- Email: [email protected]
APOE and cholesterol homeostasis in Alzheimer's disease
Review
APOE and cholesterol homeostasis in Alzheimer’s disease Vale´rie Leduc1, Ste´phanie Jasmin-Be´langer1 and Judes Poirier1,2 1 2
Douglas Mental Health University Institute, 6875 Lasalle, Montreal (Verdun), Quebec, H4H 1R3, Canada Center for Studies in Aging, McGill University, Verdun, Quebec, H4H 1R3, Canada
Converging evidence from clinical and pathological studies indicate the presence of important relationships between the ongoing deterioration of brain lipid homeostasis, vascular changes and the pathophysiology of sporadic Alzheimer’s disease (AD). These associations include the recognition of cholesterol transporters apolipoprotein E (APOE), APOC1 and APOJ as major genetic risk factors for common AD and observations associating risk factors for cardiovascular disease such as high midlife plasma cholesterol, diabetes, stroke, obesity and hypertension to dementia. Moreover, recent clinical findings lend support to the notion that progressive deterioration of cholesterol homeostasis in AD is a central player in the disease pathophysiology and is, therefore, a potential therapeutic target for disease prevention. Alzheimer’s disease and cholesterol homeostasis Dementias are progressive neurodegenerative disorders characterized by a decline in cognitive functions beyond what might be expected from normal aging. Of these dementias, Alzheimer’s disease (AD) is the most common type among the elderly accounting for approximately 60% of all dementia cases diagnosed [1]. Pathologically, AD is characterized by the presence of extracellular amyloid plaques, intracellular neurofibrillary tangles (Box 1) as well as extensive neuronal and synaptic losses [2]. Although gradual atrophy of the brain is observed, degeneration occurs preferentially in neuronal populations of cholinergic phenotype [3]. There are several reasons why these particular neurons might be more vulnerable to AD, notably their high-energy requirement, their reliance on axonal transport for sustained function and trophic support, and their large and poorly protected (sparsely myelinated) cell surface area that increases their exposure to toxic environmental conditions and oxidative stresses [3,4]. Accordingly, neuronal loss usually starts in brain structures rich in long cholinergic fibers, that is the entorhinal cortex and hippocampus of the limbic system, spreads to the temporal lobe, then to the frontal cortex and ultimately to primary sensory and visual areas [3,4]. Etiologically, AD is considered a multifactorial disease with a strong genetic component. The disease can be subdivided into two distinct categories, the so-called familial and sporadic forms of the disease. The identification of specific gene mutations has provided new insight into the molecular changes responsible for the pathophysiological Corresponding author: Poirier, J. ([email protected])
processes of AD. Mutations in the genes encoding the amyloid precursor protein (APP), presenilin 1 and presenilin 2 associate with rare early-onset forms of familial AD [5]. However, coding mutations in these genes do not increase the risk of the more prevalent late-onset disease, which accounts for more than 95% of all AD cases worldwide [1]. APOE is the only known locus that affects the risk of the late-onset form of the disease, with the e4 allele (encoding apolipoprotein 4, APOE4) and the e2 allele (encoding APOE2) increasing and decreasing the risk level, respectively [6,7]. Recently, several independent genomewide association studies (GWAS) in homogeneous and heterogeneous populations of AD with age-matched control cases in North America, Europe and Asia associated the e4 allele with AD [8–12]. Surprisingly, the remaining genetic associations uncovered in these GWAS failed to replicate across studies, except for the gene encoding APOJ, which is an accessory protein to APOE in the maintenance of cholesterol homeostasis. APOE coordinates the mobilization and redistribution of cholesterol in repair, growth and maintenance of myelin and neuronal membranes during development or after injury in the peripheral nervous system [13,14]. In the central nervous system (CNS), APOE in partnership with APOJ and APOC1 plays a pivotal role in cholesterol delivery during the membrane remodeling associated with synaptic turnover and dendritic reorganization [15,16]. The near complete absence of other key plasma apolipoproteins such as APOA1 and APOB in the brain further emphasizes the critical and unique role of APOE for cholesterol transport in the normal or injured CNS. Cardiovascular contributions to AD pathology The prevalence and incidence of degenerative and vascular dementias increase almost exponentially with age, from 70 years onward. In view of the increasing longevity of humans, both types of dementia have progressively evolved into major public health problems worldwide [17]. The integrity of the cerebral vasculature is crucial for the maintenance of cognitive function; vascular factors, such as hypertension, myocardial infarction, diabetes, obesity, hyper- or dyslipidemia, ischemic white matter lesions and generalized atherosclerosis, associate with dementia and cognitive decline [18]. The connection between vascular factors and cognition remain largely unknown, although recent evidence associates the e4 allele with regional cerebral blood flow (rCBF) disturbances in regions of the brain affected by AD [19]. Indeed, apparently
1471-4914/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2010.07.008 Trends in Molecular Medicine, October 2010, Vol. 16, No. 10
469
Review Box 1. Neurofibrillary tangles One other famous pathological feature of AD is the neurofibrillary tangle. These intraneuronal filamentous deposits primarily result from the hyperphosphorylation of the microtubule-stabilizing Tau proteins and elicit neuronal death by disrupting axonal transport and inducing widespread metabolic decline [100].
clinically normal APOE4 carriers 50 years and older are more likely to show significant cognitive impairment than APOE4-negative subjects and exhibit greater decline in rCBF over time by position-emission tomography [20]. Interestingly, APOE4 carriers have higher baseline rCBF than noncarriers in regions exhibiting significant longitudinal decline in rCBF, suggesting that APOE4 carriers might have to exert greater cognitive effort to perform similar to their noncarrier counterparts [21]. However, because relatively young healthy adult carriers of APOE4 (315 years) also exhibit greater baseline rCBF than noncarriers, these increments might reflect mechanisms to compensate for inefficient neuronal processes, such as reduced synaptic plasticity and neuronal growth in APOE4 carriers. Accordingly, the e4 allele dose impacts the age of onset in both familial and sporadic cases [6,22], again suggesting the possibility that APOE4 carriers are developmentally different from their noncarrier counterparts. The emergence of decreased CBF and hypertension as major risk factors for sporadic AD has garnered interest from a therapeutic angle [23,24]. Epidemiological associations between common AD and vascular disease risk factors could reflect an overdiagnosis of AD in individuals with silent cerebrovascular disease or alternatively cerebrovascular disease could affect the clinical expression and onset of AD. Further possibilities include: AD might increase the risk of vascular diseases; vascular disease might silently stimulate AD processes; or similar mechanisms, such as disturbances in lipid homeostatic processes and abnormal cholesterol transport, distribution or accumulation could contribute to the pathogenesis of both disorders. According to several longitudinal studies, hypertension predisposes individuals to cognitive impairment and ensuing dementia after a period varying from a few years to several decades [25]. The e4 allele drastically enhances the dementia risk level in hypertensive subjects [25]. Conversely, antihypertensive drugs, such as angiotensin-converting enzyme inhibitors and diuretics, markedly reduce the risk of cognitive impairment and in some studies delay the onset of AD itself [26]. Estrogen, which lowers serum cholesterol levels and the risk of coronary heart disease, decreases the incidence of dementia in cross-sectional epidemiological studies but does not affect disease progression [27] in diagnosed mild-to-moderate AD. Cholesterol synthesis, transport and degradation Cholesterol homeostasis in the brain is carefully maintained through a series of interdependent processes that include synthesis, storage, degradation and transport. Cholesterol and other lipids are used for membrane synthesis and for many other anabolic or catabolic activities by cells throughout the body including those of the CNS, a site 470
Trends in Molecular Medicine Vol.16 No.10
of high lipid turnover. The cholesterol requirements of most mammalian cells are met by two separate but inter-related processes. One process is the endogenous synthesis of cholesterol, which involves over 20 reactions and is regulated primarily by the activity of the 3-hydroxy3-methylglutaryl coenzyme A reductase (HMGCR) that catalyzes the formation of mevalonate, the key precursor molecule in the synthesis of cholesterol (Figure 1). Brain cells, particularly astrocytes and neurons, cultured in vitro synthesize cholesterol at a rate that is inversely proportional to the cholesterol content in the growth environment [28]. The other central process regulating cholesterol levels in the brain involves the utilization of lipoprotein-derived cholesterol, following internalization of a lipoprotein bound to its surface receptor [usually an APOE-rich high-density lipoprotein (HDL)-like complex]. The ATPbinding cassette transporter types A1 and G1 (ABCA1/ ABCG1) coordinate the mobilization of cholesterol from cytoplasmic pools to the cell surface membrane [29], where it is eventually combined with APOE, APOJ and phospholipids to produce a functional HDL-like particle through the action of the lipoprotein lipase (LPL) [30]. Newly synthesized lipoprotein particles are released into the extracellular space where they migrate, in a gradientdriven manner, toward both APOE and APOJ receptors on ependymal cells surrounding the ventricles or toward APOE receptors on neuronal and glial cells. Following binding of the APOE lipoprotein particle to neuronal surface low-density lipoprotein receptor (LDLR), the APOE lipoprotein/LDLR complex is internalized and degraded, releasing cholesterol that can be used for synapse formation and terminal proliferation (Figure 2). Excess cholesterol can also be converted into the more hydrophilic oxysterol 24S-hydroxycholesterol by the neuron-specific enzyme cholesterol 24S-hydroxylase, which is encoded by CYP46A1. This oxysterol leaves the brain via the blood–brain barrier (BBB) and directly enters the bloodstream [31]. Thus, almost all the 24S-hydroxycholesterol present in human circulation has a cerebral origin [32]. Cholesterol levels in the different brain areas in AD have been documented over the past 25 years. Brain cholesterol levels (total, free and esterified cholesterol) are significantly reduced in hippocampal and cortical areas in AD patients compared to age-matched neurological controls but no differences are detected in the pathologyfree cerebellum [33,34]. Cholesterol levels thus differ depending on the brain region examined and rather interestingly the brain areas that are more susceptible to AD are those that contain the greatest amounts of synaptic membrane cholesterol, that is hippocampus>cortex> cerebellum [35]. Consistent with these observations, HMGCR activity in the brain of autopsy-confirmed AD cases is significantly reduced in cortical and hippocampal areas as compared to age-matched control subjects [36]. The reduction of HMGCR enzymatic activity is, however, APOE-genotype independent. HMGCR mRNA levels are not different in AD versus control subjects but are inversely correlated with APOE protein levels, suggesting a compensatory upregulation of cholesterol internalization by surviving cortical and hippocampal neurons [36].
(Figure_1)TD$IG][ Review
Trends in Molecular Medicine
Vol.16 No.10
Figure 1. Cholesterol synthesis in eukaryotic cells. Schematic representation of the mevalonate pathway and key enzymes regulating cholesterol synthesis and esterification. The isoprenoid cascade (dotted box) provides a biochemical association through which cholesterol synthesis can indirectly modulate production of phosphotau in the CNS. Abbreviations: GGPP, geranylgeranyl diphosphate; CdK5, cyclin-dependent kinase 5; PKA, protein kinase A; GSK3B, glycogen synthase kinase 3 beta; HMG CoA, 3-hydroxy-3-methylglutaryl-CoA; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
Ultrastructural studies have shown that with the loss of neuronal input to the hippocampus, astrocytes and microglia progressively engulf both presynaptic terminals and preterminal axons (Figure 2) to rapidly clear the area and allow synaptic replacement [37]. Once metabolized, these terminal-derived ovoids generate a large glial store of lipids that are readily available for the synthesis of the membrane components necessary for new synapses and dendrites (Figure 2) in surviving adjacent neurons [37]. The local accumulation of high concentrations of cellular cholesterol in astrocytes induces APOE and APOJ synthesis and secretion in combination with cholesterol (Figure 2). As the intracellular concentration of cholesterol increases in glial cells, cholesterol synthesis is progressively repressed at the level of HMGCR [38]. Deafferented granular cell neurons in the hippocampus exhibit marked increases of LDLR during the early and middle phases of the reinnervation process (Figure 2) [38]. Both the LDLR-related protein (LRP) and the APOE receptor type 2 expressed by neurons indirectly modulate terminal proliferation and axonal extension [39], primarily through a signaling pathway [40] in contrast to the APOE/LDLR, which uses direct lipoprotein internalization mechanisms. In response to the increased internalization of cholesterol via the APOE/LDLR pathway,
neurons of the injured hippocampal formation in AD reduce their HMGCR activity (Figure 2). The apparent contradiction of depressed cholesterol synthesis and increased APOE expression in the presence of active synaptogenesis can be reconciled by postulating that cholesterol from degenerating terminals is salvaged through the APOE transport/LDLR uptake pathway [38,41] and reutilized. The absence of either APOE or LDLR in deficient (knockout) mice compromises synaptic integrity and compensatory synaptogenesis and causes age-dependent cognitive deficits, presumably owing to an impaired ability of the brain to produce or bind to functional HDL-like particles [42,43]. Neurodegeneration combined with brain atrophy also associate with significantly reduced concentrations of other lipid products. Compared to age-matched neurological control subjects, patients with AD have marked increases and decreases in 24S-hydroxycholesterol in cerebrospinal fluid (CSF) and circulation, respectively [44]. However, genetic association studies examining the relationship between the CYP46A1 gene and sporadic AD are in conflict. Similarly, the phospholipid content of membrane structures in AD brains is reduced compared to age-matched neurological control subjects [34]. APOE, which is the main phospholipid carrier in the mature CNS, affects 471
(Figure_2)TD$IG][ Review
Trends in Molecular Medicine Vol.16 No.10
Figure 2. Cholesterol recycling in the injured CNS. Degenerating nerve terminals are internalized and degraded. Nonesterified cholesterol (i) is used as free cholesterol (FC) for the assembly of an APOE+APOJ/cholesterol/lipoprotein complex (ii) or converted into cholesterol esters (CEs) for storage. The APOE+APOJ/cholesterol/lipoprotein complexes are then directed either into circulation, presumably through the ependymal cells surrounding the ventricles, and/or to specific brain cells requiring lipids. APOE+APOJ complexes are internalized by the neuronal LDLR pathway (iii) and the cholesterol is released (iv) for dendritic proliferation and/or synaptogenesis (vi). As a consequence of internalization, cholesterol synthesis in neurons via the HMGCR pathway (v) is progressively repressed. Abbreviations: E, APOE; J, APOJ; PL, phospholipids; CE, cholesterol ester; FC, free cholesterol. ACAT, acetyl coenzyme A cholesterol acyltransferase; ABCA1/G1, ATP binding cassette transporter A1 and G1; LPL, lipoprotein lipase; AA, amino acids; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
phospholipids concentrations in an e4 dose-dependent manner [45,46]. Interestingly, the cholesterol-to-phospholipids ratio in cortical lipid membranes is markedly decreased in the temporal and frontal cortices of AD subjects as compared to age-matched neurological control subjects [47], consistent with pronounced thinning of the plasma membrane observed in cortical areas. Cholesterol is a vital component of the CNS and is essential for axonal growth, and synaptic formation and remodeling, processes that are crucial for learning, formation of memories and neuronal repair [48]. The CNS has thus evolved two strategies to ensure adequate supplies of cholesterol: de novo synthesis and the internalization of lipoprotein particles. Synapses and dendrites are located far away from the cell body where cholesterol synthesis takes place; thus, the internalization of lipoproteins is imperative for neurons to meet their cholesterol needs [49]. Unsurprisingly, cholesterol and lipid levels are decreased in AD. Because APOE proteins are expressed at different levels according to the gradient E2>E3>E4 [50], concentration differences could explain why the APOE4 variant confers a huge risk for developing AD and results in such poor synaptic remodeling capacity [51]. 472
Cholesterol homeostasis and APP metabolism: a role for APOE-mediated beta-amyloid (Ab) transport and degradation Dominant mutations within the APP gene that alter the processing of the APP protein can result in AD. Although cleavage of the APP by a-secretase precludes the formation of Ab peptides, consecutive cleavage of APP by b-secretase and g-secretase stimulates Ab production, resulting in the accumulation of Ab peptides that eventually aggregate as extracellular amyloid plaques, one of the landmark pathologies of AD. Since Sparks’ initial observation that feeding rabbits a cholesterol-enriched diet for eight weeks leads to an accumulation of intracellular Ab immunoreactivity in hippocampal neurons [52], studies associating cholesterol and Ab metabolism have overrun the field of AD research. The finding that cholesterol enrichment enhances Ab production has been widely replicated in animals [53,54] and cultured-cell models strongly suggesting that lipids, especially cholesterol, modulate Ab production. Accordingly, lowering cholesterol levels apparently decreases Ab production [55–59]. Exposure to cholesterol-lowering agents such as probucol [55,56], statins [57], BM15.766 [58] and cyclodextrin [57,59] reduces the amyloid load in
Review cultured cells and in vivo mammalian models. On the one hand, the mechanisms by which probucol and statins reduce Ab load could be pleiotropic. Probucol affects cholesterol metabolism through multiple mechanisms but most probably induces APOE synthesis [56]. On the other hand, BM15.766 inhibits the enzyme catalyzing the final step of cholesterol biosynthesis, namely 7-dehydrocholesterol reductase, suggesting that the amyloid-reducing effect might involve sterol depletion [58]. These findings further support clinical and epidemiological evidence suggesting that individuals with elevated midlife plasma cholesterol have increased susceptibility to dementia and AD when compared to normal cholesterolemic subjects [60,61]. The mechanism by which changes in plasma cholesterol affect AD and Ab production in the brain is, however, obscure. Although brain endothelial cells can transport peripheral LDL cholesterol through LDLR-mediated uptake [62], this pathway appears to be of little importance under normal conditions. Indeed, owing to the BBB, cholesterol metabolism in the brain is regulated independently from the periphery; isotope-labeled cholesterol or deuterium do not accumulate in the brain [63,64]. Furthermore, the absence of APOB, the main apolipoprotein associated with LDL particles, within the CNS [65] argues against a significant transport of LDL cholesterol across the BBB. However, under pathological conditions, small amounts of cholesterol might enter the CNS. A small but significant increase in brain cholesterol was observed in hypercholesterolemic transgenic mice [53]. Additionally, cholestanol and APOB abnormally accumulate in the CSF of patients with cerebrotendinous xanthomatosis, a rare disorder engendered by a defect in the biosynthesis of bile acids [66]. If increasing cholesterol levels promotes Ab generation, how can AD associate with decreased total cholesterol levels and increased Ab levels in the brain? Total cholesterol measurements only mirror bulk cholesterol within the CNS, which is highly compartmentalized. Hence, cholesterol levels in specific subcellular compartments might be a more important factor in the pathophysiology of AD than bulk cholesterol. Indeed, cholesterol is asymmetrically distributed in the exofacial and cytofacial leaflets of cell membranes [67] and is not evenly distributed in cellular organelles [68]. Additionally, lipid domains enriched in cholesterol such as lipid rafts have been identified in the plasma membrane. Thus, changes in cholesterol distribution or the formation of cholesterol-rich domains might not be reflected in the total amount of brain cholesterol. Accordingly, the selective removal of cholesterol from the plasma membranes of hippocampal and mixed cortical neurons by cyclodextrin is sufficient to inhibit Ab production [57], pinpointing lipid rafts as potential critical elements for Ab production. In agreement with this concept, lipid rafts – membrane microdomains rich in lipids and cholesterol – modulate APP processing. Both g-secretase and b-secretase BACE-1 activities have been isolated from lipids rafts [69], whereas the majority of a-secretase ADAM10 activity occurs in nonraft regions of the membrane [70]. Accordingly, increased intracellular cholesterol levels, which favor lipid raft formation, negatively regulate
Trends in Molecular Medicine
Vol.16 No.10
a-secretase activity [71] but stimulate both b- and g-secretase activities, resulting in an increase in Ab production. This raft/nonraft model of APP processing might help to explain cholesterol-dependent regulation of Ab production, but the explanation is imperfect as a modest reduction in cellular cholesterol paradoxically accelerates b-secretase cleavage and Ab production [72]. In line with these findings, total cholesterol might not regulate secretase activity but rather the ratio of free cholesterol (FC) to cholesterol esters (CEs), which is regulated by acyl coenzyme A cholesterol acyltransferase (ACAT) [73]. Moreover, ACAT activity correlates with changes in a-, b- and g-secretase APP products [73], indicating that ACAT activity affects APP processing. Whether ACAT activity modulates Ab generation by directly modulating the activity of the secretases or the availability of APP for cleavage through changes in cholesterol distribution and compartmentalization remains to be clarified. Nevertheless, because these effects are only observed in cells containing physiological or supraphysiological levels of FC, a role for FC in Ab modulation cannot be ruled out [73]. The pharmacological inhibition of ACAT causes a concomitant increase in FC levels in the cytoplasm and a reduction in both CE levels and Ab production [74], similar to the findings that APOE, which promotes cholesterol transport and internalization in an APOE2>APOE3>>APOE4 manner, strongly modulates Ab burden [75–77]. Numerous studies show a positive correlation between amyloid burden and e4 allele dose. Indeed, the e4 allele associates with increased plaque density in humans [78] and in transgenic mice [76,77,79]. The exact mechanism by which APOE4 promotes Ab production and accumulation remains elusive, although the isoform-dependent modulation of the local steady-state levels of APOE remains the primary suspect, as APOE concentrations follow an E2>E3>E4 gradient [36,77]. Because APOE acts as a powerful Ab scavenger in the extracellular space, low steady-state levels associated with the presence of the e4 allele are bound to affect the normal removal of the extracellular peptide. Numerous in vitro studies have demonstrated that human APOE facilitates cellular Ab uptake and degradation [80,81]. Additionally, APOE promotes brain to blood removal of Ab peptides by transport across the BBB, a clearing process that follows an APOE2>APOE3>APOE4 gradient [82]. Whereas the predominant role of APOE in the scavenging of Ab in the CNS is generally accepted and well documented, the presumed effect of APOE on Ab aggregation is controversial. In vitro studies suggest that APOE isoforms, with APOE4 being the least effective, might decrease Ab fibrillogenesis by interfering with Ab nucleation [83]. These observations contrast with animal studies, where the knockdown of the murine ApoE gene drastically decreases Ab deposition, along with a marked loss of thioflavin-S-positive amyloid plaques, which are plaques containing amyloid in the form of fibrils [84]. However, transgenic mice expressing human APOE show a marked delay in Ab deposition and plaque formation relative to mice expressing murine APOE or no APOE [79]. This suggests that human APOE exhibits physiological properties that are different from murine APOE with regard to amyloid metabolism, removal and/or deposition in the mature brain. 473
Review Cholesterol homeostasis disturbances occur in AD and modulate Ab production, but the presence of the BBB, the complexity of the biosynthesis and regulation of cholesterol and the lack of specific and sensitive methods to quantify brain cholesterol levels at subcellular resolution impedes our understanding of the global mechanism involved. Although it has long been recognized that brain cholesterol is independent from changes in plasma cholesterol, growing evidence suggests the contrary. Changes in cholesterol domains rather than bulk cholesterol levels could be involved in AD pathology and Ab modulation, but other key factors such as CE levels and ACAT should not be underestimated. Cholesterol metabolism: a potential therapeutic target for the treatment of sporadic AD Under normal circumstances, cholesterol synthesis via the HMGCR pathway (Figure 1) is required only when the amount of lipoprotein internalization by the APOE/LDLR pathway is insufficient to meet the cholesterol requirement of a cell. To maintain cellular cholesterol homeostasis, a potent negative feedback on HMGCR activity (Figure 1) and gene expression decreases cholesterol synthesis in response to excess intracellular sterol internalization via members of the LDLR family. This first and most important feedback on HMGCR occurs through a decrease in gene
Trends in Molecular Medicine Vol.16 No.10
transcription or the modulation of the translational efficiency of HMGCR; the latter can decrease or increase protein levels. Interestingly, a hereditary cholesterol storage disorder known as Niemann–Pick C (NPC) shows Alzheimer-like tau pathology in youth or adolescence without the amyloid plaque deposition. NPC results from a mutation in either of two genes, NPC1 or NPC2/HE1, and associates with the intracellular misrouting and accumulation of FC [85]. Tangle-bearing neurons in NPC subjects display high levels of cholesterol that resemble the tangle-bearing neurons in AD [86,87]. The administration of HMGCR inhibitors (statins) in NPC mice, which normally carry high concentrations of phospho-tau in the brain, reduces brain phospho-tau to near normal levels [86]. Figure 1 illustrates the geranylgeranyl diphosphate cascade believed to be responsible, at least in part, for the coordinated phosphorylation of tau in neurons [86,88]. Evidence from multiple prospective epidemiological studies, suggesting that the presymptomatic use of statins in middle-age subjects confers some levels of protection against sporadic AD later in life [89–91], provides us with a potentially interesting biochemical target for prevention studies. Furthermore, a recent populationbased study examining the effect of statins in the cognitively intact elderly revealed that over time, statin users
[(Figure_3)TD$IG]
Figure 3. An APOE4/amyloid hypothesis of AD pathophysiology. The sequence of pathogenic events leading to neuronal cell loss and synaptic damage is based on the well-established amyloid cascade hypothesis (blue background), which proposed that the accumulation of beta amyloid in the brain is the primary influence driving AD pathology. The different modulators of beta amyloid metabolism affecting lipid homeostasis, such as APOE, HMGCR and APOJ, have been added to the cascade. Finally, the emerging roles of cholesterol and HMGCR have been put in perspective in relation to their respective contribution to the pathophysiology of AD. Abbreviations: +, enhancer, , inhibitor; CLA, cholesterol-lowering agent; APP, amyloid precursor protein; PS1/PS2, presenilin 1/2; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
474
Review accumulate significantly less neurofibrillary tangles (but not amyloid plaques) when compared to age-matched statin-free subjects [92]. Together, these data point toward HMGCR as a potential key player in the pathophysiology of tau phosphorylation and tangle deposition in the aging and diseased brain. Statins competitively inhibit HMGCR activity, thereby inhibiting the synthesis of cholesterol by preventing the conversion of HMG CoA to mevalonate. This family of inhibitors reduces the formation and internalization of LDL cholesterol particles in circulation and upregulates LDLR activity at the cell surface of liver cells, thereby reducing serum LDL cholesterol and triglycerides and increasing HDL cholesterol concentration [93]. It is unclear whether the beneficial effects of statins are mediated through BBB penetration or via alterations of plasma cholesterol levels. In an attempt to clarify this issue, two large prospective population studies examined the effects of lipophilic versus lipophobic statins on AD risk. Because only lipophilic statins are believed to cross the BBB, a beneficial effect of lipophilic over lipophobic compounds is expected if this effect is mediated through direct modulation of brain cholesterol levels. One study detected preferential benefits from lipophilic compounds, such as simvastatin, over lipophobic ones [90], whereas the other did not [91]. Interestingly, however, both reconfirmed the association between statin use and reduced risk for AD. Hence, it remains unclear how statins exert their beneficial effect against AD and whether the effect is mediated through modulation of brain and/or plasma cholesterol levels. In addition to the statin-mediated LDLR upregulation, statins might induce APOE synthesis and secretion, thereby facilitating both lipoprotein delivery and internalization in neurons and Ab removal from the extracellular space. The potential use of an APOE-inducer drug in the treatment of AD has been reviewed elsewhere [36]. Briefly, two major approaches have been proposed to facilitate cholesterol mobilization and transport in AD, particularly in e4 allele carriers. One proposes to induce APOE synthesis and secretion in the brain with probucol [36,56], a cholesterol-lowering agent used in the mid 1980 s to treat familial hypercholesterolemia [94]; the other strategy proposes to administer liver X receptor agonists that stimulate ABCA1, APOE and the LDLR production through a common signaling pathway [95,96]. These independent and yet complementary approaches indicate that the modulation of brain cholesterol homeostasis might interfere with disease onset in subjects exposed to cholesterol-lowering agents prior to the diagnosis of AD. Unfortunately, the most recent large-scale, randomized double-blind placebo-controlled clinical trials with atorvastatin [97] and simvastatin [98] failed to slow disease progression in mild-to-moderate AD, suggesting that statins provide beneficial effects only in the presymptomatic stages of the disease. It is still unclear whether the risk-reducing effect of statins is specific to dementia of the Alzheimer’s type or more general. It is possible that the beneficial effect of cholesterol-lowering agents stems from a completely distinct pathway involving an independent cardiovascular risk factor (other than cholesterol) or other
Trends in Molecular Medicine
Vol.16 No.10
pleiotropic effects (i.e. increased endothelial nitric oxide synthase activity, decreased endothelin-1 activity or antiinflammatory activity) that modulate disease onset in atrisk individuals (i.e. e4 allele carriers). In this scenario, cholesterol-lowering agents would indirectly prevent vascular risk factors, such as circulating levels of cholesterol or atherosclerotic plaque deposition, from modulating the age of onset in AD. Although it remains difficult to determine the exact mode of action by which cholesterol-lowering agents affect the pathophysiology of AD, the recent discovery of polymorphic genetic variants in the HMGCR gene [99] provide us with a possible explanation as to what could be the molecular basis and target for the beneficial effect against AD. Recent preliminary evidence suggests that some of these polymorphisms affect either HMGCR splicing efficiency or the production of abnormal spliced variants that indirectly affects enzymatic activity in the brain [99]. Is there a unifying hypothesis of AD pathophysiology? Although age is a key determinant in AD pathology, genetic risk factors play a central role in this process (Figure 3). Factors such as APOE4, ACAT, APOJ, LRP, LPL or paraoxonase 1 all modulate cholesterol homeostasis and their genes are risk factors for common AD. In recent years, it has become obvious that traditional vascular risk factors such as hypertension, myocardial infarction, diabetes, obesity, hyper- or dyslipidemia, ischemic white matter lesions and generalized atherosclerosis, definitively associate with the common form of dementia as well as cognitive decline associated with normal aging. In contrast to the aforementioned genetic risk factors, it is possible to pharmacologically intervene to reduce and even eliminate many of the vascular factors associated with AD. In practice, this resulted in the absence of cognitive decline of placebo groups in nearly all of the 6-month randomized clinical trials performed in AD these past five to six years. However, this does not stop disease progression over the long term. Although indirect, these replicated observations (from multiple clinical studies performed by several independent pharmaceutical corporations) clearly show that, similar to the APOE4 findings, these vascular risk factors (including circulating cholesterol levels) significantly contribute to AD pathophysiology (Figure 3) but most probably do not cause the disease itself. In the same context, we believe that strategies designed to minimize or eliminate the physiological impact of the risk factors in the decades preceding the onset of AD in ‘‘at-risk’’ subjects will delay the onset of the symptoms by several years but will not prevent the disease itself. That being said, it has been estimated that a delay in the onset of AD by five years would result in a 50% reduction of the incidence within one generation. Although this approach might not cure the disease, it would definitively impact its overall socioeconomic burden. Acknowledgements We wish to thank our long-time collaborators for their intellectual input to this manuscript as well as the funding agencies supporting this research: the Canadian Institute for Health Research, Natural Sciences and Engineering Research Council of Canada, the Fonds de la recherche en sante´ du Que´bec and the Alzheimer Society of Canada. 475
Review References 1 Terry, R.D. (2006) Alzheimer’s disease and the aging brain. J. Geriatr. Psychiatry Neurol. 19, 125–128 2 Adlard, P.A. and Cummings, B.J. (2004) Alzheimer’s disease – a sum greater than its parts? Neurobiol. Aging 25, 725–733 3 Mattson, M.P. and Magnus, T. (2006) Ageing and neuronal vulnerability. Nat. Rev. Neurosci. 7, 278–294 4 Braak, H. et al. (2000) Vulnerability of select neuronal types to Alzheimer’s disease. Ann. N. Y. Acad. Sci. 924, 53–61 5 Hardy, J. (1997) Amyloid, the presenilins and Alzheimer’s disease. Trends Neurosci. 20, 154–159 6 Poirier, J. et al. (1993) Apolipoprotein E polymorphism and Alzheimer’s disease. Lancet 342, 697–699 7 Corder, E.H. et al. (1994) Protective effect of apolipoprotein E type 2 allele for late onset Alzheimer disease. Nat. Genet. 7, 180–184 8 Carrasquillo, M.M. et al. (2009) Genetic variation in PCDH11X is associated with susceptibility to late-onset Alzheimer’s disease. Nat. Genet. 41, 192–198 9 Reiman, E.M. et al. (2007) GAB2 alleles modify Alzheimer’s risk in APOE epsilon4 carriers. Neuron. 54, 713–720 10 Li, H. et al. (2008) Candidate single-nucleotide polymorphisms from a genomewide association study of Alzheimer disease. Arch. Neurol. 65, 45–53 11 Harold, D. et al. (2009) Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer’s disease. Nat. Genet. 41, 1088–1093 12 Lambert, J.C. et al. (2009) Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer’s disease. Nat. Genet. 41, 1094–1099 13 Boyles, J.K. et al. (1990) Accumulation of apolipoproteins in the regenerating and remyelinating mammalian peripheral nerve. Identification of apolipoprotein D, apolipoprotein A-IV, apolipoprotein E, and apolipoprotein A-I. J. Biol. Chem. 265, 17805–17815 14 Poirier, J. et al. (1991) Cloning of hippocampal poly(A) RNA sequences that increase after entorhinal cortex lesion in adult rat. Brain Res. Mol. Brain Res. 9, 191–195 15 May, P.C. et al. (1990) Dynamics of gene expression for a hippocampal glycoprotein elevated in Alzheimer’s disease and in response to experimental lesions in rat. Neuron 5, 831–839 16 Poirier, J. et al. (1991) Astrocytic apolipoprotein E mRNA and GFAP mRNA in hippocampus after entorhinal cortex lesioning. Brain Res. Mol. Brain Res. 11, 97–106 17 Kalaria, R.N. et al. (2008) Alzheimer’s disease and vascular dementia in developing countries: prevalence, management, and risk factors. Lancet Neurol. 7, 812–826 18 Skoog, I. and Gustafson, D. (2006) Update on hypertension and Alzheimer’s disease. Neurol. Res. 28, 605–611 19 Thambisetty, M. et al. (2010) APOE epsilon4 genotype and longitudinal changes in cerebral blood flow in normal aging. Arch. Neurol. 67, 93–98 20 de Leon, M.J. et al. (2001) Prediction of cognitive decline in normal elderly subjects with 2-[(18)F]fluoro-2-deoxy-D-glucose/positronemission tomography (FDG/PET). Proc. Natl. Acad. Sci. U. S. A. 98, 10966–10971 21 Bondi, M.W. et al. (2005) fMRI evidence of compensatory mechanisms in older adults at genetic risk for Alzheimer disease. Neurology 64, 501–508 22 Corder, E.H. et al. (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261, 921–923 23 Forette, F. et al. (1998) Prevention of dementia in randomised doubleblind placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial. Lancet 352, 1347–1351 24 Hoffman, L.B. et al. (2009) Less Alzheimer disease neuropathology in medicated hypertensive than nonhypertensive persons. Neurology 72, 1720–1726 25 Peila, R. et al. (2001) Joint effect of the APOE gene and midlife systolic blood pressure on late-life cognitive impairment: the Honolulu–Asia aging study. Stroke 32, 2882–2889 26 Shah, K. et al. (2009) Does use of antihypertensive drugs affect the incidence or progression of dementia? A systematic review. Am. J. Geriatr. Pharmacother. 7, 250–261
476
Trends in Molecular Medicine Vol.16 No.10 27 Mulnard, R.A. et al. (2000) Estrogen replacement therapy for treatment of mild to moderate Alzheimer’s disease: a 1-year randomized controlled trial. JAMA 283, 1007–1015 28 Poirier, J. (2005) Apolipoprotein E, cholesterol transport and synthesis in sporadic Alzheimer’s disease. Neurobiol. Aging 26, 355–361 29 Hirsch-Reinshagen, V. et al. (2005) The absence of ABCA1 decreases soluble ApoE levels but does not diminish amyloid deposition in two murine models of Alzheimer disease. J. Biol. Chem. 280, 43243–43256 30 Blain, J.F. et al. (2004) A role for lipoprotein lipase during synaptic remodeling in the adult mouse brain. Neurobiol. Dis. 15, 510–519 31 Lutjohann, D. et al. (1996) Cholesterol homeostasis in human brain: evidence for an age-dependent flux of 24S-hydroxycholesterol from the brain into the circulation. Proc. Natl. Acad. Sci. U. S. A. 93, 9799– 9804 32 Leoni, V. (2009) Oxysterols as markers of neurological disease – a review. Scand. J. Clin. Lab Invest. 69, 22–25 33 Mulder, M. et al. (1998) Reduced levels of cholesterol, phospholipids, and fatty acids in cerebrospinal fluid of Alzheimer disease patients are not related to apolipoprotein E4. Alzheimer Dis. Assoc. Disord. 12, 198–203 34 Svennerholm, L. and Gottfries, C.G. (1994) Membrane lipids, selectively diminished in Alzheimer brains, suggest synapse loss as a primary event in early-onset form (type I) and demyelination in lateonset form (type II). J. Neurochem. 62, 1039–1047 35 Chochina, S.V. et al. (2001) Amyloid beta-peptide1-40 increases neuronal membrane fluidity: role of cholesterol and brain region. J. Lipid Res. 42, 1292–1297 36 Poirier, J. (2008) Apolipoprotein E represents a potent gene-based therapeutic target for the treatment of sporadic Alzheimer’s disease. Alzheimers Dement. 4, S91–97 37 Beffert, U. et al. (1998) The neurobiology of apolipoproteins and their receptors in the CNS and Alzheimer’s disease. Brain Res. Brain Res. Rev. 27, 119–142 38 Poirier, J. et al. (1993) Cholesterol synthesis and lipoprotein reuptake during synaptic remodelling in hippocampus in adult rats. Neuroscience 55, 81–90 39 Nathan, B.P. et al. (1994) Differential effects of apolipoproteins E3 and E4 on neuronal growth in vitro. Science 264, 850–852 40 Beffert, U. et al. (2004) Functions of lipoprotein receptors in neurons. J. Lipid Res. 45, 403–409 41 Posse De Chaves, E.I. et al. (2000) Uptake of lipoproteins for axonal growth of sympathetic neurons. J. Biol. Chem. 275, 19883–19890 42 Champagne, D. et al. (2002) Apolipoprotein E knockout mice display procedural deficits in the Morris water maze: analysis of learning strategies in three versions of the task. Neuroscience 114, 641–654 43 Mulder, M. et al. (2004) Low-density lipoprotein receptor-knockout mice display impaired spatial memory associated with a decreased synaptic density in the hippocampus. Neurobiol. Dis. 16, 212–219 44 Papassotiropoulos, A. et al. (2002) 24S-hydroxycholesterol in cerebrospinal fluid is elevated in early stages of dementia. J. Psychiatr. Res. 36, 27–32 45 Igbavboa, U. et al. (2002) A new role for apolipoprotein E: modulating transport of polyunsaturated phospholipid molecular species in synaptic plasma membranes. J. Neurochem. 80, 255–261 46 Klunk, W.E. et al. (1998) Metabolic alterations in postmortem Alzheimer’s disease brain are exaggerated by Apo-E4. Neurobiol. Aging 19, 511–515 47 Mason, R.P. et al. (1992) Evidence for changes in the Alzheimer’s disease brain cortical membrane structure mediated by cholesterol. Neurobiol. Aging 13, 413–419 48 Mauch, D.H. et al. (2001) CNS synaptogenesis promoted by gliaderived cholesterol. Science 294, 1354–1357 49 Dietschy, J.M. and Turley, S.D. (2004) Thematic review series: brain lipids. Cholesterol metabolism in the central nervous system during early development and in the mature animal. J. Lipid Res. 45, 1375– 1397 50 Utermann, G. et al. (1979) Polymorphism of apolipoprotein E. III. Effect of a single polymorphic gene locus on plasma lipid levels in man. Clin. Genet. 15, 63–72 51 Arendt, T. et al. (1997) Plastic neuronal remodeling is impaired in patients with Alzheimer’s disease carrying apolipoprotein epsilon 4 allele. J. Neurosci. 17, 516–529
Review 52 Sparks, D.L. et al. (1994) Induction of Alzheimer-like beta-amyloid immunoreactivity in the brains of rabbits with dietary cholesterol. Exp. Neurol. 126, 88–94 53 Refolo, L.M. et al. (2000) Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model. Neurobiol. Dis. 7, 321–331 54 Ghribi, O. et al. (2006) Deposition of iron and beta-amyloid plaques is associated with cortical cellular damage in rabbits fed with long-term cholesterol-enriched diets. J. Neurochem. 99, 438–449 55 Poirier, J. (2003) Apolipoprotein E and cholesterol metabolism in the pathogenesis and treatment of Alzheimer’s disease. Trends Mol. Med. 9, 94–101 56 Champagne, D. et al. (2003) The cholesterol-lowering drug probucol increases apolipoprotein E production in the hippocampus of aged rats: implications for Alzheimer’s disease. Neuroscience 121, 99–110 57 Fassbender, K. et al. (2001) Simvastatin strongly reduces levels of Alzheimer’s disease beta-amyloid peptides Abeta 42 and Abeta 40 in vitro and in vivo. Proc. Natl. Acad. Sci. U. S. A. 98, 5856–5861 58 Refolo, L.M. et al. (2001) A cholesterol-lowering drug reduces betaamyloid pathology in a transgenic mouse model of Alzheimer’s disease. Neurobiol. Dis. 8, 890–899 59 Bodovitz, S. and Klein, W.L. (1996) Cholesterol modulates alphasecretase cleavage of amyloid precursor protein. J. Biol. Chem. 271, 4436–4440 60 Giubilei, F. et al. (1990) Serum lipoprotein pattern variations in dementia and ischemic stroke. Acta Neurol. Scand. 81, 84–86 61 Moroney, J.T. et al. (1999) Low-density lipoprotein cholesterol and the risk of dementia with stroke. JAMA 282, 254–260 62 Dehouck, B. et al. (1997) A new function for the LDL receptor: transcytosis of LDL across the blood–brain barrier. J. Cell Biol. 138, 877–889 63 Waelsch, H. et al. (1940) A study of the synthesis and deposition of lipids in brain and other tissues with deuterium as an indicator. J. Biol. Chem. 135, 291–296 64 Chobanian, A.V. and Hollander, W. (1962) Body cholesterol metabolism in man. I. The equilibration of serum and tissue cholesterol. J. Clin. Invest. 41, 1732–1737 65 Roheim, P.S. et al. (1979) Apolipoproteins in human cerebrospinal fluid. Proc. Natl. Acad. Sci. U. S. A. 76, 4646–4649 66 Salen, G. et al. (1987) Increased concentrations of cholestanol and apolipoprotein B in the cerebrospinal fluid of patients with cerebrotendinous xanthomatosis. Effect of chenodeoxycholic acid. N. Engl. J. Med. 316, 1233–1238 67 Igbavboa, U. et al. (1997) Transbilayer distribution of cholesterol is modified in brain synaptic plasma membranes of knockout mice deficient in the low-density lipoprotein receptor, apolipoprotein E, or both proteins. J. Neurochem. 69, 1661–1667 68 Gibson Wood, W. et al. (2003) Amyloid beta-protein interactions with membranes and cholesterol: causes or casualties of Alzheimer’s disease. Biochim. Biophys. Acta 1610, 281–290 69 Riddell, D.R. et al. (2001) Compartmentalization of beta-secretase (Asp2) into low-buoyant density, noncaveolar lipid rafts. Curr. Biol. 11, 1288–1293 70 Kojro, E. et al. (2001) Low cholesterol stimulates the nonamyloidogenic pathway by its effect on the alpha-secretase ADAM 10. Proc. Natl. Acad. Sci. U. S. A. 98, 5815–5820 71 Xiu, J. et al. (2006) Influence of cholesterol and lovastatin on alphaform of secreted amyloid precursor protein and expression of alpha7 nicotinic receptor on astrocytes. Neurochem. Int. 49, 459–465 72 Abad-Rodriguez, J. et al. (2004) Neuronal membrane cholesterol loss enhances amyloid peptide generation. J. Cell Biol. 167, 953–960 73 Puglielli, L. et al. (2001) Acyl-coenzyme A: cholesterol acyltransferase modulates the generation of the amyloid beta-peptide. Nat. Cell Biol. 3, 905–912 74 Pani, A. et al. (2009) Altered cholesterol ester cycle in skin fibroblasts from patients with Alzheimer’s disease. J. Alzheimers Dis. 18, 829–841 75 Beffert, U. et al. (1999) Apolipoprotein E and beta-amyloid levels in the hippocampus and frontal cortex of Alzheimer’s disease subjects are disease-related and apolipoprotein E genotype dependent. Brain Res. 843, 87–94
Trends in Molecular Medicine
Vol.16 No.10
76 Riddell, D.R. et al. (2008) Impact of apolipoprotein E (ApoE) polymorphism on brain ApoE levels. J. Neurosci. 28, 11445–11453 77 Bales, K.R. et al. (2009) Human APOE isoform-dependent effects on brain beta-amyloid levels in PDAPP transgenic mice. J. Neurosci. 29, 6771–6779 78 Rebeck, G.W. et al. (1993) Apolipoprotein E in sporadic Alzheimer’s disease: allelic variation and receptor interactions. Neuron 11, 575– 580 79 Holtzman, D.M. et al. (1999) Expression of human apolipoprotein E reduces amyloid-beta deposition in a mouse model of Alzheimer’s disease. J. Clin. Invest. 103, R15–21 80 Beffert, U. et al. (1999) Apolipoprotein E isoform-specific reduction of extracellular amyloid in neuronal cultures. Brain Res. Mol. Brain Res. 68, 181–185 81 Yang, D.S. et al. (1999) Apolipoprotein E promotes the binding and uptake of beta-amyloid into Chinese hamster ovary cells in an isoform-specific manner. Neuroscience 90, 1217–1226 82 Deane, R. et al. (2008) apoE isoform-specific disruption of amyloid beta peptide clearance from mouse brain. J. Clin. Invest. 118, 4002– 4013 83 Evans, K.C. et al. (1995) Apolipoprotein E is a kinetic but not a thermodynamic inhibitor of amyloid formation: implications for the pathogenesis and treatment of Alzheimer disease. Proc. Natl. Acad. Sci. U. S. A. 92, 763–767 84 Holtzman, D.M. et al. (2000) Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A. 97, 2892–2897 85 Morris, J.A. et al. (1999) The genomic organization and polymorphism analysis of the human Niemann-Pick C1 gene. Biochem. Biophys. Res. Commun. 261, 493–498 86 Ohm, T.G. et al. (2003) Cholesterol and tau protein – findings in Alzheimer’s and Niemann Pick C’s disease. Pharmacopsychiatry 36 (Suppl. 2), S120–126 87 Distl, R. et al. (2001) Tangle-bearing neurons contain more free cholesterol than adjacent tangle-free neurons. Acta Neuropathol. 101, 547–554 88 Chauhan, N.B. (2006) Effect of aged garlic extract on APP processing and tau phosphorylation in Alzheimer’s transgenic model Tg2576. J. Ethnopharmacol. 108, 385–394 89 Jick, H. et al. (2000) Statins and the risk of dementia. Lancet 356, 1627–1631 90 Wolozin, B. et al. (2007) Simvastatin is associated with a reduced incidence of dementia and Parkinson’s disease. BMC Med. 5, 20 91 Haag, M.D. et al. (2009) Statins are associated with a reduced risk of Alzheimer disease regardless of lipophilicity. The Rotterdam Study. J. Neurol. Neurosurg. Psychiatry 80, 13–17 92 Li, G. et al. (2007) Statin therapy is associated with reduced neuropathologic changes of Alzheimer disease. Neurology 69, 878–885 93 Goldstein, J.L. and Brown, M.S. (1990) Regulation of the mevalonate pathway. Nature 343, 425–430 94 Enjalbert, M. et al. (1980) Usefulness of probucol in treating primary hypercholesterolemia. Can. Med. Assoc. J. 123, 754–757 95 Liang, Y. et al. (2004) A liver X receptor and retinoid X receptor heterodimer mediates apolipoprotein E expression, secretion and cholesterol homeostasis in astrocytes. J. Neurochem. 88, 623–634 96 Riddell, D.R. et al. (2007) The LXR agonist TO901317 selectively lowers hippocampal Abeta42 and improves memory in the Tg2576 mouse model of Alzheimer’s disease. Mol. Cell. Neurosci. 34, 621–628 97 Feldman, H.H. et al. (2010) Randomized controlled trial of atorvastatin in mild to moderate Alzheimer disease: LEADe. Neurology 74, 956–964 98 Arvanitakis, Z. and Knopman, D.S. (2010) Clinical trial efforts in Alzheimer disease. Why test statins? Neurology 74, 945–946 99 Poirier, J. (2002) Cholesterol transport and synthesis are compromised in the brain in sporadic Alzheimer’s disease: from risk factors to therapeutic targets. In Alzheimer’s Disease and Related Disorders Annual (Gauthier, S. and Cummings, J.L., eds), pp. 1–23, Dunitz 100 Thompson, P.M. et al. (2003) Dynamics of gray matter loss in Alzheimer’s disease. J. Neurosci. 23, 994–1005
477
APOE and cholesterol homeostasis in Alzheimer’s disease Vale´rie Leduc1, Ste´phanie Jasmin-Be´langer1 and Judes Poirier1,2 1 2
Douglas Mental Health University Institute, 6875 Lasalle, Montreal (Verdun), Quebec, H4H 1R3, Canada Center for Studies in Aging, McGill University, Verdun, Quebec, H4H 1R3, Canada
Converging evidence from clinical and pathological studies indicate the presence of important relationships between the ongoing deterioration of brain lipid homeostasis, vascular changes and the pathophysiology of sporadic Alzheimer’s disease (AD). These associations include the recognition of cholesterol transporters apolipoprotein E (APOE), APOC1 and APOJ as major genetic risk factors for common AD and observations associating risk factors for cardiovascular disease such as high midlife plasma cholesterol, diabetes, stroke, obesity and hypertension to dementia. Moreover, recent clinical findings lend support to the notion that progressive deterioration of cholesterol homeostasis in AD is a central player in the disease pathophysiology and is, therefore, a potential therapeutic target for disease prevention. Alzheimer’s disease and cholesterol homeostasis Dementias are progressive neurodegenerative disorders characterized by a decline in cognitive functions beyond what might be expected from normal aging. Of these dementias, Alzheimer’s disease (AD) is the most common type among the elderly accounting for approximately 60% of all dementia cases diagnosed [1]. Pathologically, AD is characterized by the presence of extracellular amyloid plaques, intracellular neurofibrillary tangles (Box 1) as well as extensive neuronal and synaptic losses [2]. Although gradual atrophy of the brain is observed, degeneration occurs preferentially in neuronal populations of cholinergic phenotype [3]. There are several reasons why these particular neurons might be more vulnerable to AD, notably their high-energy requirement, their reliance on axonal transport for sustained function and trophic support, and their large and poorly protected (sparsely myelinated) cell surface area that increases their exposure to toxic environmental conditions and oxidative stresses [3,4]. Accordingly, neuronal loss usually starts in brain structures rich in long cholinergic fibers, that is the entorhinal cortex and hippocampus of the limbic system, spreads to the temporal lobe, then to the frontal cortex and ultimately to primary sensory and visual areas [3,4]. Etiologically, AD is considered a multifactorial disease with a strong genetic component. The disease can be subdivided into two distinct categories, the so-called familial and sporadic forms of the disease. The identification of specific gene mutations has provided new insight into the molecular changes responsible for the pathophysiological Corresponding author: Poirier, J. ([email protected])
processes of AD. Mutations in the genes encoding the amyloid precursor protein (APP), presenilin 1 and presenilin 2 associate with rare early-onset forms of familial AD [5]. However, coding mutations in these genes do not increase the risk of the more prevalent late-onset disease, which accounts for more than 95% of all AD cases worldwide [1]. APOE is the only known locus that affects the risk of the late-onset form of the disease, with the e4 allele (encoding apolipoprotein 4, APOE4) and the e2 allele (encoding APOE2) increasing and decreasing the risk level, respectively [6,7]. Recently, several independent genomewide association studies (GWAS) in homogeneous and heterogeneous populations of AD with age-matched control cases in North America, Europe and Asia associated the e4 allele with AD [8–12]. Surprisingly, the remaining genetic associations uncovered in these GWAS failed to replicate across studies, except for the gene encoding APOJ, which is an accessory protein to APOE in the maintenance of cholesterol homeostasis. APOE coordinates the mobilization and redistribution of cholesterol in repair, growth and maintenance of myelin and neuronal membranes during development or after injury in the peripheral nervous system [13,14]. In the central nervous system (CNS), APOE in partnership with APOJ and APOC1 plays a pivotal role in cholesterol delivery during the membrane remodeling associated with synaptic turnover and dendritic reorganization [15,16]. The near complete absence of other key plasma apolipoproteins such as APOA1 and APOB in the brain further emphasizes the critical and unique role of APOE for cholesterol transport in the normal or injured CNS. Cardiovascular contributions to AD pathology The prevalence and incidence of degenerative and vascular dementias increase almost exponentially with age, from 70 years onward. In view of the increasing longevity of humans, both types of dementia have progressively evolved into major public health problems worldwide [17]. The integrity of the cerebral vasculature is crucial for the maintenance of cognitive function; vascular factors, such as hypertension, myocardial infarction, diabetes, obesity, hyper- or dyslipidemia, ischemic white matter lesions and generalized atherosclerosis, associate with dementia and cognitive decline [18]. The connection between vascular factors and cognition remain largely unknown, although recent evidence associates the e4 allele with regional cerebral blood flow (rCBF) disturbances in regions of the brain affected by AD [19]. Indeed, apparently
1471-4914/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2010.07.008 Trends in Molecular Medicine, October 2010, Vol. 16, No. 10
469
Review Box 1. Neurofibrillary tangles One other famous pathological feature of AD is the neurofibrillary tangle. These intraneuronal filamentous deposits primarily result from the hyperphosphorylation of the microtubule-stabilizing Tau proteins and elicit neuronal death by disrupting axonal transport and inducing widespread metabolic decline [100].
clinically normal APOE4 carriers 50 years and older are more likely to show significant cognitive impairment than APOE4-negative subjects and exhibit greater decline in rCBF over time by position-emission tomography [20]. Interestingly, APOE4 carriers have higher baseline rCBF than noncarriers in regions exhibiting significant longitudinal decline in rCBF, suggesting that APOE4 carriers might have to exert greater cognitive effort to perform similar to their noncarrier counterparts [21]. However, because relatively young healthy adult carriers of APOE4 (315 years) also exhibit greater baseline rCBF than noncarriers, these increments might reflect mechanisms to compensate for inefficient neuronal processes, such as reduced synaptic plasticity and neuronal growth in APOE4 carriers. Accordingly, the e4 allele dose impacts the age of onset in both familial and sporadic cases [6,22], again suggesting the possibility that APOE4 carriers are developmentally different from their noncarrier counterparts. The emergence of decreased CBF and hypertension as major risk factors for sporadic AD has garnered interest from a therapeutic angle [23,24]. Epidemiological associations between common AD and vascular disease risk factors could reflect an overdiagnosis of AD in individuals with silent cerebrovascular disease or alternatively cerebrovascular disease could affect the clinical expression and onset of AD. Further possibilities include: AD might increase the risk of vascular diseases; vascular disease might silently stimulate AD processes; or similar mechanisms, such as disturbances in lipid homeostatic processes and abnormal cholesterol transport, distribution or accumulation could contribute to the pathogenesis of both disorders. According to several longitudinal studies, hypertension predisposes individuals to cognitive impairment and ensuing dementia after a period varying from a few years to several decades [25]. The e4 allele drastically enhances the dementia risk level in hypertensive subjects [25]. Conversely, antihypertensive drugs, such as angiotensin-converting enzyme inhibitors and diuretics, markedly reduce the risk of cognitive impairment and in some studies delay the onset of AD itself [26]. Estrogen, which lowers serum cholesterol levels and the risk of coronary heart disease, decreases the incidence of dementia in cross-sectional epidemiological studies but does not affect disease progression [27] in diagnosed mild-to-moderate AD. Cholesterol synthesis, transport and degradation Cholesterol homeostasis in the brain is carefully maintained through a series of interdependent processes that include synthesis, storage, degradation and transport. Cholesterol and other lipids are used for membrane synthesis and for many other anabolic or catabolic activities by cells throughout the body including those of the CNS, a site 470
Trends in Molecular Medicine Vol.16 No.10
of high lipid turnover. The cholesterol requirements of most mammalian cells are met by two separate but inter-related processes. One process is the endogenous synthesis of cholesterol, which involves over 20 reactions and is regulated primarily by the activity of the 3-hydroxy3-methylglutaryl coenzyme A reductase (HMGCR) that catalyzes the formation of mevalonate, the key precursor molecule in the synthesis of cholesterol (Figure 1). Brain cells, particularly astrocytes and neurons, cultured in vitro synthesize cholesterol at a rate that is inversely proportional to the cholesterol content in the growth environment [28]. The other central process regulating cholesterol levels in the brain involves the utilization of lipoprotein-derived cholesterol, following internalization of a lipoprotein bound to its surface receptor [usually an APOE-rich high-density lipoprotein (HDL)-like complex]. The ATPbinding cassette transporter types A1 and G1 (ABCA1/ ABCG1) coordinate the mobilization of cholesterol from cytoplasmic pools to the cell surface membrane [29], where it is eventually combined with APOE, APOJ and phospholipids to produce a functional HDL-like particle through the action of the lipoprotein lipase (LPL) [30]. Newly synthesized lipoprotein particles are released into the extracellular space where they migrate, in a gradientdriven manner, toward both APOE and APOJ receptors on ependymal cells surrounding the ventricles or toward APOE receptors on neuronal and glial cells. Following binding of the APOE lipoprotein particle to neuronal surface low-density lipoprotein receptor (LDLR), the APOE lipoprotein/LDLR complex is internalized and degraded, releasing cholesterol that can be used for synapse formation and terminal proliferation (Figure 2). Excess cholesterol can also be converted into the more hydrophilic oxysterol 24S-hydroxycholesterol by the neuron-specific enzyme cholesterol 24S-hydroxylase, which is encoded by CYP46A1. This oxysterol leaves the brain via the blood–brain barrier (BBB) and directly enters the bloodstream [31]. Thus, almost all the 24S-hydroxycholesterol present in human circulation has a cerebral origin [32]. Cholesterol levels in the different brain areas in AD have been documented over the past 25 years. Brain cholesterol levels (total, free and esterified cholesterol) are significantly reduced in hippocampal and cortical areas in AD patients compared to age-matched neurological controls but no differences are detected in the pathologyfree cerebellum [33,34]. Cholesterol levels thus differ depending on the brain region examined and rather interestingly the brain areas that are more susceptible to AD are those that contain the greatest amounts of synaptic membrane cholesterol, that is hippocampus>cortex> cerebellum [35]. Consistent with these observations, HMGCR activity in the brain of autopsy-confirmed AD cases is significantly reduced in cortical and hippocampal areas as compared to age-matched control subjects [36]. The reduction of HMGCR enzymatic activity is, however, APOE-genotype independent. HMGCR mRNA levels are not different in AD versus control subjects but are inversely correlated with APOE protein levels, suggesting a compensatory upregulation of cholesterol internalization by surviving cortical and hippocampal neurons [36].
(Figure_1)TD$IG][ Review
Trends in Molecular Medicine
Vol.16 No.10
Figure 1. Cholesterol synthesis in eukaryotic cells. Schematic representation of the mevalonate pathway and key enzymes regulating cholesterol synthesis and esterification. The isoprenoid cascade (dotted box) provides a biochemical association through which cholesterol synthesis can indirectly modulate production of phosphotau in the CNS. Abbreviations: GGPP, geranylgeranyl diphosphate; CdK5, cyclin-dependent kinase 5; PKA, protein kinase A; GSK3B, glycogen synthase kinase 3 beta; HMG CoA, 3-hydroxy-3-methylglutaryl-CoA; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
Ultrastructural studies have shown that with the loss of neuronal input to the hippocampus, astrocytes and microglia progressively engulf both presynaptic terminals and preterminal axons (Figure 2) to rapidly clear the area and allow synaptic replacement [37]. Once metabolized, these terminal-derived ovoids generate a large glial store of lipids that are readily available for the synthesis of the membrane components necessary for new synapses and dendrites (Figure 2) in surviving adjacent neurons [37]. The local accumulation of high concentrations of cellular cholesterol in astrocytes induces APOE and APOJ synthesis and secretion in combination with cholesterol (Figure 2). As the intracellular concentration of cholesterol increases in glial cells, cholesterol synthesis is progressively repressed at the level of HMGCR [38]. Deafferented granular cell neurons in the hippocampus exhibit marked increases of LDLR during the early and middle phases of the reinnervation process (Figure 2) [38]. Both the LDLR-related protein (LRP) and the APOE receptor type 2 expressed by neurons indirectly modulate terminal proliferation and axonal extension [39], primarily through a signaling pathway [40] in contrast to the APOE/LDLR, which uses direct lipoprotein internalization mechanisms. In response to the increased internalization of cholesterol via the APOE/LDLR pathway,
neurons of the injured hippocampal formation in AD reduce their HMGCR activity (Figure 2). The apparent contradiction of depressed cholesterol synthesis and increased APOE expression in the presence of active synaptogenesis can be reconciled by postulating that cholesterol from degenerating terminals is salvaged through the APOE transport/LDLR uptake pathway [38,41] and reutilized. The absence of either APOE or LDLR in deficient (knockout) mice compromises synaptic integrity and compensatory synaptogenesis and causes age-dependent cognitive deficits, presumably owing to an impaired ability of the brain to produce or bind to functional HDL-like particles [42,43]. Neurodegeneration combined with brain atrophy also associate with significantly reduced concentrations of other lipid products. Compared to age-matched neurological control subjects, patients with AD have marked increases and decreases in 24S-hydroxycholesterol in cerebrospinal fluid (CSF) and circulation, respectively [44]. However, genetic association studies examining the relationship between the CYP46A1 gene and sporadic AD are in conflict. Similarly, the phospholipid content of membrane structures in AD brains is reduced compared to age-matched neurological control subjects [34]. APOE, which is the main phospholipid carrier in the mature CNS, affects 471
(Figure_2)TD$IG][ Review
Trends in Molecular Medicine Vol.16 No.10
Figure 2. Cholesterol recycling in the injured CNS. Degenerating nerve terminals are internalized and degraded. Nonesterified cholesterol (i) is used as free cholesterol (FC) for the assembly of an APOE+APOJ/cholesterol/lipoprotein complex (ii) or converted into cholesterol esters (CEs) for storage. The APOE+APOJ/cholesterol/lipoprotein complexes are then directed either into circulation, presumably through the ependymal cells surrounding the ventricles, and/or to specific brain cells requiring lipids. APOE+APOJ complexes are internalized by the neuronal LDLR pathway (iii) and the cholesterol is released (iv) for dendritic proliferation and/or synaptogenesis (vi). As a consequence of internalization, cholesterol synthesis in neurons via the HMGCR pathway (v) is progressively repressed. Abbreviations: E, APOE; J, APOJ; PL, phospholipids; CE, cholesterol ester; FC, free cholesterol. ACAT, acetyl coenzyme A cholesterol acyltransferase; ABCA1/G1, ATP binding cassette transporter A1 and G1; LPL, lipoprotein lipase; AA, amino acids; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
phospholipids concentrations in an e4 dose-dependent manner [45,46]. Interestingly, the cholesterol-to-phospholipids ratio in cortical lipid membranes is markedly decreased in the temporal and frontal cortices of AD subjects as compared to age-matched neurological control subjects [47], consistent with pronounced thinning of the plasma membrane observed in cortical areas. Cholesterol is a vital component of the CNS and is essential for axonal growth, and synaptic formation and remodeling, processes that are crucial for learning, formation of memories and neuronal repair [48]. The CNS has thus evolved two strategies to ensure adequate supplies of cholesterol: de novo synthesis and the internalization of lipoprotein particles. Synapses and dendrites are located far away from the cell body where cholesterol synthesis takes place; thus, the internalization of lipoproteins is imperative for neurons to meet their cholesterol needs [49]. Unsurprisingly, cholesterol and lipid levels are decreased in AD. Because APOE proteins are expressed at different levels according to the gradient E2>E3>E4 [50], concentration differences could explain why the APOE4 variant confers a huge risk for developing AD and results in such poor synaptic remodeling capacity [51]. 472
Cholesterol homeostasis and APP metabolism: a role for APOE-mediated beta-amyloid (Ab) transport and degradation Dominant mutations within the APP gene that alter the processing of the APP protein can result in AD. Although cleavage of the APP by a-secretase precludes the formation of Ab peptides, consecutive cleavage of APP by b-secretase and g-secretase stimulates Ab production, resulting in the accumulation of Ab peptides that eventually aggregate as extracellular amyloid plaques, one of the landmark pathologies of AD. Since Sparks’ initial observation that feeding rabbits a cholesterol-enriched diet for eight weeks leads to an accumulation of intracellular Ab immunoreactivity in hippocampal neurons [52], studies associating cholesterol and Ab metabolism have overrun the field of AD research. The finding that cholesterol enrichment enhances Ab production has been widely replicated in animals [53,54] and cultured-cell models strongly suggesting that lipids, especially cholesterol, modulate Ab production. Accordingly, lowering cholesterol levels apparently decreases Ab production [55–59]. Exposure to cholesterol-lowering agents such as probucol [55,56], statins [57], BM15.766 [58] and cyclodextrin [57,59] reduces the amyloid load in
Review cultured cells and in vivo mammalian models. On the one hand, the mechanisms by which probucol and statins reduce Ab load could be pleiotropic. Probucol affects cholesterol metabolism through multiple mechanisms but most probably induces APOE synthesis [56]. On the other hand, BM15.766 inhibits the enzyme catalyzing the final step of cholesterol biosynthesis, namely 7-dehydrocholesterol reductase, suggesting that the amyloid-reducing effect might involve sterol depletion [58]. These findings further support clinical and epidemiological evidence suggesting that individuals with elevated midlife plasma cholesterol have increased susceptibility to dementia and AD when compared to normal cholesterolemic subjects [60,61]. The mechanism by which changes in plasma cholesterol affect AD and Ab production in the brain is, however, obscure. Although brain endothelial cells can transport peripheral LDL cholesterol through LDLR-mediated uptake [62], this pathway appears to be of little importance under normal conditions. Indeed, owing to the BBB, cholesterol metabolism in the brain is regulated independently from the periphery; isotope-labeled cholesterol or deuterium do not accumulate in the brain [63,64]. Furthermore, the absence of APOB, the main apolipoprotein associated with LDL particles, within the CNS [65] argues against a significant transport of LDL cholesterol across the BBB. However, under pathological conditions, small amounts of cholesterol might enter the CNS. A small but significant increase in brain cholesterol was observed in hypercholesterolemic transgenic mice [53]. Additionally, cholestanol and APOB abnormally accumulate in the CSF of patients with cerebrotendinous xanthomatosis, a rare disorder engendered by a defect in the biosynthesis of bile acids [66]. If increasing cholesterol levels promotes Ab generation, how can AD associate with decreased total cholesterol levels and increased Ab levels in the brain? Total cholesterol measurements only mirror bulk cholesterol within the CNS, which is highly compartmentalized. Hence, cholesterol levels in specific subcellular compartments might be a more important factor in the pathophysiology of AD than bulk cholesterol. Indeed, cholesterol is asymmetrically distributed in the exofacial and cytofacial leaflets of cell membranes [67] and is not evenly distributed in cellular organelles [68]. Additionally, lipid domains enriched in cholesterol such as lipid rafts have been identified in the plasma membrane. Thus, changes in cholesterol distribution or the formation of cholesterol-rich domains might not be reflected in the total amount of brain cholesterol. Accordingly, the selective removal of cholesterol from the plasma membranes of hippocampal and mixed cortical neurons by cyclodextrin is sufficient to inhibit Ab production [57], pinpointing lipid rafts as potential critical elements for Ab production. In agreement with this concept, lipid rafts – membrane microdomains rich in lipids and cholesterol – modulate APP processing. Both g-secretase and b-secretase BACE-1 activities have been isolated from lipids rafts [69], whereas the majority of a-secretase ADAM10 activity occurs in nonraft regions of the membrane [70]. Accordingly, increased intracellular cholesterol levels, which favor lipid raft formation, negatively regulate
Trends in Molecular Medicine
Vol.16 No.10
a-secretase activity [71] but stimulate both b- and g-secretase activities, resulting in an increase in Ab production. This raft/nonraft model of APP processing might help to explain cholesterol-dependent regulation of Ab production, but the explanation is imperfect as a modest reduction in cellular cholesterol paradoxically accelerates b-secretase cleavage and Ab production [72]. In line with these findings, total cholesterol might not regulate secretase activity but rather the ratio of free cholesterol (FC) to cholesterol esters (CEs), which is regulated by acyl coenzyme A cholesterol acyltransferase (ACAT) [73]. Moreover, ACAT activity correlates with changes in a-, b- and g-secretase APP products [73], indicating that ACAT activity affects APP processing. Whether ACAT activity modulates Ab generation by directly modulating the activity of the secretases or the availability of APP for cleavage through changes in cholesterol distribution and compartmentalization remains to be clarified. Nevertheless, because these effects are only observed in cells containing physiological or supraphysiological levels of FC, a role for FC in Ab modulation cannot be ruled out [73]. The pharmacological inhibition of ACAT causes a concomitant increase in FC levels in the cytoplasm and a reduction in both CE levels and Ab production [74], similar to the findings that APOE, which promotes cholesterol transport and internalization in an APOE2>APOE3>>APOE4 manner, strongly modulates Ab burden [75–77]. Numerous studies show a positive correlation between amyloid burden and e4 allele dose. Indeed, the e4 allele associates with increased plaque density in humans [78] and in transgenic mice [76,77,79]. The exact mechanism by which APOE4 promotes Ab production and accumulation remains elusive, although the isoform-dependent modulation of the local steady-state levels of APOE remains the primary suspect, as APOE concentrations follow an E2>E3>E4 gradient [36,77]. Because APOE acts as a powerful Ab scavenger in the extracellular space, low steady-state levels associated with the presence of the e4 allele are bound to affect the normal removal of the extracellular peptide. Numerous in vitro studies have demonstrated that human APOE facilitates cellular Ab uptake and degradation [80,81]. Additionally, APOE promotes brain to blood removal of Ab peptides by transport across the BBB, a clearing process that follows an APOE2>APOE3>APOE4 gradient [82]. Whereas the predominant role of APOE in the scavenging of Ab in the CNS is generally accepted and well documented, the presumed effect of APOE on Ab aggregation is controversial. In vitro studies suggest that APOE isoforms, with APOE4 being the least effective, might decrease Ab fibrillogenesis by interfering with Ab nucleation [83]. These observations contrast with animal studies, where the knockdown of the murine ApoE gene drastically decreases Ab deposition, along with a marked loss of thioflavin-S-positive amyloid plaques, which are plaques containing amyloid in the form of fibrils [84]. However, transgenic mice expressing human APOE show a marked delay in Ab deposition and plaque formation relative to mice expressing murine APOE or no APOE [79]. This suggests that human APOE exhibits physiological properties that are different from murine APOE with regard to amyloid metabolism, removal and/or deposition in the mature brain. 473
Review Cholesterol homeostasis disturbances occur in AD and modulate Ab production, but the presence of the BBB, the complexity of the biosynthesis and regulation of cholesterol and the lack of specific and sensitive methods to quantify brain cholesterol levels at subcellular resolution impedes our understanding of the global mechanism involved. Although it has long been recognized that brain cholesterol is independent from changes in plasma cholesterol, growing evidence suggests the contrary. Changes in cholesterol domains rather than bulk cholesterol levels could be involved in AD pathology and Ab modulation, but other key factors such as CE levels and ACAT should not be underestimated. Cholesterol metabolism: a potential therapeutic target for the treatment of sporadic AD Under normal circumstances, cholesterol synthesis via the HMGCR pathway (Figure 1) is required only when the amount of lipoprotein internalization by the APOE/LDLR pathway is insufficient to meet the cholesterol requirement of a cell. To maintain cellular cholesterol homeostasis, a potent negative feedback on HMGCR activity (Figure 1) and gene expression decreases cholesterol synthesis in response to excess intracellular sterol internalization via members of the LDLR family. This first and most important feedback on HMGCR occurs through a decrease in gene
Trends in Molecular Medicine Vol.16 No.10
transcription or the modulation of the translational efficiency of HMGCR; the latter can decrease or increase protein levels. Interestingly, a hereditary cholesterol storage disorder known as Niemann–Pick C (NPC) shows Alzheimer-like tau pathology in youth or adolescence without the amyloid plaque deposition. NPC results from a mutation in either of two genes, NPC1 or NPC2/HE1, and associates with the intracellular misrouting and accumulation of FC [85]. Tangle-bearing neurons in NPC subjects display high levels of cholesterol that resemble the tangle-bearing neurons in AD [86,87]. The administration of HMGCR inhibitors (statins) in NPC mice, which normally carry high concentrations of phospho-tau in the brain, reduces brain phospho-tau to near normal levels [86]. Figure 1 illustrates the geranylgeranyl diphosphate cascade believed to be responsible, at least in part, for the coordinated phosphorylation of tau in neurons [86,88]. Evidence from multiple prospective epidemiological studies, suggesting that the presymptomatic use of statins in middle-age subjects confers some levels of protection against sporadic AD later in life [89–91], provides us with a potentially interesting biochemical target for prevention studies. Furthermore, a recent populationbased study examining the effect of statins in the cognitively intact elderly revealed that over time, statin users
[(Figure_3)TD$IG]
Figure 3. An APOE4/amyloid hypothesis of AD pathophysiology. The sequence of pathogenic events leading to neuronal cell loss and synaptic damage is based on the well-established amyloid cascade hypothesis (blue background), which proposed that the accumulation of beta amyloid in the brain is the primary influence driving AD pathology. The different modulators of beta amyloid metabolism affecting lipid homeostasis, such as APOE, HMGCR and APOJ, have been added to the cascade. Finally, the emerging roles of cholesterol and HMGCR have been put in perspective in relation to their respective contribution to the pathophysiology of AD. Abbreviations: +, enhancer, , inhibitor; CLA, cholesterol-lowering agent; APP, amyloid precursor protein; PS1/PS2, presenilin 1/2; HMGCR, 3-hydroxy-3-methylglutaryl-CoA reductase.
474
Review accumulate significantly less neurofibrillary tangles (but not amyloid plaques) when compared to age-matched statin-free subjects [92]. Together, these data point toward HMGCR as a potential key player in the pathophysiology of tau phosphorylation and tangle deposition in the aging and diseased brain. Statins competitively inhibit HMGCR activity, thereby inhibiting the synthesis of cholesterol by preventing the conversion of HMG CoA to mevalonate. This family of inhibitors reduces the formation and internalization of LDL cholesterol particles in circulation and upregulates LDLR activity at the cell surface of liver cells, thereby reducing serum LDL cholesterol and triglycerides and increasing HDL cholesterol concentration [93]. It is unclear whether the beneficial effects of statins are mediated through BBB penetration or via alterations of plasma cholesterol levels. In an attempt to clarify this issue, two large prospective population studies examined the effects of lipophilic versus lipophobic statins on AD risk. Because only lipophilic statins are believed to cross the BBB, a beneficial effect of lipophilic over lipophobic compounds is expected if this effect is mediated through direct modulation of brain cholesterol levels. One study detected preferential benefits from lipophilic compounds, such as simvastatin, over lipophobic ones [90], whereas the other did not [91]. Interestingly, however, both reconfirmed the association between statin use and reduced risk for AD. Hence, it remains unclear how statins exert their beneficial effect against AD and whether the effect is mediated through modulation of brain and/or plasma cholesterol levels. In addition to the statin-mediated LDLR upregulation, statins might induce APOE synthesis and secretion, thereby facilitating both lipoprotein delivery and internalization in neurons and Ab removal from the extracellular space. The potential use of an APOE-inducer drug in the treatment of AD has been reviewed elsewhere [36]. Briefly, two major approaches have been proposed to facilitate cholesterol mobilization and transport in AD, particularly in e4 allele carriers. One proposes to induce APOE synthesis and secretion in the brain with probucol [36,56], a cholesterol-lowering agent used in the mid 1980 s to treat familial hypercholesterolemia [94]; the other strategy proposes to administer liver X receptor agonists that stimulate ABCA1, APOE and the LDLR production through a common signaling pathway [95,96]. These independent and yet complementary approaches indicate that the modulation of brain cholesterol homeostasis might interfere with disease onset in subjects exposed to cholesterol-lowering agents prior to the diagnosis of AD. Unfortunately, the most recent large-scale, randomized double-blind placebo-controlled clinical trials with atorvastatin [97] and simvastatin [98] failed to slow disease progression in mild-to-moderate AD, suggesting that statins provide beneficial effects only in the presymptomatic stages of the disease. It is still unclear whether the risk-reducing effect of statins is specific to dementia of the Alzheimer’s type or more general. It is possible that the beneficial effect of cholesterol-lowering agents stems from a completely distinct pathway involving an independent cardiovascular risk factor (other than cholesterol) or other
Trends in Molecular Medicine
Vol.16 No.10
pleiotropic effects (i.e. increased endothelial nitric oxide synthase activity, decreased endothelin-1 activity or antiinflammatory activity) that modulate disease onset in atrisk individuals (i.e. e4 allele carriers). In this scenario, cholesterol-lowering agents would indirectly prevent vascular risk factors, such as circulating levels of cholesterol or atherosclerotic plaque deposition, from modulating the age of onset in AD. Although it remains difficult to determine the exact mode of action by which cholesterol-lowering agents affect the pathophysiology of AD, the recent discovery of polymorphic genetic variants in the HMGCR gene [99] provide us with a possible explanation as to what could be the molecular basis and target for the beneficial effect against AD. Recent preliminary evidence suggests that some of these polymorphisms affect either HMGCR splicing efficiency or the production of abnormal spliced variants that indirectly affects enzymatic activity in the brain [99]. Is there a unifying hypothesis of AD pathophysiology? Although age is a key determinant in AD pathology, genetic risk factors play a central role in this process (Figure 3). Factors such as APOE4, ACAT, APOJ, LRP, LPL or paraoxonase 1 all modulate cholesterol homeostasis and their genes are risk factors for common AD. In recent years, it has become obvious that traditional vascular risk factors such as hypertension, myocardial infarction, diabetes, obesity, hyper- or dyslipidemia, ischemic white matter lesions and generalized atherosclerosis, definitively associate with the common form of dementia as well as cognitive decline associated with normal aging. In contrast to the aforementioned genetic risk factors, it is possible to pharmacologically intervene to reduce and even eliminate many of the vascular factors associated with AD. In practice, this resulted in the absence of cognitive decline of placebo groups in nearly all of the 6-month randomized clinical trials performed in AD these past five to six years. However, this does not stop disease progression over the long term. Although indirect, these replicated observations (from multiple clinical studies performed by several independent pharmaceutical corporations) clearly show that, similar to the APOE4 findings, these vascular risk factors (including circulating cholesterol levels) significantly contribute to AD pathophysiology (Figure 3) but most probably do not cause the disease itself. In the same context, we believe that strategies designed to minimize or eliminate the physiological impact of the risk factors in the decades preceding the onset of AD in ‘‘at-risk’’ subjects will delay the onset of the symptoms by several years but will not prevent the disease itself. That being said, it has been estimated that a delay in the onset of AD by five years would result in a 50% reduction of the incidence within one generation. Although this approach might not cure the disease, it would definitively impact its overall socioeconomic burden. Acknowledgements We wish to thank our long-time collaborators for their intellectual input to this manuscript as well as the funding agencies supporting this research: the Canadian Institute for Health Research, Natural Sciences and Engineering Research Council of Canada, the Fonds de la recherche en sante´ du Que´bec and the Alzheimer Society of Canada. 475
Review References 1 Terry, R.D. (2006) Alzheimer’s disease and the aging brain. J. Geriatr. Psychiatry Neurol. 19, 125–128 2 Adlard, P.A. and Cummings, B.J. (2004) Alzheimer’s disease – a sum greater than its parts? Neurobiol. Aging 25, 725–733 3 Mattson, M.P. and Magnus, T. (2006) Ageing and neuronal vulnerability. Nat. Rev. Neurosci. 7, 278–294 4 Braak, H. et al. (2000) Vulnerability of select neuronal types to Alzheimer’s disease. Ann. N. Y. Acad. Sci. 924, 53–61 5 Hardy, J. (1997) Amyloid, the presenilins and Alzheimer’s disease. Trends Neurosci. 20, 154–159 6 Poirier, J. et al. (1993) Apolipoprotein E polymorphism and Alzheimer’s disease. Lancet 342, 697–699 7 Corder, E.H. et al. (1994) Protective effect of apolipoprotein E type 2 allele for late onset Alzheimer disease. Nat. Genet. 7, 180–184 8 Carrasquillo, M.M. et al. (2009) Genetic variation in PCDH11X is associated with susceptibility to late-onset Alzheimer’s disease. Nat. Genet. 41, 192–198 9 Reiman, E.M. et al. (2007) GAB2 alleles modify Alzheimer’s risk in APOE epsilon4 carriers. Neuron. 54, 713–720 10 Li, H. et al. (2008) Candidate single-nucleotide polymorphisms from a genomewide association study of Alzheimer disease. Arch. Neurol. 65, 45–53 11 Harold, D. et al. (2009) Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer’s disease. Nat. Genet. 41, 1088–1093 12 Lambert, J.C. et al. (2009) Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer’s disease. Nat. Genet. 41, 1094–1099 13 Boyles, J.K. et al. (1990) Accumulation of apolipoproteins in the regenerating and remyelinating mammalian peripheral nerve. Identification of apolipoprotein D, apolipoprotein A-IV, apolipoprotein E, and apolipoprotein A-I. J. Biol. Chem. 265, 17805–17815 14 Poirier, J. et al. (1991) Cloning of hippocampal poly(A) RNA sequences that increase after entorhinal cortex lesion in adult rat. Brain Res. Mol. Brain Res. 9, 191–195 15 May, P.C. et al. (1990) Dynamics of gene expression for a hippocampal glycoprotein elevated in Alzheimer’s disease and in response to experimental lesions in rat. Neuron 5, 831–839 16 Poirier, J. et al. (1991) Astrocytic apolipoprotein E mRNA and GFAP mRNA in hippocampus after entorhinal cortex lesioning. Brain Res. Mol. Brain Res. 11, 97–106 17 Kalaria, R.N. et al. (2008) Alzheimer’s disease and vascular dementia in developing countries: prevalence, management, and risk factors. Lancet Neurol. 7, 812–826 18 Skoog, I. and Gustafson, D. (2006) Update on hypertension and Alzheimer’s disease. Neurol. Res. 28, 605–611 19 Thambisetty, M. et al. (2010) APOE epsilon4 genotype and longitudinal changes in cerebral blood flow in normal aging. Arch. Neurol. 67, 93–98 20 de Leon, M.J. et al. (2001) Prediction of cognitive decline in normal elderly subjects with 2-[(18)F]fluoro-2-deoxy-D-glucose/positronemission tomography (FDG/PET). Proc. Natl. Acad. Sci. U. S. A. 98, 10966–10971 21 Bondi, M.W. et al. (2005) fMRI evidence of compensatory mechanisms in older adults at genetic risk for Alzheimer disease. Neurology 64, 501–508 22 Corder, E.H. et al. (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261, 921–923 23 Forette, F. et al. (1998) Prevention of dementia in randomised doubleblind placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial. Lancet 352, 1347–1351 24 Hoffman, L.B. et al. (2009) Less Alzheimer disease neuropathology in medicated hypertensive than nonhypertensive persons. Neurology 72, 1720–1726 25 Peila, R. et al. (2001) Joint effect of the APOE gene and midlife systolic blood pressure on late-life cognitive impairment: the Honolulu–Asia aging study. Stroke 32, 2882–2889 26 Shah, K. et al. (2009) Does use of antihypertensive drugs affect the incidence or progression of dementia? A systematic review. Am. J. Geriatr. Pharmacother. 7, 250–261
476
Trends in Molecular Medicine Vol.16 No.10 27 Mulnard, R.A. et al. (2000) Estrogen replacement therapy for treatment of mild to moderate Alzheimer’s disease: a 1-year randomized controlled trial. JAMA 283, 1007–1015 28 Poirier, J. (2005) Apolipoprotein E, cholesterol transport and synthesis in sporadic Alzheimer’s disease. Neurobiol. Aging 26, 355–361 29 Hirsch-Reinshagen, V. et al. (2005) The absence of ABCA1 decreases soluble ApoE levels but does not diminish amyloid deposition in two murine models of Alzheimer disease. J. Biol. Chem. 280, 43243–43256 30 Blain, J.F. et al. (2004) A role for lipoprotein lipase during synaptic remodeling in the adult mouse brain. Neurobiol. Dis. 15, 510–519 31 Lutjohann, D. et al. (1996) Cholesterol homeostasis in human brain: evidence for an age-dependent flux of 24S-hydroxycholesterol from the brain into the circulation. Proc. Natl. Acad. Sci. U. S. A. 93, 9799– 9804 32 Leoni, V. (2009) Oxysterols as markers of neurological disease – a review. Scand. J. Clin. Lab Invest. 69, 22–25 33 Mulder, M. et al. (1998) Reduced levels of cholesterol, phospholipids, and fatty acids in cerebrospinal fluid of Alzheimer disease patients are not related to apolipoprotein E4. Alzheimer Dis. Assoc. Disord. 12, 198–203 34 Svennerholm, L. and Gottfries, C.G. (1994) Membrane lipids, selectively diminished in Alzheimer brains, suggest synapse loss as a primary event in early-onset form (type I) and demyelination in lateonset form (type II). J. Neurochem. 62, 1039–1047 35 Chochina, S.V. et al. (2001) Amyloid beta-peptide1-40 increases neuronal membrane fluidity: role of cholesterol and brain region. J. Lipid Res. 42, 1292–1297 36 Poirier, J. (2008) Apolipoprotein E represents a potent gene-based therapeutic target for the treatment of sporadic Alzheimer’s disease. Alzheimers Dement. 4, S91–97 37 Beffert, U. et al. (1998) The neurobiology of apolipoproteins and their receptors in the CNS and Alzheimer’s disease. Brain Res. Brain Res. Rev. 27, 119–142 38 Poirier, J. et al. (1993) Cholesterol synthesis and lipoprotein reuptake during synaptic remodelling in hippocampus in adult rats. Neuroscience 55, 81–90 39 Nathan, B.P. et al. (1994) Differential effects of apolipoproteins E3 and E4 on neuronal growth in vitro. Science 264, 850–852 40 Beffert, U. et al. (2004) Functions of lipoprotein receptors in neurons. J. Lipid Res. 45, 403–409 41 Posse De Chaves, E.I. et al. (2000) Uptake of lipoproteins for axonal growth of sympathetic neurons. J. Biol. Chem. 275, 19883–19890 42 Champagne, D. et al. (2002) Apolipoprotein E knockout mice display procedural deficits in the Morris water maze: analysis of learning strategies in three versions of the task. Neuroscience 114, 641–654 43 Mulder, M. et al. (2004) Low-density lipoprotein receptor-knockout mice display impaired spatial memory associated with a decreased synaptic density in the hippocampus. Neurobiol. Dis. 16, 212–219 44 Papassotiropoulos, A. et al. (2002) 24S-hydroxycholesterol in cerebrospinal fluid is elevated in early stages of dementia. J. Psychiatr. Res. 36, 27–32 45 Igbavboa, U. et al. (2002) A new role for apolipoprotein E: modulating transport of polyunsaturated phospholipid molecular species in synaptic plasma membranes. J. Neurochem. 80, 255–261 46 Klunk, W.E. et al. (1998) Metabolic alterations in postmortem Alzheimer’s disease brain are exaggerated by Apo-E4. Neurobiol. Aging 19, 511–515 47 Mason, R.P. et al. (1992) Evidence for changes in the Alzheimer’s disease brain cortical membrane structure mediated by cholesterol. Neurobiol. Aging 13, 413–419 48 Mauch, D.H. et al. (2001) CNS synaptogenesis promoted by gliaderived cholesterol. Science 294, 1354–1357 49 Dietschy, J.M. and Turley, S.D. (2004) Thematic review series: brain lipids. Cholesterol metabolism in the central nervous system during early development and in the mature animal. J. Lipid Res. 45, 1375– 1397 50 Utermann, G. et al. (1979) Polymorphism of apolipoprotein E. III. Effect of a single polymorphic gene locus on plasma lipid levels in man. Clin. Genet. 15, 63–72 51 Arendt, T. et al. (1997) Plastic neuronal remodeling is impaired in patients with Alzheimer’s disease carrying apolipoprotein epsilon 4 allele. J. Neurosci. 17, 516–529
Review 52 Sparks, D.L. et al. (1994) Induction of Alzheimer-like beta-amyloid immunoreactivity in the brains of rabbits with dietary cholesterol. Exp. Neurol. 126, 88–94 53 Refolo, L.M. et al. (2000) Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model. Neurobiol. Dis. 7, 321–331 54 Ghribi, O. et al. (2006) Deposition of iron and beta-amyloid plaques is associated with cortical cellular damage in rabbits fed with long-term cholesterol-enriched diets. J. Neurochem. 99, 438–449 55 Poirier, J. (2003) Apolipoprotein E and cholesterol metabolism in the pathogenesis and treatment of Alzheimer’s disease. Trends Mol. Med. 9, 94–101 56 Champagne, D. et al. (2003) The cholesterol-lowering drug probucol increases apolipoprotein E production in the hippocampus of aged rats: implications for Alzheimer’s disease. Neuroscience 121, 99–110 57 Fassbender, K. et al. (2001) Simvastatin strongly reduces levels of Alzheimer’s disease beta-amyloid peptides Abeta 42 and Abeta 40 in vitro and in vivo. Proc. Natl. Acad. Sci. U. S. A. 98, 5856–5861 58 Refolo, L.M. et al. (2001) A cholesterol-lowering drug reduces betaamyloid pathology in a transgenic mouse model of Alzheimer’s disease. Neurobiol. Dis. 8, 890–899 59 Bodovitz, S. and Klein, W.L. (1996) Cholesterol modulates alphasecretase cleavage of amyloid precursor protein. J. Biol. Chem. 271, 4436–4440 60 Giubilei, F. et al. (1990) Serum lipoprotein pattern variations in dementia and ischemic stroke. Acta Neurol. Scand. 81, 84–86 61 Moroney, J.T. et al. (1999) Low-density lipoprotein cholesterol and the risk of dementia with stroke. JAMA 282, 254–260 62 Dehouck, B. et al. (1997) A new function for the LDL receptor: transcytosis of LDL across the blood–brain barrier. J. Cell Biol. 138, 877–889 63 Waelsch, H. et al. (1940) A study of the synthesis and deposition of lipids in brain and other tissues with deuterium as an indicator. J. Biol. Chem. 135, 291–296 64 Chobanian, A.V. and Hollander, W. (1962) Body cholesterol metabolism in man. I. The equilibration of serum and tissue cholesterol. J. Clin. Invest. 41, 1732–1737 65 Roheim, P.S. et al. (1979) Apolipoproteins in human cerebrospinal fluid. Proc. Natl. Acad. Sci. U. S. A. 76, 4646–4649 66 Salen, G. et al. (1987) Increased concentrations of cholestanol and apolipoprotein B in the cerebrospinal fluid of patients with cerebrotendinous xanthomatosis. Effect of chenodeoxycholic acid. N. Engl. J. Med. 316, 1233–1238 67 Igbavboa, U. et al. (1997) Transbilayer distribution of cholesterol is modified in brain synaptic plasma membranes of knockout mice deficient in the low-density lipoprotein receptor, apolipoprotein E, or both proteins. J. Neurochem. 69, 1661–1667 68 Gibson Wood, W. et al. (2003) Amyloid beta-protein interactions with membranes and cholesterol: causes or casualties of Alzheimer’s disease. Biochim. Biophys. Acta 1610, 281–290 69 Riddell, D.R. et al. (2001) Compartmentalization of beta-secretase (Asp2) into low-buoyant density, noncaveolar lipid rafts. Curr. Biol. 11, 1288–1293 70 Kojro, E. et al. (2001) Low cholesterol stimulates the nonamyloidogenic pathway by its effect on the alpha-secretase ADAM 10. Proc. Natl. Acad. Sci. U. S. A. 98, 5815–5820 71 Xiu, J. et al. (2006) Influence of cholesterol and lovastatin on alphaform of secreted amyloid precursor protein and expression of alpha7 nicotinic receptor on astrocytes. Neurochem. Int. 49, 459–465 72 Abad-Rodriguez, J. et al. (2004) Neuronal membrane cholesterol loss enhances amyloid peptide generation. J. Cell Biol. 167, 953–960 73 Puglielli, L. et al. (2001) Acyl-coenzyme A: cholesterol acyltransferase modulates the generation of the amyloid beta-peptide. Nat. Cell Biol. 3, 905–912 74 Pani, A. et al. (2009) Altered cholesterol ester cycle in skin fibroblasts from patients with Alzheimer’s disease. J. Alzheimers Dis. 18, 829–841 75 Beffert, U. et al. (1999) Apolipoprotein E and beta-amyloid levels in the hippocampus and frontal cortex of Alzheimer’s disease subjects are disease-related and apolipoprotein E genotype dependent. Brain Res. 843, 87–94
Trends in Molecular Medicine
Vol.16 No.10
76 Riddell, D.R. et al. (2008) Impact of apolipoprotein E (ApoE) polymorphism on brain ApoE levels. J. Neurosci. 28, 11445–11453 77 Bales, K.R. et al. (2009) Human APOE isoform-dependent effects on brain beta-amyloid levels in PDAPP transgenic mice. J. Neurosci. 29, 6771–6779 78 Rebeck, G.W. et al. (1993) Apolipoprotein E in sporadic Alzheimer’s disease: allelic variation and receptor interactions. Neuron 11, 575– 580 79 Holtzman, D.M. et al. (1999) Expression of human apolipoprotein E reduces amyloid-beta deposition in a mouse model of Alzheimer’s disease. J. Clin. Invest. 103, R15–21 80 Beffert, U. et al. (1999) Apolipoprotein E isoform-specific reduction of extracellular amyloid in neuronal cultures. Brain Res. Mol. Brain Res. 68, 181–185 81 Yang, D.S. et al. (1999) Apolipoprotein E promotes the binding and uptake of beta-amyloid into Chinese hamster ovary cells in an isoform-specific manner. Neuroscience 90, 1217–1226 82 Deane, R. et al. (2008) apoE isoform-specific disruption of amyloid beta peptide clearance from mouse brain. J. Clin. Invest. 118, 4002– 4013 83 Evans, K.C. et al. (1995) Apolipoprotein E is a kinetic but not a thermodynamic inhibitor of amyloid formation: implications for the pathogenesis and treatment of Alzheimer disease. Proc. Natl. Acad. Sci. U. S. A. 92, 763–767 84 Holtzman, D.M. et al. (2000) Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U. S. A. 97, 2892–2897 85 Morris, J.A. et al. (1999) The genomic organization and polymorphism analysis of the human Niemann-Pick C1 gene. Biochem. Biophys. Res. Commun. 261, 493–498 86 Ohm, T.G. et al. (2003) Cholesterol and tau protein – findings in Alzheimer’s and Niemann Pick C’s disease. Pharmacopsychiatry 36 (Suppl. 2), S120–126 87 Distl, R. et al. (2001) Tangle-bearing neurons contain more free cholesterol than adjacent tangle-free neurons. Acta Neuropathol. 101, 547–554 88 Chauhan, N.B. (2006) Effect of aged garlic extract on APP processing and tau phosphorylation in Alzheimer’s transgenic model Tg2576. J. Ethnopharmacol. 108, 385–394 89 Jick, H. et al. (2000) Statins and the risk of dementia. Lancet 356, 1627–1631 90 Wolozin, B. et al. (2007) Simvastatin is associated with a reduced incidence of dementia and Parkinson’s disease. BMC Med. 5, 20 91 Haag, M.D. et al. (2009) Statins are associated with a reduced risk of Alzheimer disease regardless of lipophilicity. The Rotterdam Study. J. Neurol. Neurosurg. Psychiatry 80, 13–17 92 Li, G. et al. (2007) Statin therapy is associated with reduced neuropathologic changes of Alzheimer disease. Neurology 69, 878–885 93 Goldstein, J.L. and Brown, M.S. (1990) Regulation of the mevalonate pathway. Nature 343, 425–430 94 Enjalbert, M. et al. (1980) Usefulness of probucol in treating primary hypercholesterolemia. Can. Med. Assoc. J. 123, 754–757 95 Liang, Y. et al. (2004) A liver X receptor and retinoid X receptor heterodimer mediates apolipoprotein E expression, secretion and cholesterol homeostasis in astrocytes. J. Neurochem. 88, 623–634 96 Riddell, D.R. et al. (2007) The LXR agonist TO901317 selectively lowers hippocampal Abeta42 and improves memory in the Tg2576 mouse model of Alzheimer’s disease. Mol. Cell. Neurosci. 34, 621–628 97 Feldman, H.H. et al. (2010) Randomized controlled trial of atorvastatin in mild to moderate Alzheimer disease: LEADe. Neurology 74, 956–964 98 Arvanitakis, Z. and Knopman, D.S. (2010) Clinical trial efforts in Alzheimer disease. Why test statins? Neurology 74, 945–946 99 Poirier, J. (2002) Cholesterol transport and synthesis are compromised in the brain in sporadic Alzheimer’s disease: from risk factors to therapeutic targets. In Alzheimer’s Disease and Related Disorders Annual (Gauthier, S. and Cummings, J.L., eds), pp. 1–23, Dunitz 100 Thompson, P.M. et al. (2003) Dynamics of gray matter loss in Alzheimer’s disease. J. Neurosci. 23, 994–1005
477