Cholesterol and Alzheimer's disease: A still poorly understood correlation
A large amount of evidence suggests a pathogenic link between cholesterol homeostasis dysregulation and Alzheimer's disease (AD). In cell culture systems, the production of amyloid-β (Aβ) is modulated by cholesterol, and studies on animal models have consistently demonstrated that hypercholesterolemia is associated with an increased deposition of cerebral Aβ peptides. Consequently, a number of epidemiological studies have examined the effects of cholesterol-lowering drugs (i.e., statins) in the prevention and the treatment of AD. However, while retrospective studies suggested a potential benefit of statin therapy, clinical trials produced inconsistent results. Here, we summarize the main findings from in vitro and in vivo research where the correlation between cholesterol and the neurodegenerative disorder was investigated. Recognition of this correlation could be an important step forward for our understanding of AD pathogenesis and, possibly, for the development of new therapeutic strategies. © 2012 IUBMB IUBMB Life, 64(12): 931–935, 2012
Amyloid-β (Aβ) is a 39–42 amino acid peptide that takes two prevalent forms in humans, Aβ40 and Aβ42, although longer and shorter peptides also exist. Aβ is the natural product of the cleavage of a much larger protein, the amyloid-β precursor protein (APP) (1), by proteases called β- and γ-secretases (2–4). Much attention has been focused on this peptide because mutations in the APP and γ-secretase genes are associated with early onset familial forms of Alzheimer's disease (AD) and are known to cause cerebral Aβ accumulation. However, while these genetic mutations are responsible for the accumulation of Aβ in familial AD, the causative factors for Aβ load in sporadic forms of AD are still not known. This leads to the possibility that, in the absence of genetic mutations, the identification of risk factors and mechanisms by which these factors contribute to the accumulation of Aβ may help in preventing the onset of this devastating disorder. Hypercholesterolemia is such a factor and has been shown by epidemiological and laboratory studies to increase the production of Aβ. However, the molecular events by which cholesterol causes Aβ accumulation and the real contribution of this steroid to the pathogenesis of AD are still poorly understood.
In this review, we summarize the main findings from cell culture, animal model, observational, and clinical studies where the correlation between cholesterol and the neurodegenerative disorder was investigated.
CELLULAR CHOLESTEROL AND AMYLOIDOGENESIS
Because of the low permeability of the blood-brain barrier to peripheral cholesterol, most of the cholesterol present in the brain derives from de novo synthesis in the central nervous system (5). Mature neurons are thought to reduce the cholesterol production and rely on glial cells, mainly astrocytes, for their cholesterol supply (6–8). The astrocytic compartment meets neuronal cholesterol demands by secreting cholesterol–apolipoprotein E (apoE) complexes, a mechanism that may involve the activation of the ATP-binding cassette transporter A1 (ABCA1) (9). Moreover, increased levels of ABCA1 were found to lower Aβ production in neuronal cultured cells (10), whereas the deletion of ABCA1 gene in AD mouse models was associated with greater Aβ deposits (11, 12), supporting the idea that cellular cholesterol efflux might influence Aβ production. Conversely, Aβ peptides have been recently found to exert an inhibitory effect on both cholesterol synthesis (13) and astrocytic ABCA1 expression (14), therefore suggesting the existence of a regulatory cycle connecting APP processing and cholesterol homeostasis.
The amyloidogenic processing of APP is believed to occur at, or in close proximity to, lipid rafts, cholesterol-rich membrane microdomains where both β- and γ-secretase complexes reside (15). Because the enzymatic activity of β- and γ-secretases is cholesterol-dependent (16, 17), it is likely that the production of Aβ reflects a biological response to the increased cholesterol availability.
Excess free cholesterol in the cell is converted into cholesteryl esters by the enzyme acyl-coenzymeA: cholesterol acyltransferase (ACAT). Using genetic, biochemical, and metabolic approaches, it was found that cholesteryl esters are directly correlated with Aβ production, as increasing levels of cholesteryl esters enhanced Aβ release in cultured cells, whereas pharmacological inhibition of ACAT led to the reduction of both cholesteryl esters and Aβ production (18, 19). Accordingly, the inhibition of the 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase, a key enzyme in the cholesterol de novo synthesis, reduced both intracellular and extracellular Aβ levels (20, 21).
Despite the multiple lines of evidence, over the last two decades, indicating a role of cholesterol in the production of Aβ, the molecular mechanisms involved have so far only partially been uncovered. Of particular interest is the very recent observation that cholesterol forms an avid complex with the Aβ domain of APP (residues 672–711), which has provided an insight into the amyloidogenic process, implying that strategies aimed at preventing the binding of cholesterol to APP may have a prophylactic utility in AD (22).
In addition to being involved in the amyloidogenic processing of APP, cholesterol and its oxidation products were found to amplify the cytotoxic effects of Aβ peptides in neuronal cultured cells (23–25). Indeed, the idea that oxysterols, a class of cholesterol oxidation derivatives, might be the link between altered cholesterol metabolism and AD has been widely supported, and several excellent reviews on this topic are available (26–28).
Experimental studies performed on different animal models have consistently demonstrated that hypercholesterolemia is associated with an increased deposition of cerebral Aβ peptides. Sparks et al. in 1994 (29) were the first to report that, compared with control animals, rabbits fed a cholesterol-rich diet displayed a higher Aβ immunoreactivity in hyppocampal neurons. A few years later, hypercholesterolemia was found to accelerate the Alzheimer's amyloid pathology in an APP/PS1 double-transgenic mouse model (30). Similarly, mice overexpressing the human APP carrying the K670N/M671L Swedish mutation, and fed a high-cholesterol diet, were found to enhance cerebral accumulation of Aβ (31). Due to the fact that also APP transgenic mice lacking the apoE developed hypercholesterolemia, but in this case associated with a dramatic reduction of cerebral amyloidogenesis (32), it was suggested that the effects of cholesterol on the processing of APP may require the presence of apoE. Nevertheless, this tempting hypothesis was contradicted 1 year later by experiments in mice genetically modified to express endogenous levels of APP and abundant human Aβ. In these animals, in fact, the diet-induced hypercholesterolemia lowered the brain content of APP derivatives, including Aβ40 and Aβ42, despite the high levels of apoE in the serum and brain (33).
ApoE is the main cholesterol-carrier protein in the brain and is particularly involved in the transport of cholesterol from astrocytes to neurons. The influence of apoE on in vivo Aβ deposition was originally suggested by Strittmatter et al. in 1993 (34, 35) and subsequently confirmed by experimental evidence in several animal studies (32, 36, 37). Although many molecular events are still to be clarified, these findings, together with the notion that allelic variation in the apoE gene is the most influential genetic risk factor for sporadic AD, have strongly supported the existence of an intrinsic relationship between cholesterol homeostasis, Aβ production and clearance, and apoE metabolism (38).
Over the course of several years, also a number of cognitive studies, evaluating the influence of dietary cholesterol on memory, have been performed on different animal models. Behavioral assessments, for example, demonstrated that AD mice fed a high fat/high cholesterol diet had more spatial learning impairment than those fed a normal diet (39). In the same study, it was also shown that this fat-rich diet promoted atherosclerosis in the transgenic AD animals, thus providing an additional biological mechanism for the cholesterol-induced cognitive impairment (39).
In a series of elegant studies, Greenwood and Winocur (40–42) reported that rats fed a nutritionally adequate diet, but with a high cholesterol content, were impaired in a wide range of learning and memory functions. More recently, this cholesterol-induced memory impairment has been correlated with a loss of dendritic integrity, cholinergic dysfunction, inflammation (43), enhanced cortical Aβ and phosphorylated tau (44), all indications which resemble an AD-like pathology. In line with such observations, cholesterol-lowering drugs (i.e., statins) were found to reduce cerebral Aβ (21, 45, 46), phosphorylated tau (46), inflammation (45), and memory deficits (46, 47) in a variety of animal models. Whether statins act in the brain simply by lowering cholesterol biosynthesis or by a different mechanism, however, still remains an open question.
Although a number of epidemiological studies have shown that hypercholesterolemia is related to an increased risk of developing AD or mild cognitive impairment (48–50), other results, which do not fit in with this conclusion, have also been reported. In 2005, for example, Reitz et al. (51) observed no effects of cholesterol and triglyceride levels on the cognitive ability of aged individuals, while other studies demonstrated that high levels of total cholesterol or low-density lipoprotein (LDL)-cholesterol correlated with a decreased AD (52) and dementia risk (53). Such discrepancies may be explained by differences in the assessment of cholesterol analyses, patient's characteristics, as well as follow-up times. Several studies, in fact, considered total cholesterol levels without evaluating the impact of high-density lipoprotein (HDL)- and LDL-cholesterol. It should also be noted that studies finding a negative correlation between cholesterolemia and dementia were principally carried out late in the patient's lives, whereas studies reporting a positive correlation were conducted on younger individuals. Furthermore, it needs to be taken into account that when the enrolled participants are already affected by AD, it might be difficult to establish if a change in cholesterol levels has an effect on the progression of the disease or, vice versa, the AD-related changes affect the patient's cholesterol metabolism.
Statin Therapy and AD
Together with the data obtained from animal research, some observational studies have bolstered the hypothesis that statins may prevent or mitigate the course of AD.
Statins are competitive inhibitors of HMG-CoA reductase, the enzyme that catalyzes the rate-limiting step in the cholesterol biosynthesis. Therefore, the direct effect of statin therapy is the reduction of circulating cholesterol and also the inhibition of de novo cholesterol biosynthesis in the brain.
According to a 2000 report published in The Lancet, individuals of 50 years or older who were prescribed statins had a substantially lower risk of developing dementia, regardless of the presence or absence of hyperlipidaemia. The available data, however, did not distinguish between AD and other forms of dementia (54). Similarly, patients older than 60 years taking lovastatin or pravastatin showed a lower AD risk, compared to the general population or patients taking other medications (55). Nevertheless, although it is likely that statins affect amyloidogenesis by reducing the cerebral levels of cholesterol, the mechanisms underlying the therapeutic benefits of statins in the brain remain unclear. For example, both lovastatin and pravastatin were found to reduce the prevalence of AD, despite lovastatin having an almost 50-fold higher ability to cross the blood-brain barrier compared to pravastatin. Interestingly, the fact that statins exert a lipophilicity-independent protection against AD has also been recently observed in the prospective, population-based Rotterdam Study (56). In addition, an observational study on postmenopausal women with coronary heart disease reported a trend for a lower likelihood of cognitive impairment in statin users, which was independent of plasma lipid levels (57).
Not all observational studies, however, lead to the conclusion that statins protect against AD. In fact, the Cache County Study, which began in 1995 with the enrollment of 5,092 elderly residents, has found no correlation between the use of statins and the onset of dementia or AD (58). Likewise, neither in the Cardiovascular Health Study (59) nor in the Religious Orders Study (60) was statin therapy associated with a decreased risk of dementia, although people taking statins appeared less likely to develop amyloid plaques (60).
Regarding the results of intervention studies, two large, randomized, controlled trials on pravastatin (PROSPER study) (61) and simvastatin (MCR/BHF Heart Protection Study) (62) have not confirmed a clinically demonstrable cognitive benefit of statins in the treatment of AD. Moreover, the data published in 2005 by Sparks et al. (63), showing that treatment with atorvastatin had a beneficial effect on cognition and behavior in patients with mild to moderate AD, have not been confirmed by the recent, randomized controlled, LEADe study (64).
Prompted largely by results of cellular and animal studies, the concept of altered cholesterol homeostasis has emerged as an important factor in the pathogenesis of AD. However, despite the well-established influence of cholesterol on the amyloidogenic processing of APP, further investigation is needed to better understand the molecular details and the causal link, if any, between cholesterol homeostasis dysregulation and neurodegeneration. Importantly, in epidemiological studies, methodological issues, including criteria for subject selection, drug dosage, duration of treatment, and methods of cognitive evaluation, should be more thoroughly considered to avoid a lack of relevant outcomes and inconsistent results. On this point, Wolozin and colleagues (65) have recently concluded that the complexity of late-onset AD pharmacological treatment might result from the cumulative effects of at least four different pathophysiological factors: Aβ accumulation, cardiovascular disease, age-associated loss of synaptic plasticity, and inflammation. Is it by pure chance that at least two of these factors are influenced by cholesterol?
This work was supported by grants from the Alzheimer's Association and PRIN (2009M8FKBB_002 and 2008N9N9KL_002).