining with F4 / 80, an antibody marker for microglia (Serotec; Raleigh, NC, USA) revealed low levels of microglia, roughly 5–8% of the culture. Organotypic hippocampal slice cultures were prepared essentially as described by Stoppini et al. [46]. Briefly, brains from 8 day old pups of the JU strain or bAPP V717F transgenic mouse were dissected into Minimal Essential medium containing 10 mM Tris–HCl, pH 7.2, supplemented with penicillin and streptomycin. Hippocampi were isolated under a dissecting microscope and 400 mm sections made using a MacIllwain tissue chopper. Sections were separated by vigorously swirling the culture dish. Four sections were placed on Millipore culture inserts set into the well of a six-well tissue culture plate containing 1.2 ml culture medium (Delbecco’s minimal essential medium with high glucose, 12.5 mM HEPES, 25% HBSS with Ca 21 Mg 21 , 1 mM L-glutamine, 25% horse serum, 50 units / ml penicillin
102
A. Naidu et al. / Brain Research 958 (2002) 100–111
and 50 mg / ml streptomycin) and incubated at 37 8C in 5% CO 2 .
2.3. ApoE measurement Conditioned media and lysates were assayed for apoE levels using Western blot or ELISA. For Western blot analysis, samples were electrophoresed on reducing 12% SDS–PAGE. After membrane transfer, apoE was detected with a rabbit anti-murine apoE antiserum purchased from Biodesign International (Kennebunk, ME, USA). The antigen antibody reaction was visualized by using a secondary antibody conjugated with horse-radish peroxidase and enhanced chemiluminesence detection reagents (Amersham Pharmacia Biotech). We previously described quantifying rodent apoE by ELISA [25].
2.4. Inhibitor treatment and lipid rescue BV2, primary glia, and hippocampal slice cultures were maintained in serum-free medium during exposure to inhibitors. A concentrated stock solution of each inhibitor was prepared and diluted into serum-free medium to the designated final concentration. Medium containing inhibitor was added to cultures washed free of serum and incubated at 37 8C in 5% CO 2 for the times indicated in each experiment. After treatment, the conditioned medium was collected, centrifuged at low speed to remove cellular debris, and assayed for apoE. Treated cells were washed twice with phosphate buffered saline, lysed by three cycles of freeze–thaw at 280 and 27 8C in serum-free culture medium, and then assayed for cell-associated apoE. For experiments involving rescue of prenylation inhibition, mevalonate, farnesyl-pyrophosphate (FPP) or geranylgeranyl-pyrophosphate (GGPP), both solubilized in culture medium, cultures were pre-treated overnight with 10 mM of each intermediate. Cultures were changed to serum-free medium containing 5 mM lovastatin and 10 mM GGPP or FPP, incubated for 5 h, and media collected for analysis. To deliver GGPP and FPP using liposomes, liposomes were prepared by mixing 5 mmol dipalmitoylphosphatidylcholine with 200 mg GGPP or FPP and drying by rotary evaporation. The dried mixtures were resuspended in phosphate buffered saline, warmed to 50 8C and sonicated for 30 s. Control liposomes containing only dipalmitoylphosphatidylcholine were similarly prepared.
2.5. Toxicity assessment For cell cultures, toxicity was measured by the 3-(4,5dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrozolium salt (MTS) assay using a commercially available kit (CellTiter 96 AQ ueus NonRadioactive Cell Proliferation Assay, Promega, Madison, WI, USA). This is a colorimetric assay for determining the number of viable cells whereby the MTS is bioreduced to
formazan by dehydrogenase enzymes in metabolically active cells. In brief, the MTS reagent was added to cells after treatment with the various inhibitors. The cells were incubated for 1–2 h at 37 8C in the presence of the MTS reagent. After appropriate color development, absorbency was read at 490 nm. For hippocampal slice cultures, toxicity was measured by lactate dehydrogenase (LDH) release assay using a commercially available kit (CytoTox 96 NON-radioactive Cytotoxicity Assay, Promega). The kit quantitatively measures LDH release upon cell lysis. Aliquots (75 ml) of slice culture medium were assayed in the LDH assay after the slices were treated with various inhibitors. For a positive control for LDH release, slice cultures were treated with 10% ethanol for 24, 48, and 72 h and the conditioned medium was assayed in the LDH assay. To determine the total amount of LDH present in the slices, lysates from untreated slices were prepared by freezing at 280 8C and thawing the slices three times in culture medium and assaying the lysates in the LDH assay.
3. Results Microglia and apoE have each been implicated in the pathogenesis of senile plaques in AD. We recently showed that cultures of BV2 murine microglial cells release apoE [53] and that apoE release could be enhanced by inflammatory stimuli, including Ab [3]. In the present study, we sought to define the molecular process governing apoE production by BV2 microglia and by primary glial cultures. We investigated the effects of statins on apoE secretion by BV2 microglia for two reasons. ApoE secretion by BV2 cells utilizes a NF-kB-dependent mechanism identical to that for the release of inflammatory cytokines [3] (our unpublished observations). HMG-CoA reductase inhibitors (statins) have been reported to inhibit cytokine secretion from rat primary astrocytes, microglia and macrophages [29].
3.1. Statins block apoE release from cultured glia Using Western blot analysis, cultures of BV2 cells in serum-free medium release apoE which accumulates in a time-dependent manner (Fig. 1A). The observed doublet of secreted apoE immunoreactivity represents sialyated and non-sialyated forms of apoE [53]. Accumulation of apoE in the conditioned medium was also evident when assaying samples by ELISA (Fig. 1B). Cell-associated levels, however, were relatively constant after an initial increase (Fig. 1C). ApoE production by BV2 cultures after treatment with the HMG-CoA reductase inhibitors, mevastatin and lovastatin, was assessed. Both statins inhibited apoE secretion in a dose-dependent fashion (Fig. 2A). Approximately 85% of the apoE released into the medium was inhibited at the highest dose tested. Identical results were obtained by Western blot (Fig. 2C). Cell-associated levels
A. Naidu et al. / Brain Research 958 (2002) 100–111
103
Fig. 1. Production of apoE by BV-2 microglial cells. Cultures of |10 5 BV-2 cells in serum-free medium were sampled for apoE at the times indicated. Conditioned media were used to assess secreted apoE and cell lysates were tested for cell-associated apoE by Western blot and ELISA as detailed in Materials and methods. The data represent one of four identical experiments. (A and B) Time-dependent release of apoE into medium using Western blot and ELISA detection, respectively. (C) Cell-associated apoE.
were unperturbed, indicating that statins were not depleting intracellular apoE pools (Fig. 2B). Statin suppression of apoE release from BV2 cells could not be attributed to a general suppression of protein secretion from the cells, as silver staining of the conditioned medium showed no gross differences in protein levels or pattern (data not shown). This effect was not due to toxicity as control and statintreated cultures were equally viable. To further verify that the statin effect on apoE release was authentic, we used mevalonate to rescue the inhibition. Since mevalonate is the product of HMG-CoA reductase, addition of mevalonate to the cells should overcome the statin inhibition. BV2 cells were treated with mevastatin in the presence or absence of 100 mM exogenous mevalonate. Secreted apoE was restored to levels equivalent to that of the untreated control medium when mevalonate was added to the mevastatin-treated cultures (Fig. 2D). We next examined the effects of statins on apoE release from primary murine mixed glial cultures. The cultures were assessed for the approximate proportion of astrocytes and microglia using immunocytochemistry that indicated that the cultures were predominantly populated by astrocytes with minor representation of microglia (10–15% of all cells). Similar to results with BV2 microglial cells, primary mixed glia release apoE in a time-dependent manner (Fig. 3A). This is in agreement with previous reports indicating that apoE is released by both microglia and astrocytes [6,26,32,53]. Exposing the cultures to 10 mM mevastatin resulted in partial inhibition of apoE as assessed by ELISA (Fig. 3B) or by Western blot (Fig. 3C).
3.2. Geranylgeranylation required for apoE release from glial cells The production of mevalonate by HMG-CoA reductase provides the precursor leading to the formation of cholesterol, dolichol, ubiquinone, as well as isoprenoids (Fig. 4). Isoprenoids can be covalently linked to proteins, a process referred to as protein prenylation. Protein prenylation is recognized as a mechanism to promote membrane interactions and biological activities of a variety of cellular proteins [41]. Because statins act at an early step in the mevalonate pathway, we wanted to determine if the inhibition of glial apoE secretion was due to inhibition of cholesterol preventing the formation of apoE-containing lipoprotein particles or, alternatively, was a result of blocking the prenylation of a protein required for apoE secretion. This latter possibility would be a downstream effect of inhibiting HMG-CoA reductase with statins. To address this issue, we tested whether lovastatin inhibition of apoE secretion from BV2 microglial cells could be rescued with farnesyl-pyrophosphate (FPP) or geranylgeranyl-pyrophosphate (GGPP), specific intermediates to farnesylation and geranylgeranylation, respectively. Rescue with mevalonate served as a control. Exposure of treated BV2 cells to 10 mM FPP or 10 mM GGPP indicated that GGPP, but not FPP, rescued |40% of the inhibition of apoE secretion generated by lovastatin. To better characterize the specific sterol requirement for apoE release by glia, we investigated the effects of wellcharacterized protein prenylation inhibitors. Several far-
104
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 2. Statin inhibition of apoE release by BV-2 microglia and rescue by mevalonate. Approximately 10 5 BV-2 cells were pretreated for 16–24 h in the presence of serum with the doses of mevastatin or lovastatin indicated. After pretreatment, cells were placed in serum-free medium and incubated for 4 h. ApoE, released into medium or remaining cell-associated, was measured by Western blot and ELISA. For mevalonate rescue, 100 mM exogenous mevalonate was included during the treatment period. The figure shows one representative experiment of several performed. (A and C) Dose-dependent inhibition of apoE secretion by mevastatin and lovastatin, respectively. (B) Unchanged cell-associated apoE levels during statin treatment. (D) Rescue of statin-induced inhibition of apoE secretion by mevalonate.
nesyl protein transferase inhibitors, FTase-I, FPT-II and FPT-III, were tested for inhibition of apoE secretion. The inhibitors differ in structure and all are reported to be cell permeable. FPT-III is a prodrug of FPT-II. All inhibitors were assessed using concentrations previously shown to be inhibitory in other cell systems [12,24]. Dose-dependent inhibition of apoE secretion from BV2 cells by FPT-II and FPT-III was observed (Fig. 5A and C). In contrast, the FTase-I inhibitor did not block apoE release (Fig. 5B). Primary murine mixed glia were also tested for response to FPT-II and FTase-I farnesyl transferase inhibitors. Similar to the results obtained with BV2 microglial cultures, FPT-
II treatment resulted in reduced apoE release, albeit partial, reaching a maximum of |50%, and FPTase-1 was without effect up to 200 mM (Fig. 6). The active doses of FPT-II and FPT-III in the glial apoE release assay were in the 100 mM range which have been reported to inhibit both farnesyl transferase and geranylgeranyl transferase [24]. To address which type of protein prenylation is required for apoE secretion, we tested more selective inhibitors. Two peptidomimetic farnesyl transferase inhibitors, FTI-276 and FTI-277, have been described as 100-fold more selective for farnesyl transferase over geranylgeranyl transferase. Likewise, two
A. Naidu et al. / Brain Research 958 (2002) 100–111
105
Fig. 3. Secretion of apoE by primary murine mixed glia and its inhibition by statins. Primary glial cultures containing |10 5 cells were assessed for apoE release in the presence or absence of statins. For statin treatment, 10 mM mevastatin or lovastatin was applied to the cultures in complete medium for 16–24 h, after which the cultures were placed in serum-free medium for 4–5 h containing 10 mM statin. Media were scored for apoE using Western blot and ELISA. The data show one of three separate experiments. (A) Time-dependent secretion of apoE. (B and C) Statin inhibition of apoE secretion.
geranylgeranyl transferase I inhibitors, GGTI-286 and GGTI-287, are selective inhibitors. GGTI-286 is a more cell-permeable derivative of GGTI-287. GGTI-286 is described as 25-fold more selective for geranylgeranyl transferase versus farnesyl transferase [22,48,49]. All four inhibitors were tested with BV2 microglia. Little or no inhibition of apoE release was seen with FTI-276, FTI-277 and GGTI-287, unlike GGTI-286 which showed partial inhibition (Fig. 7A–E). Extended treatment times resulted in greater inhibition up to 80%. The inhibitory effect of GGTI-286 on apoE release from primary mixed murine glia was similar (Fig. 7F). The poor cell permeability of GGTI-287 compared to its methyl ester derivative, GGTI-286, reconciles the different responses obtained with this pair of geranylgeranyl transferase inhibitors. Last, we tested alendronate, a nitrogen-containing bisphosphonate that has been reported to inhibit the production of geranylgeraniol [11]. The target of alendronate appears to be the enzyme involved in the conversion of mevalonate to GGPP [11]. Because alendronate has a different mode of action compared to the inhibitors tested thus far, we examined its influence on apoE secretion from BV2 cells and found that it was inhibitory in the absence
of toxicity (Fig. 7E). These results, together with the observed inhibition by GGTI-286, the lack of inhibition by FPT-276 and FPT-277, and the partial rescue by exogenous GGPP but not FPP, are consistent with the requirement for protein geranylgeranylation in the secretory machinery used for apoE release from glia.
3.3. Protein prenylation inhibition in hippocampal cultures blocks apoE release Because of the role of apoE in the pathogenesis of AD, we were interested in determining whether protein prenylation is required for apoE secretion by hippocampal tissue. To do so, we established conditions yielding quantitative and consistent production of apoE from murine hippocampal slice cultures. Inhibitors active in suppressing apoE secretion from cultured glia (lovastatin, FPT-II, GGTI-286, and alendronate) were each tested in the hippocampal slice culture system. Three media harvests were made and evaluated for apoE levels at 24, 48, and 72 h of cumulative exposure. Exposure to each prenylation inhibitor resulted in a time-dependent reduction of apoE in the conditioned medium over the 72 h treatment period compared to untreated control samples (Fig. 8). In addition, lovastatin
106
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 4. Protein prenylation pathway showing inhibitor targets.
suppression of apoE release in the hippocampal slice cultures could be rescued by addition of 200 mM mevalonate (data not shown). These data indicate that protein prenylation, specifically geranylgeranylation, is essential to the secretory machinery used to release apoE from cultured hippocampi, a target tissue of AD plaque pathology.
4. Discussion ApoE contributes to the development of AD in a number of ways. The APOE-´4 allele is a genetic risk modifier that decreases the age of onset in a dose-dependent manner [5,7,38]. The pathophysiological consequence of APOE-´4 is an increased Ab plaque burden in humans and in transgenic mice [13,16]. Noteworthy is the absence of Ab plaques in APP V717F transgenic mice on an APOE null background [1,2]. From in vitro studies, apoE appears to facilitate Ab fibril formation [20,23,50]. However, the normal physiological function of apoE in regulating plasma lipid and cholesterol metabolism, serving both as a
ligand for lipoprotein receptors and as a particle for lipid transport [5,38], may also contribute to AD. Since neurons and glia express a variety of receptors in the LDL family and express lipoproteins [15], apoE may function in a similar capacity in the brain parenchyma as it does in the periphery. The primary source of apoE in brain is glia. In vitro primary murine astrocytes secrete apoE as a component of a HDL-like particle [6,21,32]. In contrast, the BV2 murine microglial cell line releases lipid-poor protein aggregates of apoE [53]. Microglia are probably similar to peripheral macrophages which independently secrete apoE and cholesterol but, upon release, associate to form a particle [4]. In this study, we employed cultured BV2 cells, primary murine glia (predominately astrocytes), and murine organotypic hippocampal cultures. In each of these culture systems, we demonstrated a time-dependent release and accumulation of apoE into the medium. We found that inhibition of mevalonate production by HMG-CoA reductase inhibitors (lovastatin and mevastatin), general inhibition of protein prenylation with non-selective inhibitors (FPT-II, FPT-III, and alendronate), and inhibition of
A. Naidu et al. / Brain Research 958 (2002) 100–111
107
Fig. 5. Inhibition of apoE secretion by BV-2 cells after treatment with protein prenylation inhibitors. Cultures of |10 5 BV-2 cells in serum-free medium were treated with the indicated doses of FPT-II (A), FTase-I (B), or FPT-III (C) for 4–6 h, after which the conditioned media were assayed for apoE using ELISA. The data represent one of three independent determinations.
Fig. 6. Effects of protein prenylation inhibitors on apoE secretion by primary mixed glia. Murine primary mixed glial cultures containing |10 5 cells in serum-free medium were treated with FPT-II (A) or FTase-I (B) at the doses indicated for 4 h. ApoE released into the medium during this period was determined by ELISA.
108
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 7. Effects of protein prenylation inhibitors on BV-2 microglia and on primary mixed glia. Glial cultures containing |10 5 cells in serum-free medium were pretreated for 1 h with the doses of prenylation inhibitors indicated, after which the cultures were moved to serum-free medium containing the same dose of inhibitor. Treatments with FTI-276, FTI-277, GGTI-286 and GGTI-287 were for 4–6 h and alendronate treatment was for 24 h. Media and cell lysates were assayed by ELISA and Western blot. The data provided are representative of three separate determinations. BV-2 microglia were treated with increasing doses of FTI-276 (A), FTI-277, (B), GGTI-286 (C), GGTI-287 (D), and alendronate (E). Murine primary mixed glia were treated with 20 mM GGTI-286 (F).
protein geranylgeranylation using selective cell-penetrant agents against geranylgeranyl transferase (GGTI-286) blocked apoE secretion. Mevalonate is the up-stream precursor to GGPP and FPP and, indeed, we found mevalonate could rescue the inhibition by these compounds. GGPP and FPP are lipids used in prenylation, a post-translation mechanism to target proteins, such as those in the Ras superfamily, to cellular membranes [41]. Based on our results, it appears that the secretory machinery used to release apoE from glial cells contains an essential geranylgeranlyated protein. The observed effects
are not due to direct prenylation of apoE or to the depletion of cholesterol. Inhibition of apoE secretion by statins and prenylation inhibitors was quantitatively different between the BV2 microglial cells and cultures where the majority of the glial population is astrocytes, such as with primary murine mixed glia and organotypic hippocampal sections. In the pure microglial system, near complete suppression of apoE release could be achieved by the active inhibitors described. This was in contrast to the cultures rich in astrocytes where inhibition by these agents was partial,
A. Naidu et al. / Brain Research 958 (2002) 100–111
109
Fig. 8. Inhibition of apoE secretion by organotypic hippocampal cultures treated with inhibitors of protein prenylation. Microtiter wells each containing three murine hippocampal tissue slices in serum-free media were treated with the protein prenylation inhibitors found to block apoE release from BV-2 and primary glia. Media were harvested and replaced every 24 h with fresh drug-containing media. Media from quadruplicate samples were assayed from 24, 48, and 72 h harvest or drug treatment periods using ELISA. ApoE values in drug-treated samples were compared to media from mock-treated cultures similarly handled. The data reflect one of three similar experiments. (A) 100 mM lovastatin. (B) 100 mM FPT-II. (C) 100 mM alendronate. (D) 100 mM GGTI-286.
ranging between 40 and 60%. The partial inhibition in the astrocyte-rich cultures was consistent irrespective of the active inhibitor used. This suggests that apoE secretion from microglia relies solely on a mechanism(s) requiring a prenylated protein, whereas astrocytes use an identical or similar secretory mechanism, in part, as well as a secretory process that does not involve protein prenylation. In the context of lowering extracellular apoE in brain as a possible therapeutic approach to mitigate Ab plaque formation, such a partial inhibition may be effective. Moderate suppression of apoE levels could limit Ab nucleation and amyloid fibril formation. Support of this concept is seen in bAPP V717F transgenic mice carrying only one APOE allele in which the plaque frequency is 80% lower compared to transgenic mice with two alleles [2]. It is of interest that HMG-CoA reductase inhibitors,
statins, in addition to their beneficial lipid-lowering effects for treating cardiovascular disease, have been shown to reduce the risk of AD [52] and dementia [18] in two recent epidemiological studies. Statins were evaluated in AD patients because of the link between cholesterol metabolism and the pathophysiology of this disease. For example, elevated plasma cholesterol levels have been show to be a risk factor for AD. APOE-´4 polymorphism increases not only the risk of AD [7] but also elevates cholesterol levels and increases the risk of cardiovascular disease [8,17,28]. Furthermore, cardiovascular disease is a risk factor for AD [42,43]. Cholesterol levels have also been found to influence bAPP processing in cell culture [10,40] as well as Ab load in vivo [10,35]. Despite these associated observations, the effects of statins on cholesterol-lowering action and on bAPP processing do not fully explain their beneficial effects on reducing the risk of dementia. Drach-
110
A. Naidu et al. / Brain Research 958 (2002) 100–111
man and colleagues [18] showed that while statins lowered the risk of developing dementia by 70%, non-statin lipidlowering agents did not show a reduced risk. Hence, the mechanism by which statins reduce the risk of developing dementia is unclear. From the data presented in this work, one explanation is that statins suppress extracellular levels of apoE, thereby reducing Ab deposition and plaque formation. While viable hypotheses have been put forward to explain the positive statin effects on the development of dementia [18,52], statin-mediated inhibition of glial apoE secretion should be considered among them.
[11]
[12]
[13]
[14]
Acknowledgements [15]
We thank Merck for providing the alendronate. Eli Lilly and Co funded this research.
References [1] K.R. Bales, T. Verina, R.C. Dodel, Y. Du, L. Alstiel, M. Bender, P. Hyslop, E.M. Johnstone, S.P. Little, D.J. Cummins, P. Piccardo, B. Ghetti, S.M. Paul, Lack of apolipoprotein E dramatically reduces amyloid, Nature Genet. 17 (1997) 263–264. [2] K.R. Bales, T. Verina, D.J. Cummins, Y. Du, R.C. Dodel, J. Saura, C.E. Fishman, C.A. DeLong, P. Piccardo, V. Petegnief, B. Ghetti, S.M. Paul, Apolipoprotein E is essential for amyloid deposition in the APPV 717F transgenic mouse model of Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 96 (1999) 15233–15238. [3] K.R. Bales, Y. Du, D. Holtzman, B. Cordell, S.M. Paul, Neuroinflammation and Alzheimer’s disease: critical roles for cytokine / Ab-induced glial activation, NF-kB, and apolipoprotein E, Neurobiol. Aging 21 (2000) 427–432. [4] S.K. Basu, J.L. Goldstein, M.S. Brown, Independent pathways for secretion of cholesterol and apolipoprotein E by macrophage, Science 219 (1983) 871–873. [5] U. Beffert, M. Danik, P. Krzywkowski, C. Ramassamy, F. Berrada, J. Poirier, The neurobiology of apolipoproteins and their receptors in the CNS and Alzheimer’s disease, Brain Res. Rev. 27 (1998) 119–142. [6] J.K. Boyles, R.E. Pitas, E. Wilson, R.W. Mahley, J.M. Taylor, Apolipoprotein E associates with astrocytic glia of the central nervous system and with nonmyelinating glia of the peripheral nervous system, J. Clin. Invest. 76 (1985) 1501–1513. [7] E.H. Corder, A.M. Saunders, W.J. Strittmatter, D.E. Schmechel, P.C. Gaskell, G.W. Small, A.D. Roses, J.L. Haines, M.A. Pericak-Vance, Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families, Science 261 (1993) 921–923. [8] J. Davignon, R.E. Gregg, C.F. Sing, Apolipoprotein E polymorphism and atherosclerosis, Atherosclerosis 8 (1988) 1–12. [9] M. Endres, U. Laufs, Z. Huang, T. Nakamura, P. Huang, M.A. Moskowitz, J.K. Liao, Stroke protection by 3-hydroxy-3methylglutaryl (HMG)-CoA reductase inhibitors mediated by endothelial nitric oxide synthase, Proc. Natl. Acad. Sci. USA 95 (1998) 8880–8885. [10] K. Fassbender, M. Simons, C. Bergmann, M. Stroick, D. Lutjohann, P. Keller, H. Runz, S. Kuhl, T. Bertsch, K. von Bergmann, M. Hennerici, K. Beyreuther, T. Hartmann, Simvastatin strongly reduces levels of Alzheimer’s disease beta-amyloid peptides Abeta42
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
and Abeta40 in vitro and in vivo, Proc. Natl. Acad. Sci. USA 98 (2001) 5856–5861. J.E. Fisher, M.J. Rogers, J.M. Halasy, S.P. Luckman, D.E. Hughes, P.J. Masarachia, G. Wesolowski, R.G.G. Russell, G.A. Rodan, A.A. Reszka, Alendronate mechanism of action: geranylgeraniol, an intermediate in the mevalonate pathway, prevents inhibition of osteoclast formation, bone resorption, and kinase activation in vitro, Proc. Natl. Acad. Sci. USA 96 (1999) 133–138. A.M. Garcia, C. Rowell, K. Ackermann, J.J. Kowalczyk, M.D. Lewis, Peptidomimetic inhibitors of Ras farnesylation and function in whole cells, J. Biol. Chem. 268 (1995) 18415–18418. M. Gearing, H. Mori, S.S. Mirra, Abeta-peptide length and apolipoprotein E genotype in Alzheimer disease, Ann. Neurol. 39 (1996) 395–399. W.S.T. Griffin, J.G. Sheng, M.C. Royston, S.M. Gentleman, J.E. McKenzie, D.I. Graham, G.W. Roberts, R.E. Mrak, Glial–neuronal interactions in Alzheimer’s disease: the potential role of a ‘cytokine cycle’ in disease progression, Brain Pathol. 8 (1998) 65–72. J. Herz, The LDL receptor gene family: (un)expected signal transducers in the brain, Neuron 29 (2001) 571–581. D.M. Holtzman, K.R. Bales, T. Tenkova, A.M. Fagan, M. Parsadanian, L.J. Sartorius, B. Mackey, J. Olney, D. McKeel, D. Wozniak, S.M. Paul, Apolipoprotein E isoform-dependent amyloid deposition and neurite degeneration in a mouse model of Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 97 (2000) 2892–2897. G.P. Jarvik, E.M. Wijsman, W.A. Kukull, G.D. Schellenberg, C. Yu, E.B. Larson, Interactions of apolipoprotein E genotype, total cholesterol level, age and sex in prediction of Alzheimer disease: a case-control study, Neurology 45 (1995) 1092–1096. H. Jick, G.L. Zornberg, S.S. Jick, S. Seshadri, D.A. Drachman, Statins and the risk of dementia, Lancet 356 (2000) 1627–1631. B. Kwak, F. Mulhaupt, S. Myit, F. Mach, Statins as a newly recognized type of immunomodulator, Nature Med. 6 (2000) 1399– 1402. M.J. LaDu, M.T. Falduto, A.M. Manelli, C.A. Reardon, G.S. Getz, D.E. Frail, Isoform-specific binding of apolipoprotein E to betaamyloid, J. Biol. Chem. 269 (1994) 23403–23406. M.J. LaDu, C.A. Reardon, L.J. Van Eldik, A.M. Fagan, G. Bu, D.M. Holtzman, G.S. Getz, Lipoproteins in the central nervous system, Ann. NY Acad. Sci. 903 (2000) 167–175. E.C. Lerner, Y. Qian, M.A. Blaskovich, R.D. Fossum, A. Vogt, J. Sun, A.D. Cox, C.J. Der, A.D. Hamilton, S.M. Sebti, Ras CAAX peptidomimetic FTI-277 selectively blocks oncogenic Ras signaling by inducing cytoplasmic accumulation of inactive Ras–Raf complexes, J. Biol. Chem. 270 (1995) 26802–26806. J. Ma, A. Wee, H.B. Brewer, S. Das, H. Potter, Amyloid-associated proteins alpha-1-antichymotrypsin and apolipoprotein E promote assembly of Alzheimer beta-protein into filaments, Nature 372 (1994) 92–94. V. Manne, C.S. Ricca, J.G. Brown, A. Tuomari, N. Yan, D.V. Patel, R. Schmidt, M.J. Lynch, J. Ciosek, J.M. Carboni, S. Robinson, E.M. Gordon, M. Barbacid, B.R. Seizinger, S.A. Biller, Ras farnesylation as a target for novel antitumor agents: potent and selective farnesyl diphosphate analog inhibitors of farnesyltransferase, Drug Dev. Res. 34 (1995) 121–137. A. Naidu, R. Catalano, K. Bales, S. Wu, S.M. Paul, S.M.B. Cordell, Conversion of brain apolipoprotein E to an insoluble form in a mouse model of Alzheimer disease, NeuroReport 12 (2001) 1265– 1270. M. Nakai, T. Kawamata, K. Maeda, C. Tanaka, Expression of apolipoprotein E mRNA in rat microglia, Neurosci. Lett. 211 (1996) 41–44. J. Naslund, J. Thyberg, L.O. Tjernberg, C. Wernstedt, A.R. Karlstrom, N. Bogdanovic, S.E. Gandy, L. Lannfelt, L. Terenius, C. Nordstedt, Characterization of stable complexes involving apolipoprotein E and the amyloid beta peptide in Alzheimer’s disease brain, Neuron 15 (1995) 219–228.
A. Naidu et al. / Brain Research 958 (2002) 100–111 [28] I.L. Notkola, R. Sulkava, J. Pekkanen, T. Erkinjuntti, C. Ehnholm, P. Kivinen, J. Tuomilehto, A. Nissinen, Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer’s disease, Neuroepidemiology 17 (1998) 14–20. [29] K. Pahan, F.G. Sheikh, A.M.S. Namboodiri, I. Singh, Lovastatin and phenylacetate inhibit the induction of nitric oxide synthase and cytokines in rat primary astrocytes, microglia, and macrophages, J. Clin. Invest. 100 (1997) 2671–2679. [30] H.-J. Park, J.B. Galper, 3-Hydroxy-3-methylglutaryl CoA reductase inhibitors up-regulate transforming growth factor-beta signaling in cultured heart cells via inhibition of geranylgeranylation of RhoA GTPase, Proc. Natl. Acad. Sci. USA 96 (1999) 11525–11530. [31] B. Permanne, C. Perez, C. Soto, B. Frangione, T. Wisniewski, Detection of apolipoprotein E / dimeric soluble amyloid beta complexes in Alzheimer’s disease brain supernatants, Biochem. Biophys. Res. Commun. 240 (1997) 715–720. [32] R.E. Pitas, J.K. Boyles, S.H. Lee, D. Foss, R.W. Mahley, Astrocytes synthesize apolipoprotein E and metabolize apolipoprotein E-containing lipoproteins, Biochim. Biophys. Acta 917 (1987) 148–161. [33] J. Poirier, Apolipoprotein E and Alzheimer’s disease: a role in amyloid catabolism, Ann. NY Acad. Sci. 924 (2000) 81–90. [34] D. Quon, Y. Wang, R. Catalano, J. Marian Scardina, K. Murakami, B. Cordell, Formation of b-amyloid protein deposits in brains of transgenic mice, Nature 352 (1991) 239–241. [35] L.M. Refolo, B. Malester, J. LaFrancois, T. Thomas-Bryant, R. Wang, G.S. Tint, K. Sambamurti, K.E. Duff, M.A. Pappolla, Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model, Neurobiol. Dis. 7 (2000) 321– 331. [36] L.M. Refolo, M.A. Pappolla, J. LaFrancois, J. Malester, S.D. Schmidt, T. Thomas-Bryant, G.S. Tint, R. Wang, M. Mercken, S.S. Petanceska, K.E. Duff, A cholesterol-lowering drug reduces betaamyloid pathology in a transgenic mouse model of Alzheimer disease, Neurobiol. Dis. 8 (2001) 890–899. [37] C. Russo, D. Angelini, D. Dapino, A. Piccini, G. Piombo, G. Schettini, S. Chen, J.K. Teller, D. Zaccheo, P. Gambetti, M. Tabaton, Opposite roles of apolipoprotein E in normal brains and in Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 95 (1998) 15598– 15602. [38] A.M. Saunders, Apolipoprotein E and Alzheimer’s disease: an update on genetic and functional analyses, J. Neuropathol. Exp. Neurol. 59 (2000) 751–758. [39] J.G. Sheng, R.E. Mrak, W.T.S. Griffin, Apolipoprotein E distribution among different plaque types in Alzheimer disease: implications for its role in plaque progression, Neuropathol. Appl. Neurol. 22 (1996) 334–341. [40] M. Simons, P. Keller, B. De Strooper, K. Beyreuther, C.G. Dotti, K. Simons, Cholesterol depletion inhibits the generation of betaamyloid in hippocampal neurons, Proc. Natl. Acad. Sci. USA 95 (1998) 6460–6464.
111
[41] M. Sinensky, Functional aspects of polyisoprenoid protein substituents: roles in protein–protein interaction and trafficking, Biochim. Biophys. Acta 1529 (2000) 203–209. [42] A.J.C. Slooter, M.-X. Tang, C.M. van Duijn, Y. Stern, A. Ott, K. Bell, M.M.B. Breteler, C. van Broeckhoven, T.K. Tatemichi, B. Tycko, A. Hofman, A.R. Mayeux, Apolipoprotein E epsilon-4 and the risk of dementia with stroke. A population-based investigation, J. Am. Med. Assoc. 277 (1997) 818–821. [43] D.A. Snowdon, L.H. Greiner, J.A. Mortimer, K.P. Riley, P.A. Greiner, W.R. Markesbery, Brain infarction and the clinical expression of Alzheimers disease. The nun study, J. Am. Med. Assoc. 277 (1997) 813–817. [44] D.L. Sparks, S.W. Scheff, J.C. Hunsaker, H. Liu, T. Landers, D.R. Gross, Induction of Alzheimer-like beta-amyloid immunoreactivity in the brains of rabbits with dietary cholesterol, Exp. Neurol. 126 (1994) 88–94. [45] D.L. Sparks, Coronary artery disease, hypertension, ApoE, and cholesterol: a link to Alzheimer’s disease?, Ann. NY Acad. Sci. 826 (1997) 128–146. [46] L. Stoppini, P.A. Buchs, D. Muller, A simple method for organotypic culture of nervous tissue, J. Neurosci. Methods 37 (1991) 173–182. [47] S.D. Styren, M.I. Kamboh, J. Dekosky, Expression of differential immune factors in temporal cortex and cerebellum: the role of alpha-1-antichymotrypsin, apolipoprotein E, and reactive glia in the progression of Alzheimer disease, Comp. Neurol. 396 (1998) 511– 520. [48] J. Sun, Y. Qian, A.D. Hamilton, S.M. Sebti, Ras CAAX peptidomimetic FTI-276 selectively blocks tumor growth in nude mice of a human lung carcinoma with k-Ras mutation and p53 deletion, Cancer Res. 55 (1995) 4243–4247. [49] J. Sun, Y. Qian, A.D. Hamilton, S.M. Sebti, Both farnesyltransferase and geranylgeranyltransferase I inhibitors are required for inhibition of oncogenic k-Ras prenylation but each alone is sufficient to suppress human growth in nude mouse xenografts, Oncogene 16 (1998) 1467–1473. [50] T. Wisniewski, E.M. Castano, A. Goldbek, T. Vogel, B. Frangione, Acceleration of Alzheimer’s fibril formation by apolipoprotein E in vitro, Am. J. Pathol. 145 (1994) 1030–1035. [51] G. Weitz-Schmidt, K. Welzenbach, V. Brinkmann, T. Kamata, J. Kallen, C. Bruns, S. Cottens, Y. Takada, U. Hommel, Statins selectively inhibit leukocyte function antigen-1 by binding to a novel regulatory integrin site, Nature Med. 7 (2001) 687–692. [52] B. Wolozin, W. Kellman, P. Ruosseau, G.G. Celesia, G. Siegel, Decreased prevalence of Alzheimer disease with 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors, Arch. Neurol. 57 (2000) 1439–1443. [53] Q. Xu, C. Cyras, D.A. Sanan, B. Cordell, Isolation and characterization of apolipoproteins from murine microglia, J. Biol. Chem. 275 (2000) 31770–31777.
Research report
Secretion of apolipoprotein E by brain glia requires protein prenylation and is suppressed by statins Asha Naidu, Qiang Xu 1 , Rosanne Catalano, Barbara Cordell* Scios Inc., 820 West Maude Avenue, Sunnyvale, CA 94085, USA Accepted 19 August 2002
Abstract Apolipoprotein E (ApoE) genotype modulates the risk of Alzheimer’s disease. ApoE has been shown essential for amyloid b-peptide fibrillogenesis and deposition, a defining pathological feature of this disease. Because astrocytes and microglia represent the major source of extracellular apoE in brain, we investigated apoE secretion by glia. We determined that protein prenylation is required for apoE release from a continuous microglial cell line, primary mixed glia, and from organotypic hippocampal cultures. Using selective protein prenylation inhibitors, apoE secretion was found to require protein geranylgeranylation. This prenylation involved a protein critical to apoE secretion, not apoE proper. ApoE secretion could also be suppressed by inhibiting synthesis of mevalonate, the precursor to both types of protein prenylation, using hydroxyl-3-methylglutaryl coenzyme A reductase inhibitors (statins). Recent reports have described the beneficial effects of statins on the risk of dementia. Our finding that protein geranylgeranylation is required for apoE secretion in the brain parenchyma provides another contributing mechanism to explain the effective properties of statins against the development of dementia. In this model, statin-mediated inhibition of mevalonate synthesis, an essential reaction in forming geranylgeranyl lipid, would lower extracellular levels of parenchymal apoE. Because apoE has been found necessary for plaque development in transgenic models of Alzheimer’s disease, suppressing apoE secretion by statins could reduce plaques and, in turn, improve cognitive function. 2002 Elsevier Science B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: Alzheimer’s Keywords: ApoE; Alzheimer; Glia; Prenylation
1. Introduction Multiple genetic and environmental factors interact to generate Alzheimer’s disease (AD). Genetic defects, including both causative and susceptibility genes, have been identified. One common susceptibility gene is that encoding the type 4 isoform of apolipoprotein E (APOE), a cholesterol transport protein [5,7,38]. Epidemiological studies suggest a relationship between cholesterol levels and AD. Individuals carrying the APOE-´4 allele have elevated plasma cholesterol levels [17,28] and manifest an *Corresponding author. Tel.: 11-650-493-2715; fax: 11-925-5603901. E-mail address: [email protected] (B. Cordell). 1 Current address: Osel Inc., 1800 Wyatt Drive, Santa Clara, CA 95054, USA.
increased risk of cardiovascular disease [8]. Correspondingly, cardiovascular disease has also been shown to predispose to AD, vascular dementia, and stroke [42,43]. This association between cholesterol and AD is further supported by animal studies. Sparks et al. found that rabbits fed a high fat cholesterol diet developed amyloid b-peptide (Ab) immunoreactive deposits in the frontal cortex and hippocampus, the brain regions displaying pathology in AD [44,45]. Transgenic studies have also demonstrated a direct correlation between elevated cholesterol levels and increased Ab plaque burden upon increased dietary cholesterol intake [36]. In addition to the role of apoE in cholesterol transport, this protein has also been implicated in AD pathogenesis by other mechanisms. The APOE-´4 allele not only increases the risk of developing AD but also the plaque burden in a gene dosage-dependent manner [13], as well as
0006-8993 / 02 / $ – see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03480-7
A. Naidu et al. / Brain Research 958 (2002) 100–111
Ab production [33]. Neuropathological analyses have revealed extracellular apoE immunoreactivity associated with Ab amyloid plaques in the parenchyma and in the cerebrovasculature, although not with non-fibrillar Ab deposits [14,39,47]. Moreover, stable complexes between apoE and Ab have been isolated from AD brain [27,31,37]. These in vivo observations, plus the high avidity binding of apoE to Ab in vitro promoting fibrillogenesis [20,23,50], suggest that apoE promotes plaque formation through the conversion of Ab to a fibrillar structure. This concept is further supported by transgenic crosses which have shown the absence of fibrillar plaques in transgenic mice on an APOE null background, in contrast to abundant plaques in the context of APOE alleles [1,2]. Second to the liver, the brain produces the largest quantities of apoE. Astrocytes and microglia are the major sources of extracellular apoE [6,26,32,53]. Glia release apoE as free protein and as a component of lipoprotein particles [32,53]. Because apoE appears to play a critical role in AD pathogenesis by modulating cholesterol levels and Ab plaque formation, we sought to better define the molecular factors required for apoE secretion. Our initial approach was to assess the effects of hydroxyl-3methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) on apoE secretion from cultured glial systems. Statins are commonly prescribed to reduce plasma cholesterol levels and consequently the risk of cardiovascular disease. Our reasons for testing statins in this system are several. First, there is growing evidence for additional benefits of statins that cannot be completely explained by their lipid-lowering effects. Indeed, studies have indicated immunomodulatory actions of this drug class [19,51]. Clinical studies involving transplant recipients indicate possible immunosuppressive effects of statins. Additionally, anti-inflammatory actions of statins have been reported in which NF-kB activation and subsequent induction of nitric oxide synthase and cytokines in isolated macrophages and glia could be prevented by lovastatin [29]. In this context, it is noteworthy that we reported that pro-inflammatory agents, including Ab and interleukin-6, can promote the secretion of glial apoE via NF-kB activation [3], providing a rationale for testing statins effects on apoE secretion. Second, a lower prevalence of AD and a lower risk for dementia in general have been observed with individuals taking statins [18,52]. Whether this effect is due solely to a reduction in cholesterol levels or whether statins affect Ab plaque formation, or both, is unclear. Third, HMG-CoA reductase inhibitors have been shown capable of altering both gene transcription and signaling [9,30]. Here, we show that statins block apoE release from glial cell lines, primary mixed glia, and organotypic hippocampal cultures. The inhibition of glial apoE secretion by statins was determined to be due to prenylation of an unidentified protein required for apoE secretion, not apoE itself.
101
2. Materials and methods
2.1. Reagents Lovastatin, mevastatin, FTI-277, FTI-276, GGTI-287, GGTI-286, FPT-II, FPT-III, and FTase-I were purchased from CalBiochem. Mevalonate, geranylgeranyl pyrophosphate, and farnesyl pyrophosphate were purchased from Sigma. Alendronate (4-amino-1-hydroxybutylidene bisphosphonate) was a gift from Merck Research Laboratories.
2.2. Cell and tissue culture The growth and propagation of the BV-2 murine microglial cell line has been described previously [53]. Primary mixed glia were prepared from JU [34] or bAPP V717F mice [1,2] using cortices from 1-day to 3-day old pups. Dissected cortices, from which the meninges were removed, were washed twice with Hank’s balanced salt solution containing 20 mM Ca 21 and 20 mM Mg 21 (HBSS with Ca 21 Mg 21 ). Tissue was trypsinized with 0.05% trypsin in HBSS Ca 21 Mg 21 containing 50 mM EDTA, 100 units / ml penicillin and 100 mg / ml streptomycin for 20 min at room temperature, after which the trypsin solution was removed and the tissue rinsed twice with HBSS Ca 21 Mg 21 . The cortices were triturated in growth medium (Delbecco’s minimal essential medium with high glucose, 1 mM sodium pyruvate, 2 mM Lglutamine, 10% fetal bovine serum, 50 units / ml penicillin and 50 mg / ml streptomycin) then seeded into flasks at a density of six cortices per T75 flask. The cultures were incubated in 5% CO 2 at 37 8C. Approximate cell populations in the isolated primary glial cultures were assessed using immunohistochemistry. Staining for glial acidic fibrillary protein (antibody purchased from Sigma) to identify astrocytes indicated that this glial cell type represented 80–90% of the total culture over a 10-day period. Staining with F4 / 80, an antibody marker for microglia (Serotec; Raleigh, NC, USA) revealed low levels of microglia, roughly 5–8% of the culture. Organotypic hippocampal slice cultures were prepared essentially as described by Stoppini et al. [46]. Briefly, brains from 8 day old pups of the JU strain or bAPP V717F transgenic mouse were dissected into Minimal Essential medium containing 10 mM Tris–HCl, pH 7.2, supplemented with penicillin and streptomycin. Hippocampi were isolated under a dissecting microscope and 400 mm sections made using a MacIllwain tissue chopper. Sections were separated by vigorously swirling the culture dish. Four sections were placed on Millipore culture inserts set into the well of a six-well tissue culture plate containing 1.2 ml culture medium (Delbecco’s minimal essential medium with high glucose, 12.5 mM HEPES, 25% HBSS with Ca 21 Mg 21 , 1 mM L-glutamine, 25% horse serum, 50 units / ml penicillin
102
A. Naidu et al. / Brain Research 958 (2002) 100–111
and 50 mg / ml streptomycin) and incubated at 37 8C in 5% CO 2 .
2.3. ApoE measurement Conditioned media and lysates were assayed for apoE levels using Western blot or ELISA. For Western blot analysis, samples were electrophoresed on reducing 12% SDS–PAGE. After membrane transfer, apoE was detected with a rabbit anti-murine apoE antiserum purchased from Biodesign International (Kennebunk, ME, USA). The antigen antibody reaction was visualized by using a secondary antibody conjugated with horse-radish peroxidase and enhanced chemiluminesence detection reagents (Amersham Pharmacia Biotech). We previously described quantifying rodent apoE by ELISA [25].
2.4. Inhibitor treatment and lipid rescue BV2, primary glia, and hippocampal slice cultures were maintained in serum-free medium during exposure to inhibitors. A concentrated stock solution of each inhibitor was prepared and diluted into serum-free medium to the designated final concentration. Medium containing inhibitor was added to cultures washed free of serum and incubated at 37 8C in 5% CO 2 for the times indicated in each experiment. After treatment, the conditioned medium was collected, centrifuged at low speed to remove cellular debris, and assayed for apoE. Treated cells were washed twice with phosphate buffered saline, lysed by three cycles of freeze–thaw at 280 and 27 8C in serum-free culture medium, and then assayed for cell-associated apoE. For experiments involving rescue of prenylation inhibition, mevalonate, farnesyl-pyrophosphate (FPP) or geranylgeranyl-pyrophosphate (GGPP), both solubilized in culture medium, cultures were pre-treated overnight with 10 mM of each intermediate. Cultures were changed to serum-free medium containing 5 mM lovastatin and 10 mM GGPP or FPP, incubated for 5 h, and media collected for analysis. To deliver GGPP and FPP using liposomes, liposomes were prepared by mixing 5 mmol dipalmitoylphosphatidylcholine with 200 mg GGPP or FPP and drying by rotary evaporation. The dried mixtures were resuspended in phosphate buffered saline, warmed to 50 8C and sonicated for 30 s. Control liposomes containing only dipalmitoylphosphatidylcholine were similarly prepared.
2.5. Toxicity assessment For cell cultures, toxicity was measured by the 3-(4,5dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrozolium salt (MTS) assay using a commercially available kit (CellTiter 96 AQ ueus NonRadioactive Cell Proliferation Assay, Promega, Madison, WI, USA). This is a colorimetric assay for determining the number of viable cells whereby the MTS is bioreduced to
formazan by dehydrogenase enzymes in metabolically active cells. In brief, the MTS reagent was added to cells after treatment with the various inhibitors. The cells were incubated for 1–2 h at 37 8C in the presence of the MTS reagent. After appropriate color development, absorbency was read at 490 nm. For hippocampal slice cultures, toxicity was measured by lactate dehydrogenase (LDH) release assay using a commercially available kit (CytoTox 96 NON-radioactive Cytotoxicity Assay, Promega). The kit quantitatively measures LDH release upon cell lysis. Aliquots (75 ml) of slice culture medium were assayed in the LDH assay after the slices were treated with various inhibitors. For a positive control for LDH release, slice cultures were treated with 10% ethanol for 24, 48, and 72 h and the conditioned medium was assayed in the LDH assay. To determine the total amount of LDH present in the slices, lysates from untreated slices were prepared by freezing at 280 8C and thawing the slices three times in culture medium and assaying the lysates in the LDH assay.
3. Results Microglia and apoE have each been implicated in the pathogenesis of senile plaques in AD. We recently showed that cultures of BV2 murine microglial cells release apoE [53] and that apoE release could be enhanced by inflammatory stimuli, including Ab [3]. In the present study, we sought to define the molecular process governing apoE production by BV2 microglia and by primary glial cultures. We investigated the effects of statins on apoE secretion by BV2 microglia for two reasons. ApoE secretion by BV2 cells utilizes a NF-kB-dependent mechanism identical to that for the release of inflammatory cytokines [3] (our unpublished observations). HMG-CoA reductase inhibitors (statins) have been reported to inhibit cytokine secretion from rat primary astrocytes, microglia and macrophages [29].
3.1. Statins block apoE release from cultured glia Using Western blot analysis, cultures of BV2 cells in serum-free medium release apoE which accumulates in a time-dependent manner (Fig. 1A). The observed doublet of secreted apoE immunoreactivity represents sialyated and non-sialyated forms of apoE [53]. Accumulation of apoE in the conditioned medium was also evident when assaying samples by ELISA (Fig. 1B). Cell-associated levels, however, were relatively constant after an initial increase (Fig. 1C). ApoE production by BV2 cultures after treatment with the HMG-CoA reductase inhibitors, mevastatin and lovastatin, was assessed. Both statins inhibited apoE secretion in a dose-dependent fashion (Fig. 2A). Approximately 85% of the apoE released into the medium was inhibited at the highest dose tested. Identical results were obtained by Western blot (Fig. 2C). Cell-associated levels
A. Naidu et al. / Brain Research 958 (2002) 100–111
103
Fig. 1. Production of apoE by BV-2 microglial cells. Cultures of |10 5 BV-2 cells in serum-free medium were sampled for apoE at the times indicated. Conditioned media were used to assess secreted apoE and cell lysates were tested for cell-associated apoE by Western blot and ELISA as detailed in Materials and methods. The data represent one of four identical experiments. (A and B) Time-dependent release of apoE into medium using Western blot and ELISA detection, respectively. (C) Cell-associated apoE.
were unperturbed, indicating that statins were not depleting intracellular apoE pools (Fig. 2B). Statin suppression of apoE release from BV2 cells could not be attributed to a general suppression of protein secretion from the cells, as silver staining of the conditioned medium showed no gross differences in protein levels or pattern (data not shown). This effect was not due to toxicity as control and statintreated cultures were equally viable. To further verify that the statin effect on apoE release was authentic, we used mevalonate to rescue the inhibition. Since mevalonate is the product of HMG-CoA reductase, addition of mevalonate to the cells should overcome the statin inhibition. BV2 cells were treated with mevastatin in the presence or absence of 100 mM exogenous mevalonate. Secreted apoE was restored to levels equivalent to that of the untreated control medium when mevalonate was added to the mevastatin-treated cultures (Fig. 2D). We next examined the effects of statins on apoE release from primary murine mixed glial cultures. The cultures were assessed for the approximate proportion of astrocytes and microglia using immunocytochemistry that indicated that the cultures were predominantly populated by astrocytes with minor representation of microglia (10–15% of all cells). Similar to results with BV2 microglial cells, primary mixed glia release apoE in a time-dependent manner (Fig. 3A). This is in agreement with previous reports indicating that apoE is released by both microglia and astrocytes [6,26,32,53]. Exposing the cultures to 10 mM mevastatin resulted in partial inhibition of apoE as assessed by ELISA (Fig. 3B) or by Western blot (Fig. 3C).
3.2. Geranylgeranylation required for apoE release from glial cells The production of mevalonate by HMG-CoA reductase provides the precursor leading to the formation of cholesterol, dolichol, ubiquinone, as well as isoprenoids (Fig. 4). Isoprenoids can be covalently linked to proteins, a process referred to as protein prenylation. Protein prenylation is recognized as a mechanism to promote membrane interactions and biological activities of a variety of cellular proteins [41]. Because statins act at an early step in the mevalonate pathway, we wanted to determine if the inhibition of glial apoE secretion was due to inhibition of cholesterol preventing the formation of apoE-containing lipoprotein particles or, alternatively, was a result of blocking the prenylation of a protein required for apoE secretion. This latter possibility would be a downstream effect of inhibiting HMG-CoA reductase with statins. To address this issue, we tested whether lovastatin inhibition of apoE secretion from BV2 microglial cells could be rescued with farnesyl-pyrophosphate (FPP) or geranylgeranyl-pyrophosphate (GGPP), specific intermediates to farnesylation and geranylgeranylation, respectively. Rescue with mevalonate served as a control. Exposure of treated BV2 cells to 10 mM FPP or 10 mM GGPP indicated that GGPP, but not FPP, rescued |40% of the inhibition of apoE secretion generated by lovastatin. To better characterize the specific sterol requirement for apoE release by glia, we investigated the effects of wellcharacterized protein prenylation inhibitors. Several far-
104
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 2. Statin inhibition of apoE release by BV-2 microglia and rescue by mevalonate. Approximately 10 5 BV-2 cells were pretreated for 16–24 h in the presence of serum with the doses of mevastatin or lovastatin indicated. After pretreatment, cells were placed in serum-free medium and incubated for 4 h. ApoE, released into medium or remaining cell-associated, was measured by Western blot and ELISA. For mevalonate rescue, 100 mM exogenous mevalonate was included during the treatment period. The figure shows one representative experiment of several performed. (A and C) Dose-dependent inhibition of apoE secretion by mevastatin and lovastatin, respectively. (B) Unchanged cell-associated apoE levels during statin treatment. (D) Rescue of statin-induced inhibition of apoE secretion by mevalonate.
nesyl protein transferase inhibitors, FTase-I, FPT-II and FPT-III, were tested for inhibition of apoE secretion. The inhibitors differ in structure and all are reported to be cell permeable. FPT-III is a prodrug of FPT-II. All inhibitors were assessed using concentrations previously shown to be inhibitory in other cell systems [12,24]. Dose-dependent inhibition of apoE secretion from BV2 cells by FPT-II and FPT-III was observed (Fig. 5A and C). In contrast, the FTase-I inhibitor did not block apoE release (Fig. 5B). Primary murine mixed glia were also tested for response to FPT-II and FTase-I farnesyl transferase inhibitors. Similar to the results obtained with BV2 microglial cultures, FPT-
II treatment resulted in reduced apoE release, albeit partial, reaching a maximum of |50%, and FPTase-1 was without effect up to 200 mM (Fig. 6). The active doses of FPT-II and FPT-III in the glial apoE release assay were in the 100 mM range which have been reported to inhibit both farnesyl transferase and geranylgeranyl transferase [24]. To address which type of protein prenylation is required for apoE secretion, we tested more selective inhibitors. Two peptidomimetic farnesyl transferase inhibitors, FTI-276 and FTI-277, have been described as 100-fold more selective for farnesyl transferase over geranylgeranyl transferase. Likewise, two
A. Naidu et al. / Brain Research 958 (2002) 100–111
105
Fig. 3. Secretion of apoE by primary murine mixed glia and its inhibition by statins. Primary glial cultures containing |10 5 cells were assessed for apoE release in the presence or absence of statins. For statin treatment, 10 mM mevastatin or lovastatin was applied to the cultures in complete medium for 16–24 h, after which the cultures were placed in serum-free medium for 4–5 h containing 10 mM statin. Media were scored for apoE using Western blot and ELISA. The data show one of three separate experiments. (A) Time-dependent secretion of apoE. (B and C) Statin inhibition of apoE secretion.
geranylgeranyl transferase I inhibitors, GGTI-286 and GGTI-287, are selective inhibitors. GGTI-286 is a more cell-permeable derivative of GGTI-287. GGTI-286 is described as 25-fold more selective for geranylgeranyl transferase versus farnesyl transferase [22,48,49]. All four inhibitors were tested with BV2 microglia. Little or no inhibition of apoE release was seen with FTI-276, FTI-277 and GGTI-287, unlike GGTI-286 which showed partial inhibition (Fig. 7A–E). Extended treatment times resulted in greater inhibition up to 80%. The inhibitory effect of GGTI-286 on apoE release from primary mixed murine glia was similar (Fig. 7F). The poor cell permeability of GGTI-287 compared to its methyl ester derivative, GGTI-286, reconciles the different responses obtained with this pair of geranylgeranyl transferase inhibitors. Last, we tested alendronate, a nitrogen-containing bisphosphonate that has been reported to inhibit the production of geranylgeraniol [11]. The target of alendronate appears to be the enzyme involved in the conversion of mevalonate to GGPP [11]. Because alendronate has a different mode of action compared to the inhibitors tested thus far, we examined its influence on apoE secretion from BV2 cells and found that it was inhibitory in the absence
of toxicity (Fig. 7E). These results, together with the observed inhibition by GGTI-286, the lack of inhibition by FPT-276 and FPT-277, and the partial rescue by exogenous GGPP but not FPP, are consistent with the requirement for protein geranylgeranylation in the secretory machinery used for apoE release from glia.
3.3. Protein prenylation inhibition in hippocampal cultures blocks apoE release Because of the role of apoE in the pathogenesis of AD, we were interested in determining whether protein prenylation is required for apoE secretion by hippocampal tissue. To do so, we established conditions yielding quantitative and consistent production of apoE from murine hippocampal slice cultures. Inhibitors active in suppressing apoE secretion from cultured glia (lovastatin, FPT-II, GGTI-286, and alendronate) were each tested in the hippocampal slice culture system. Three media harvests were made and evaluated for apoE levels at 24, 48, and 72 h of cumulative exposure. Exposure to each prenylation inhibitor resulted in a time-dependent reduction of apoE in the conditioned medium over the 72 h treatment period compared to untreated control samples (Fig. 8). In addition, lovastatin
106
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 4. Protein prenylation pathway showing inhibitor targets.
suppression of apoE release in the hippocampal slice cultures could be rescued by addition of 200 mM mevalonate (data not shown). These data indicate that protein prenylation, specifically geranylgeranylation, is essential to the secretory machinery used to release apoE from cultured hippocampi, a target tissue of AD plaque pathology.
4. Discussion ApoE contributes to the development of AD in a number of ways. The APOE-´4 allele is a genetic risk modifier that decreases the age of onset in a dose-dependent manner [5,7,38]. The pathophysiological consequence of APOE-´4 is an increased Ab plaque burden in humans and in transgenic mice [13,16]. Noteworthy is the absence of Ab plaques in APP V717F transgenic mice on an APOE null background [1,2]. From in vitro studies, apoE appears to facilitate Ab fibril formation [20,23,50]. However, the normal physiological function of apoE in regulating plasma lipid and cholesterol metabolism, serving both as a
ligand for lipoprotein receptors and as a particle for lipid transport [5,38], may also contribute to AD. Since neurons and glia express a variety of receptors in the LDL family and express lipoproteins [15], apoE may function in a similar capacity in the brain parenchyma as it does in the periphery. The primary source of apoE in brain is glia. In vitro primary murine astrocytes secrete apoE as a component of a HDL-like particle [6,21,32]. In contrast, the BV2 murine microglial cell line releases lipid-poor protein aggregates of apoE [53]. Microglia are probably similar to peripheral macrophages which independently secrete apoE and cholesterol but, upon release, associate to form a particle [4]. In this study, we employed cultured BV2 cells, primary murine glia (predominately astrocytes), and murine organotypic hippocampal cultures. In each of these culture systems, we demonstrated a time-dependent release and accumulation of apoE into the medium. We found that inhibition of mevalonate production by HMG-CoA reductase inhibitors (lovastatin and mevastatin), general inhibition of protein prenylation with non-selective inhibitors (FPT-II, FPT-III, and alendronate), and inhibition of
A. Naidu et al. / Brain Research 958 (2002) 100–111
107
Fig. 5. Inhibition of apoE secretion by BV-2 cells after treatment with protein prenylation inhibitors. Cultures of |10 5 BV-2 cells in serum-free medium were treated with the indicated doses of FPT-II (A), FTase-I (B), or FPT-III (C) for 4–6 h, after which the conditioned media were assayed for apoE using ELISA. The data represent one of three independent determinations.
Fig. 6. Effects of protein prenylation inhibitors on apoE secretion by primary mixed glia. Murine primary mixed glial cultures containing |10 5 cells in serum-free medium were treated with FPT-II (A) or FTase-I (B) at the doses indicated for 4 h. ApoE released into the medium during this period was determined by ELISA.
108
A. Naidu et al. / Brain Research 958 (2002) 100–111
Fig. 7. Effects of protein prenylation inhibitors on BV-2 microglia and on primary mixed glia. Glial cultures containing |10 5 cells in serum-free medium were pretreated for 1 h with the doses of prenylation inhibitors indicated, after which the cultures were moved to serum-free medium containing the same dose of inhibitor. Treatments with FTI-276, FTI-277, GGTI-286 and GGTI-287 were for 4–6 h and alendronate treatment was for 24 h. Media and cell lysates were assayed by ELISA and Western blot. The data provided are representative of three separate determinations. BV-2 microglia were treated with increasing doses of FTI-276 (A), FTI-277, (B), GGTI-286 (C), GGTI-287 (D), and alendronate (E). Murine primary mixed glia were treated with 20 mM GGTI-286 (F).
protein geranylgeranylation using selective cell-penetrant agents against geranylgeranyl transferase (GGTI-286) blocked apoE secretion. Mevalonate is the up-stream precursor to GGPP and FPP and, indeed, we found mevalonate could rescue the inhibition by these compounds. GGPP and FPP are lipids used in prenylation, a post-translation mechanism to target proteins, such as those in the Ras superfamily, to cellular membranes [41]. Based on our results, it appears that the secretory machinery used to release apoE from glial cells contains an essential geranylgeranlyated protein. The observed effects
are not due to direct prenylation of apoE or to the depletion of cholesterol. Inhibition of apoE secretion by statins and prenylation inhibitors was quantitatively different between the BV2 microglial cells and cultures where the majority of the glial population is astrocytes, such as with primary murine mixed glia and organotypic hippocampal sections. In the pure microglial system, near complete suppression of apoE release could be achieved by the active inhibitors described. This was in contrast to the cultures rich in astrocytes where inhibition by these agents was partial,
A. Naidu et al. / Brain Research 958 (2002) 100–111
109
Fig. 8. Inhibition of apoE secretion by organotypic hippocampal cultures treated with inhibitors of protein prenylation. Microtiter wells each containing three murine hippocampal tissue slices in serum-free media were treated with the protein prenylation inhibitors found to block apoE release from BV-2 and primary glia. Media were harvested and replaced every 24 h with fresh drug-containing media. Media from quadruplicate samples were assayed from 24, 48, and 72 h harvest or drug treatment periods using ELISA. ApoE values in drug-treated samples were compared to media from mock-treated cultures similarly handled. The data reflect one of three similar experiments. (A) 100 mM lovastatin. (B) 100 mM FPT-II. (C) 100 mM alendronate. (D) 100 mM GGTI-286.
ranging between 40 and 60%. The partial inhibition in the astrocyte-rich cultures was consistent irrespective of the active inhibitor used. This suggests that apoE secretion from microglia relies solely on a mechanism(s) requiring a prenylated protein, whereas astrocytes use an identical or similar secretory mechanism, in part, as well as a secretory process that does not involve protein prenylation. In the context of lowering extracellular apoE in brain as a possible therapeutic approach to mitigate Ab plaque formation, such a partial inhibition may be effective. Moderate suppression of apoE levels could limit Ab nucleation and amyloid fibril formation. Support of this concept is seen in bAPP V717F transgenic mice carrying only one APOE allele in which the plaque frequency is 80% lower compared to transgenic mice with two alleles [2]. It is of interest that HMG-CoA reductase inhibitors,
statins, in addition to their beneficial lipid-lowering effects for treating cardiovascular disease, have been shown to reduce the risk of AD [52] and dementia [18] in two recent epidemiological studies. Statins were evaluated in AD patients because of the link between cholesterol metabolism and the pathophysiology of this disease. For example, elevated plasma cholesterol levels have been show to be a risk factor for AD. APOE-´4 polymorphism increases not only the risk of AD [7] but also elevates cholesterol levels and increases the risk of cardiovascular disease [8,17,28]. Furthermore, cardiovascular disease is a risk factor for AD [42,43]. Cholesterol levels have also been found to influence bAPP processing in cell culture [10,40] as well as Ab load in vivo [10,35]. Despite these associated observations, the effects of statins on cholesterol-lowering action and on bAPP processing do not fully explain their beneficial effects on reducing the risk of dementia. Drach-
110
A. Naidu et al. / Brain Research 958 (2002) 100–111
man and colleagues [18] showed that while statins lowered the risk of developing dementia by 70%, non-statin lipidlowering agents did not show a reduced risk. Hence, the mechanism by which statins reduce the risk of developing dementia is unclear. From the data presented in this work, one explanation is that statins suppress extracellular levels of apoE, thereby reducing Ab deposition and plaque formation. While viable hypotheses have been put forward to explain the positive statin effects on the development of dementia [18,52], statin-mediated inhibition of glial apoE secretion should be considered among them.
[11]
[12]
[13]
[14]
Acknowledgements [15]
We thank Merck for providing the alendronate. Eli Lilly and Co funded this research.
References [1] K.R. Bales, T. Verina, R.C. Dodel, Y. Du, L. Alstiel, M. Bender, P. Hyslop, E.M. Johnstone, S.P. Little, D.J. Cummins, P. Piccardo, B. Ghetti, S.M. Paul, Lack of apolipoprotein E dramatically reduces amyloid, Nature Genet. 17 (1997) 263–264. [2] K.R. Bales, T. Verina, D.J. Cummins, Y. Du, R.C. Dodel, J. Saura, C.E. Fishman, C.A. DeLong, P. Piccardo, V. Petegnief, B. Ghetti, S.M. Paul, Apolipoprotein E is essential for amyloid deposition in the APPV 717F transgenic mouse model of Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 96 (1999) 15233–15238. [3] K.R. Bales, Y. Du, D. Holtzman, B. Cordell, S.M. Paul, Neuroinflammation and Alzheimer’s disease: critical roles for cytokine / Ab-induced glial activation, NF-kB, and apolipoprotein E, Neurobiol. Aging 21 (2000) 427–432. [4] S.K. Basu, J.L. Goldstein, M.S. Brown, Independent pathways for secretion of cholesterol and apolipoprotein E by macrophage, Science 219 (1983) 871–873. [5] U. Beffert, M. Danik, P. Krzywkowski, C. Ramassamy, F. Berrada, J. Poirier, The neurobiology of apolipoproteins and their receptors in the CNS and Alzheimer’s disease, Brain Res. Rev. 27 (1998) 119–142. [6] J.K. Boyles, R.E. Pitas, E. Wilson, R.W. Mahley, J.M. Taylor, Apolipoprotein E associates with astrocytic glia of the central nervous system and with nonmyelinating glia of the peripheral nervous system, J. Clin. Invest. 76 (1985) 1501–1513. [7] E.H. Corder, A.M. Saunders, W.J. Strittmatter, D.E. Schmechel, P.C. Gaskell, G.W. Small, A.D. Roses, J.L. Haines, M.A. Pericak-Vance, Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families, Science 261 (1993) 921–923. [8] J. Davignon, R.E. Gregg, C.F. Sing, Apolipoprotein E polymorphism and atherosclerosis, Atherosclerosis 8 (1988) 1–12. [9] M. Endres, U. Laufs, Z. Huang, T. Nakamura, P. Huang, M.A. Moskowitz, J.K. Liao, Stroke protection by 3-hydroxy-3methylglutaryl (HMG)-CoA reductase inhibitors mediated by endothelial nitric oxide synthase, Proc. Natl. Acad. Sci. USA 95 (1998) 8880–8885. [10] K. Fassbender, M. Simons, C. Bergmann, M. Stroick, D. Lutjohann, P. Keller, H. Runz, S. Kuhl, T. Bertsch, K. von Bergmann, M. Hennerici, K. Beyreuther, T. Hartmann, Simvastatin strongly reduces levels of Alzheimer’s disease beta-amyloid peptides Abeta42
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
and Abeta40 in vitro and in vivo, Proc. Natl. Acad. Sci. USA 98 (2001) 5856–5861. J.E. Fisher, M.J. Rogers, J.M. Halasy, S.P. Luckman, D.E. Hughes, P.J. Masarachia, G. Wesolowski, R.G.G. Russell, G.A. Rodan, A.A. Reszka, Alendronate mechanism of action: geranylgeraniol, an intermediate in the mevalonate pathway, prevents inhibition of osteoclast formation, bone resorption, and kinase activation in vitro, Proc. Natl. Acad. Sci. USA 96 (1999) 133–138. A.M. Garcia, C. Rowell, K. Ackermann, J.J. Kowalczyk, M.D. Lewis, Peptidomimetic inhibitors of Ras farnesylation and function in whole cells, J. Biol. Chem. 268 (1995) 18415–18418. M. Gearing, H. Mori, S.S. Mirra, Abeta-peptide length and apolipoprotein E genotype in Alzheimer disease, Ann. Neurol. 39 (1996) 395–399. W.S.T. Griffin, J.G. Sheng, M.C. Royston, S.M. Gentleman, J.E. McKenzie, D.I. Graham, G.W. Roberts, R.E. Mrak, Glial–neuronal interactions in Alzheimer’s disease: the potential role of a ‘cytokine cycle’ in disease progression, Brain Pathol. 8 (1998) 65–72. J. Herz, The LDL receptor gene family: (un)expected signal transducers in the brain, Neuron 29 (2001) 571–581. D.M. Holtzman, K.R. Bales, T. Tenkova, A.M. Fagan, M. Parsadanian, L.J. Sartorius, B. Mackey, J. Olney, D. McKeel, D. Wozniak, S.M. Paul, Apolipoprotein E isoform-dependent amyloid deposition and neurite degeneration in a mouse model of Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 97 (2000) 2892–2897. G.P. Jarvik, E.M. Wijsman, W.A. Kukull, G.D. Schellenberg, C. Yu, E.B. Larson, Interactions of apolipoprotein E genotype, total cholesterol level, age and sex in prediction of Alzheimer disease: a case-control study, Neurology 45 (1995) 1092–1096. H. Jick, G.L. Zornberg, S.S. Jick, S. Seshadri, D.A. Drachman, Statins and the risk of dementia, Lancet 356 (2000) 1627–1631. B. Kwak, F. Mulhaupt, S. Myit, F. Mach, Statins as a newly recognized type of immunomodulator, Nature Med. 6 (2000) 1399– 1402. M.J. LaDu, M.T. Falduto, A.M. Manelli, C.A. Reardon, G.S. Getz, D.E. Frail, Isoform-specific binding of apolipoprotein E to betaamyloid, J. Biol. Chem. 269 (1994) 23403–23406. M.J. LaDu, C.A. Reardon, L.J. Van Eldik, A.M. Fagan, G. Bu, D.M. Holtzman, G.S. Getz, Lipoproteins in the central nervous system, Ann. NY Acad. Sci. 903 (2000) 167–175. E.C. Lerner, Y. Qian, M.A. Blaskovich, R.D. Fossum, A. Vogt, J. Sun, A.D. Cox, C.J. Der, A.D. Hamilton, S.M. Sebti, Ras CAAX peptidomimetic FTI-277 selectively blocks oncogenic Ras signaling by inducing cytoplasmic accumulation of inactive Ras–Raf complexes, J. Biol. Chem. 270 (1995) 26802–26806. J. Ma, A. Wee, H.B. Brewer, S. Das, H. Potter, Amyloid-associated proteins alpha-1-antichymotrypsin and apolipoprotein E promote assembly of Alzheimer beta-protein into filaments, Nature 372 (1994) 92–94. V. Manne, C.S. Ricca, J.G. Brown, A. Tuomari, N. Yan, D.V. Patel, R. Schmidt, M.J. Lynch, J. Ciosek, J.M. Carboni, S. Robinson, E.M. Gordon, M. Barbacid, B.R. Seizinger, S.A. Biller, Ras farnesylation as a target for novel antitumor agents: potent and selective farnesyl diphosphate analog inhibitors of farnesyltransferase, Drug Dev. Res. 34 (1995) 121–137. A. Naidu, R. Catalano, K. Bales, S. Wu, S.M. Paul, S.M.B. Cordell, Conversion of brain apolipoprotein E to an insoluble form in a mouse model of Alzheimer disease, NeuroReport 12 (2001) 1265– 1270. M. Nakai, T. Kawamata, K. Maeda, C. Tanaka, Expression of apolipoprotein E mRNA in rat microglia, Neurosci. Lett. 211 (1996) 41–44. J. Naslund, J. Thyberg, L.O. Tjernberg, C. Wernstedt, A.R. Karlstrom, N. Bogdanovic, S.E. Gandy, L. Lannfelt, L. Terenius, C. Nordstedt, Characterization of stable complexes involving apolipoprotein E and the amyloid beta peptide in Alzheimer’s disease brain, Neuron 15 (1995) 219–228.
A. Naidu et al. / Brain Research 958 (2002) 100–111 [28] I.L. Notkola, R. Sulkava, J. Pekkanen, T. Erkinjuntti, C. Ehnholm, P. Kivinen, J. Tuomilehto, A. Nissinen, Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer’s disease, Neuroepidemiology 17 (1998) 14–20. [29] K. Pahan, F.G. Sheikh, A.M.S. Namboodiri, I. Singh, Lovastatin and phenylacetate inhibit the induction of nitric oxide synthase and cytokines in rat primary astrocytes, microglia, and macrophages, J. Clin. Invest. 100 (1997) 2671–2679. [30] H.-J. Park, J.B. Galper, 3-Hydroxy-3-methylglutaryl CoA reductase inhibitors up-regulate transforming growth factor-beta signaling in cultured heart cells via inhibition of geranylgeranylation of RhoA GTPase, Proc. Natl. Acad. Sci. USA 96 (1999) 11525–11530. [31] B. Permanne, C. Perez, C. Soto, B. Frangione, T. Wisniewski, Detection of apolipoprotein E / dimeric soluble amyloid beta complexes in Alzheimer’s disease brain supernatants, Biochem. Biophys. Res. Commun. 240 (1997) 715–720. [32] R.E. Pitas, J.K. Boyles, S.H. Lee, D. Foss, R.W. Mahley, Astrocytes synthesize apolipoprotein E and metabolize apolipoprotein E-containing lipoproteins, Biochim. Biophys. Acta 917 (1987) 148–161. [33] J. Poirier, Apolipoprotein E and Alzheimer’s disease: a role in amyloid catabolism, Ann. NY Acad. Sci. 924 (2000) 81–90. [34] D. Quon, Y. Wang, R. Catalano, J. Marian Scardina, K. Murakami, B. Cordell, Formation of b-amyloid protein deposits in brains of transgenic mice, Nature 352 (1991) 239–241. [35] L.M. Refolo, B. Malester, J. LaFrancois, T. Thomas-Bryant, R. Wang, G.S. Tint, K. Sambamurti, K.E. Duff, M.A. Pappolla, Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model, Neurobiol. Dis. 7 (2000) 321– 331. [36] L.M. Refolo, M.A. Pappolla, J. LaFrancois, J. Malester, S.D. Schmidt, T. Thomas-Bryant, G.S. Tint, R. Wang, M. Mercken, S.S. Petanceska, K.E. Duff, A cholesterol-lowering drug reduces betaamyloid pathology in a transgenic mouse model of Alzheimer disease, Neurobiol. Dis. 8 (2001) 890–899. [37] C. Russo, D. Angelini, D. Dapino, A. Piccini, G. Piombo, G. Schettini, S. Chen, J.K. Teller, D. Zaccheo, P. Gambetti, M. Tabaton, Opposite roles of apolipoprotein E in normal brains and in Alzheimer’s disease, Proc. Natl. Acad. Sci. USA 95 (1998) 15598– 15602. [38] A.M. Saunders, Apolipoprotein E and Alzheimer’s disease: an update on genetic and functional analyses, J. Neuropathol. Exp. Neurol. 59 (2000) 751–758. [39] J.G. Sheng, R.E. Mrak, W.T.S. Griffin, Apolipoprotein E distribution among different plaque types in Alzheimer disease: implications for its role in plaque progression, Neuropathol. Appl. Neurol. 22 (1996) 334–341. [40] M. Simons, P. Keller, B. De Strooper, K. Beyreuther, C.G. Dotti, K. Simons, Cholesterol depletion inhibits the generation of betaamyloid in hippocampal neurons, Proc. Natl. Acad. Sci. USA 95 (1998) 6460–6464.
111
[41] M. Sinensky, Functional aspects of polyisoprenoid protein substituents: roles in protein–protein interaction and trafficking, Biochim. Biophys. Acta 1529 (2000) 203–209. [42] A.J.C. Slooter, M.-X. Tang, C.M. van Duijn, Y. Stern, A. Ott, K. Bell, M.M.B. Breteler, C. van Broeckhoven, T.K. Tatemichi, B. Tycko, A. Hofman, A.R. Mayeux, Apolipoprotein E epsilon-4 and the risk of dementia with stroke. A population-based investigation, J. Am. Med. Assoc. 277 (1997) 818–821. [43] D.A. Snowdon, L.H. Greiner, J.A. Mortimer, K.P. Riley, P.A. Greiner, W.R. Markesbery, Brain infarction and the clinical expression of Alzheimers disease. The nun study, J. Am. Med. Assoc. 277 (1997) 813–817. [44] D.L. Sparks, S.W. Scheff, J.C. Hunsaker, H. Liu, T. Landers, D.R. Gross, Induction of Alzheimer-like beta-amyloid immunoreactivity in the brains of rabbits with dietary cholesterol, Exp. Neurol. 126 (1994) 88–94. [45] D.L. Sparks, Coronary artery disease, hypertension, ApoE, and cholesterol: a link to Alzheimer’s disease?, Ann. NY Acad. Sci. 826 (1997) 128–146. [46] L. Stoppini, P.A. Buchs, D. Muller, A simple method for organotypic culture of nervous tissue, J. Neurosci. Methods 37 (1991) 173–182. [47] S.D. Styren, M.I. Kamboh, J. Dekosky, Expression of differential immune factors in temporal cortex and cerebellum: the role of alpha-1-antichymotrypsin, apolipoprotein E, and reactive glia in the progression of Alzheimer disease, Comp. Neurol. 396 (1998) 511– 520. [48] J. Sun, Y. Qian, A.D. Hamilton, S.M. Sebti, Ras CAAX peptidomimetic FTI-276 selectively blocks tumor growth in nude mice of a human lung carcinoma with k-Ras mutation and p53 deletion, Cancer Res. 55 (1995) 4243–4247. [49] J. Sun, Y. Qian, A.D. Hamilton, S.M. Sebti, Both farnesyltransferase and geranylgeranyltransferase I inhibitors are required for inhibition of oncogenic k-Ras prenylation but each alone is sufficient to suppress human growth in nude mouse xenografts, Oncogene 16 (1998) 1467–1473. [50] T. Wisniewski, E.M. Castano, A. Goldbek, T. Vogel, B. Frangione, Acceleration of Alzheimer’s fibril formation by apolipoprotein E in vitro, Am. J. Pathol. 145 (1994) 1030–1035. [51] G. Weitz-Schmidt, K. Welzenbach, V. Brinkmann, T. Kamata, J. Kallen, C. Bruns, S. Cottens, Y. Takada, U. Hommel, Statins selectively inhibit leukocyte function antigen-1 by binding to a novel regulatory integrin site, Nature Med. 7 (2001) 687–692. [52] B. Wolozin, W. Kellman, P. Ruosseau, G.G. Celesia, G. Siegel, Decreased prevalence of Alzheimer disease with 3-hydroxy-3methylglutaryl coenzyme A reductase inhibitors, Arch. Neurol. 57 (2000) 1439–1443. [53] Q. Xu, C. Cyras, D.A. Sanan, B. Cordell, Isolation and characterization of apolipoproteins from murine microglia, J. Biol. Chem. 275 (2000) 31770–31777.