The hydroxy-methyl-glutaryl CoA reductase promoter polymorphism is associated with Alzheimer's risk and cognitive deterioration

The hydroxy-methyl-glutaryl CoA reductase promoter polymorphism is associated with Alzheimer's risk and cognitive deterioration

Neuroscience Letters 416 (2007) 66–70 The hydroxy-methyl-glutaryl CoA reductase promoter polymorphism is associated with Alzheimer’s risk and cogniti...

119KB Sizes 2 Downloads 6 Views

Neuroscience Letters 416 (2007) 66–70

The hydroxy-methyl-glutaryl CoA reductase promoter polymorphism is associated with Alzheimer’s risk and cognitive deterioration Elisa Porcellini a , Elena Calabrese b , Franca Guerini b , Marzia Govoni a , Martina Chiappelli a , Emanuela Tumini a , Kevin Morgan c , Sally Chappell c , Noor Kalsheker c , Massimo Franceschi d , Federico Licastro a,∗ a b

Department of Experimental Pathology, School of Medicine University of Bologna, Via San Giacomo 14, 40126 Bologna, Italy Department of Neurology, Laboratory of Molecular Medicine and Biotechnologies IRCCS Don Gnocchi Foundation, Milan, Italy c Institute of Genetics, University of Nottingham, Nottingham, United Kingdom d Neurology Department, Multi-Medica Santa Maria, Castellanza, Varese, Milan, Italy Received 30 October 2006; received in revised form 11 January 2007; accepted 22 January 2007

Abstract A link between cholesterol and Alzheimer’s disease (AD) had been suggested. Hydroxy-methylglutaryl-coenzyme A reductase (HMGCR) is the rate limiting enzyme in the synthesis of cholesterol. A single nucleotide polymorphism (SNP) in the promoter of this gene, never described in Italian AD population, was investigated in case-control studies. Genotype distribution and allele frequency in two groups of AD patients and non demented controls were investigated. A cohort of AD patients were also followed up for 2 years, cognitive performances recorded and a possible influence of this SNP on the disease progression was tested. The CC genotype of the HMGCR gene was associated with a reduced risk of AD. Conversely the A allele of this polymorphism was over represented in AD patients. The presence of the A allele was also associated with an accelerated cognitive deterioration in AD patients followed up for 2 years. However, transfection experiments showed that this polymorphism did not directly influence functional activity in luciferase reporter gene assays. This polymorphism of the HMGCR gene appears to be linked to both AD risk and disease progression. Present findings reinforce the notion that abnormal regulation of cholesterol metabolism is a key factor in the pathogenesis of the disease. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Alzheimer’s disease; Cholesterol; HMGCR; Polymorphism; AD risk; Cognitive decline; Transfection

Epidemiological studies suggested a link between cholesterol serum levels and AD [16,19]. Apolipoprotein E (APOE) is a cholesterol transporter in the blood and brain [15] and the ␧4 allele of the APOE gene is a well known genetic risk factor for both familial and sporadic forms of AD [6]. An interaction between serum cholesterol and the APOE ␧4 allele in AD has also been found [19]. Elevated blood cholesterol levels increased the susceptibility to AD and patients with the disease showed higher serum levels of total and low density lipoprotein (LDL) cholesterol than non demented elderly [19]. In vitro studies have shown that cholesterol affects amyloid precursor protein processing and generation of the amyloidogenic A␤ peptides. A direct correlation between cholesterol and A␤ peptide secretion



Corresponding author. Tel.: +39 051 2094730; fax: +39 051 2094746. E-mail address: [email protected] (F. Licastro).

0304-3940/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2007.01.046

has been suggested since cholesterol depletion decreases A␤ secretion and cholesterol overload increases it [1,9,14,25,29]. A non secondary role of cholesterol and its derivates in AD has been also suggested by gene association studies. In fact genetic variation in the cholesterol 24-hydroxylase gene (CYP46A1), which codes for an enzyme regulating cholesterol turnover between the brain, CSF and blood, has been reported to be positively associated with the risk of AD, to influence brain amyloid load and levels of phosphorylated tau [23] and to affect CSF A␤ peptide levels [17]. Another study, however, did not confirmed an association of CY46A1 polymorphism with AD risk [7]. A polymorphism of the acyl-coenzyme A:cholesterol acyltransferase gene was also found associated with a decreased risk of developing AD [30]. Finally, polymorphic variants coding for the low density lipoprotein related-protein receptor (LRP) [18] were also reported associated with AD risk. Another gene

E. Porcellini et al. / Neuroscience Letters 416 (2007) 66–70

67

APOE ␧ genotypes were detected as previously described [20]. Cloning fragments were amplified from templates corresponding to homozygotes for each allele and inserted into the pGL3E vector (Promega, Southampton, UK). A 1001-bp fragment of the HMGCR 5 flanking region (−995 to +6) was amplified by PCR using the following primers:

regulating cholesterol metabolism, the ATP-binding cassette transporter A1 (ABCA1) [31] was found associated with the increased AD risk. HMGCR is the rate limiting enzyme in cholesterol synthesis, controls the bio-availability of cellular cholesterol and is also the biology target of statins, a drug family lowering serum cholesterol [5]. It is of interest that epidemiological investigations reported a negative association of statin use with the AD risk [26], implying a potential role of HMGCR in the development of the disease. No investigations on a possible association of HMGCR gene variations with AD are on record. Here we show that a SNP in the promoter of HMGCR gene is associated with the risk of AD and the rate of cognitive decline after the manifestation of the disease. 190 patients with clinical diagnosis of probable AD (mean age = 74 ± 9 years) and 586 non demented elderly (mean age = 73 ± 7 years) were genotyped for HMGCR −911 promoter polimorphism (Table 1). A second AD population of 97 patients from the same Italian geographic area was also investigated. Two different AD groups were investigated to confirm the frequency of the polymorphism in Italian AD population. Diagnosis of probable AD was performed according to the National Institute of Neurological Communicative Disorders, as previously described [21]. A large group of AD patients were monitored for cognitive deterioration. Cognitive performances were assessed by Mini Mental State Examination scores during a 2 year follow-up in 296 AD patients who were divided in three groups according to the rate of cognitive deterioration (fast, intermediate or slow), following the method described elsewhere [8]. Genomic DNA was purified from peripheral blood leukocytes as previously described [12]; identification of a SNP in the promoter at position −911 (RS3761740), with respect to the transcriptional start site, of HMGCR gene was detected by DNA sequencing with the aid of an automated DNA sequencing machine (CEQ 2000, Beckman). A SNP was identified and consisted of a transversion C → A. Patients and controls were genotyped by PCR DNA amplification using primer pairs:5 TTGAGTGAAATACAGATCCTGTCC 3 /5 CCATAGACCTGCATCAGCGT 3 with 5 min at 96 ◦ C for the initial denaturation and 30 cycles of 30 s at 96 ◦ C, 30 s at 54 ◦ C and 45 s at 72 ◦ C. Then, 5 min incubation at 72 ◦ C for final extension was performed. The restriction enzyme BsuRI (MBI Fermentas, Italy; 10 U/sample) resolved three different band patterns, according to the three different genotypes; e.g. at 261 and 139 bp (AA genotype), 261, 70 and 69 bp (CC genotype) or 261, 139, 70 and 69 bp (AC genotype).

Forward, 5 -ggggacaagtttgtacaaaaaagcaggcttatgactgaagaagggccagtt Reverse, 5 -ggggaccactttgtacaagaaagctgggtcgaaggagccctcaccttacg Underlined nucleotides indicate the sequences required for the Gateway BP reaction (Gateway Cloning Technology, Invitrogen, UK). Amplification was conducted with 5 min at 95 ◦ C for the initial denaturation and 33 cycles of 30 s at 95 ◦ C, 1 min at 61.1 ◦ C and 1 min at 72 ◦ C, followed by 5 min at 72 ◦ C. The PCR product was inserted into the donor vector pDONOR221 by a BP recombination reaction to create an entry clone. Then, the HMGCR promoter SNP was transferred into the destination vector pGL3 Enhancer, modified for use in Gateway cloning, via a LR recombination reaction. This generated two reporter gene vectors regulated by the HMGCR promoter, each differing by a single base at the polymorphic site, as verified by sequencing. These vectors were referred to as pGL3C and pGL3A (according to the base present at the polymorphic site). Three human cell lines were used in this study: the hepatoma cell line Hep G2 (ATCC), the astrocytic cell line U373MG (ECACC) and the mixed glial/neuronal cell line T98G (CAMR). Hep G2 and U373 MG cells were grown in Eagle’s minimum essential medium (Sigma). T98G cells were grown in Dulbecco’s modified Eagle’s medium (DMEM; Gibco BRL). All cells were maintained at 37 ◦ C in a humidified atmosphere of 5% CO2 in air. Transfections were carried out in triplicate with 3 wells of a 12-well plate containing identical transfection reactions. Typically, plates were set up with 3 wells each containing pGL3Con (control, SV40 promoter and enhancer), pGL3Enh (SV40 enhancer alone), pGL3C (HMGCR promoter SNP wild type C allele) and pGL3A (HMGCR promoter SNP mutant A allele). All were co-transfected with pRL as a transfection efficiency control. All plasmids used in transfection experiments were endotoxin free preparations (Qiagen). Transfection experiments were described elsewhere [22]. Luciferase assay reagent II (LARII) and Stop and Glo reagent (Promega) were prepared according to the manufacturer’s instructions. Dual-luciferase reporter assays were carried

Table 1 Genotype distribution and allele frequency of −911 polymorphism in the HMGCR gene from AD patients and controls (CTR) CC

Total AD (N = 287) CTR (N = 586)

CA

AA

N

%

N

%

N

%

200 474

69.7 80.9

78 102

27.2 17.4

9 10

3.1 1.7

Freq. C allele

Freq. A allele

0.83 0.89

0.16 0.11

(2 13.861 p = 0.001. A Allele (2 13.734 p = 0.0001 OR 1.841 (1.330–2.549) p = 0.0001. CC Genotype (2 13.734 p = 0.0001 OR 0.543 (0.392–0.752) p = 0.0001.

68

E. Porcellini et al. / Neuroscience Letters 416 (2007) 66–70

out on a Turner Designs Model TD-20/20 Luminometer allowing sequential readings of firefly and Renilla luciferase reporter activities. The HMGCR SNP was in Hardy-Weinberg equilibrium in the control samples. Genotype distribution and allele frequency were compared by contingency tables and χ2 analysis. Odds ratio (OR) and confidence intervals (c.i.) for AD were also calculated and statistical significance assessed by using SPSS 11.0 software package. A SNP in the promoter region of the HMGCR was investigated in the two groups of AD patients. No difference in HMGCR allele and genotype frequencies between the two populations of AD patients was observed (p = 0.118); therefore, data from these two groups of AD patients were pooled. Results of HMGCR genotype distribution and allele frequency in AD patient and controls are reported in Table 1. The CC genotype was less frequent in AD than in controls (p = 0.0001) and the OR for the disease in CC homozygotes was 0.543 (c.i. = 0.392–0.752; p = 0.0001). Conversely, the A allele frequency was increased in AD patients (p = 0.0001) and the OR for AD was 1.841 in these subjects (c.i. = 1.330–2.549; p = 0.0001). As expected, the APOE ␧4 allele was over represented in AD patients (χ2 = 62.586 p = 0.0001 OR = 3.749 p = 0.0001). However, HMGCR A allele was over represented (OR = 2.218, p = 0.002) and the CC genotype less frequent (OR = 0.451, p = 0.0001) only in AD without the APOE ␧4 allele (data not shown). A large group of AD patients were followed up for two years and cognitive performances evaluated by MMSE score determination once a year. Patients losing more than 5 points in the two year interval were categorized with a fast rate of cognitive decline (F), those losing 2.6–4.9 points showed an intermediate

rate of cognitive decline (I) and those losing 2.5 or less points were labeled having a slow decline (S). AD patients were then stratified for the HMGCR genotypes and results are reported in Table 2. AD patients carrying the A allele were over represented in the group of AD with fast cognitive decline (60.6%) in comparison with those with a slow rate of cognitive deterioration (27.6%, p = 0.0001). Patients with slow and intermediate rate of cognitive decline were pooled, since no statistically significant difference in the allele frequency between these two group were detected. The presence of the A allele of HMGCR resulted in a OR of 3.626 (c.i. = 1.974–6.659, p = 0.0001) for an accelerated cognitive decline. On the other hand the CC genotype was reduced in AD with fast rate of cognitive deterioration (39.4%) compared with those with a slow rate (72.4%) (p = 0.0001). The presence of the APOE ␧4 allele did not independently affect the rate of cognitive decline in AD patients (data not shown) and the concomitant presence or absence of the APOE ␧4 allele did not change the effect of the HMGCR promoter SNP on cognitive decline (Table 2, panels B and C). To assess the effect of this SNP upon the expression level of the HMGCR gene, three human cell lines were transfected with constructs of the 5 flanking region in a luciferase reporter vector, differing only for the polymorphic nucleotide. This region clearly demonstrates functional promoter activity. However, no difference in reporter gene expression between the wild type and mutant allele was detected in any of the cell types (data not shown). Levels of reporter gene activity at different time points from astrocyte cultures (8, 16 and 48 h) were also evaluated, but again no difference was observed. Elevated levels of blood cholesterol have been associated with an increased risk of AD [27]. Moreover, in vitro investigations showed that cholesterol influenced A␤ peptide generation [1,6,14,15,29] and decreased the secretion of amyloid precursor

Table 2 Genotype distribution and allele frequencies of −911 HMGCR polymorphism in AD patients stratified according the rate of cognitive decline (panel A) CC

CA

AA

Freq. of C allele

Freq. of A allele

N

%

N

%

N

%

Panel A – total Fast = F (N = 66) Intermediate = I (N = 114) Slow = S (N = 116)

26 71 84

39.4 62.3 72.4

38 38 29

57.6 33.3 25.0

2 5 3

3.0 4.4 2.6

0.68 0.79 0.85

0.32 0.21 0.15

Panel B – APOE 4 positive F (N = 24) I (N = 47) S (N = 45)

9 33 34

37.5 70.2 75.6

14 13 11

58.3 27.7 24.4

1 1 0

4.2 2.1 0

0.66 0.84 0.88

0.34 0.16 0.12

Panel C – APOE 4 negative F (N = 26) I (N = 67) S (N = 42)

10 40 29

38.5 59.7 69.0

15 24 12

57.7 35.8 28.6

1 3 7

3.8 4.5 2.4

0.67 0.78 0.83

0.33 0.22 0.17

Patients were divided according to the presence of the ApoE ␧4 allele (panel B and C). χ2 19.354 p = 0.0001. F vs. I + S A Allele χ2 18.419 p = 0.0001 OR 3.626 (1.974–6.659) p = 0.0001. F vs. I + S CC Genotype χ2 18.419 p = 0.0001 OR 0.276 (0.150–0.506) p = 0.0001. χ2 10.514 p = 0.0001. F vs. I + S A Allele χ2 10.514 p = 0.0001 OR 4.46 (1.73–11.50) p = 0.0001. F vs. I + S CC Genotype χ2 10.514 p = 0.0001 OR 0.224 (0.087–0.576) p = 0.001. F vs. I + S Allele A χ2 5.337 p = 0.0001 OR 2.76 (1.144–6.660) p = 0.0001. F vs. I + S CC Genotype χ2 5.373 p = 0.0001 OR 0.362 (0.150–0.874) p = 0.001. Panel B and C: 45 AD cases not included (APOE or HMGCR no genotyping).

E. Porcellini et al. / Neuroscience Letters 416 (2007) 66–70

protein [10]. Cholesterol, as well as ApoE protein, co-localizes in amyloid plaques from AD brains, suggesting that cholesterol and ApoE protein affect plaque formation [3]. Therefore, cholesterol appears to participate in several pathogenetic pathways of AD, as suggested in a recent review [2]. HMGCR is the rate limiting enzyme for cholesterol synthesis in the mevalonate pathway [11]. Mevalonate gives rise to isoprenoids compounds, such as farnesyl and geranyl groups, which are required for membrane attachment of several proteins [32]. Recently, it has been shown that atorvastatin, an HMGCR inhibitor, induced a profound reduction of neurite length, neurite loss and cell death when added in vitro to human neuronal cell lines [28]. This toxic mechanism appeared to be mediated by the depletion of geranylgeranylpyrophosphate cholesterol derivative [24]. Finally, another HMGCR inhibitor, lovastatin, enhanced A␤ peptide production and senile plaque deposition in female mice transgenic for human amyloid precursor protein (APP), an animal model for AD [24]. Genotype and allele frequency were overlapping in two independent populations of AD and therefore data from patients were pooled. Here we show that a SNP in the promoter region of the HMGCR gene is associated with the risk of developing AD in APOE ␧4 negative patients. After the clinical manifestation of the disease, the A allele of the HMGCR gene was also associated with an accelerated rate of cognitive decline. Conversely the CC genotype was associated with a decreased risk of AD and a less pronounced cognitive decline in the two years follow-up. APOE ␧4 allele was associated with an increased risk of AD in our study, however, the APOE ␧4 allele did not influence the rate of cognitive decline. It is important to note that the HMGCR allele effect on AD risk was independent from APOE ␧4 and the effect on cognitive decline was present in both APOE ␧4 negative and positive AD groups. To verify a possible effect of this SNP on gene transcription, transfection experiment were performed. Results showed that this SNP did not directly affect the expression of HMGCR gene under physiological culture conditions. The updated version of HapMap (release 18) shows that there are at least 12 other SNPs in close proximity (within 3 kb) of the SNP tested here. We studied another polymorphism at the position-655 in the promoter region (data not shown), but the frequency of the mutated allele of this SNP was too low in the Italian population (<1%) to be investigated in association with AD. The interpretation of our findings is complex. On the whole our findings suggest that the SNP at −911 in the HMGCR gene might be in linkage disequilibrium with another SNP in the same gene exerting the functional effect, as described in another report [4] or with other SNPs in different genes of the mevalonate metabolic pathway, which might affect farnesyl and/or geranyl groups attachment to APP or APP processing enzymes and by this way influence the A␤ production. Another interpretation derives from a recent report showing that A␤40 directly reduced cholesterol de novo synthesis by inhibiting HMGCR activity [13]. Different HMGCR genetic backgrounds might influence a differential sensitivity of the enzyme to the A␤40 inhibitory activity and this effect could result in an increased risk of the

69

disease and an accelerated cognitive decline of A carrier patients. Conversely, C carriers showed a reduced AD risk or rate of cognitive deterioration. Therefore, HMGCR appears to behave as a dual action gene; it may influence A␤ release itself and also modulate some biological effects mediated by A␤ release. This SNP in the HMGCR gene appears to be considered a disease modifier, since it is associated with both an increased risk of AD and an accelerated clinical progression of the disease. Finally the genetic link of HMGCR with the disease reinforces the notion that mevalonate and cholesterol metabolisms play a primary role in the pathogenesis of neuro-degenerative processes leading to cognitive deterioration and dementia. Acknowledgements Research supported by Italian Ministry for University and Technology (COFIN), Italian CURA and Italian Ministry of Health “Progetti Finalizzati Alzheimer”. NK and KM would like to thank the Alzheimer’s Research Trust for Network and Programme Grant support. References [1] S. Bodovitz, W.L. Klein, Cholesterol modulates alpha-secretase cleavage of amyloid precursor protein, J. Biol. Chem. 271 (1996) 4436–4440. [2] L. Burns, G. Konopka, E. Pack-Chung, L.A. Ingano, O. Berezovska, B.T. Hyman, T.Y. Chang, R.E. Tanzi, D.M. Kovacs, Acyl-coenzyme A: cholesterol acyltransferase modulates the generation of the amyloid beta-peptide, Nat. Cell. Biol. 3 (2001) 905–912. [3] M.P. Burns, W.J. Noble, V. Olm, Co-localization of cholesterol, apolipoprotein E and fibrillar Abeta in amyloid plaques, Brain Res. Mol. Brain. Res. 110 (2003) 119–125. [4] D.I. Chasman, D. Posada, L. Subrahmanyan, N.R. Cook, V.P. Stanton, P.M. Ridker Jr., Pharmacogenetic study of statin therapy and cholesterol reduction, JAMA 16 (2004) 2821–2827. [5] P.H. Chong, R. Kezele, C. Franklin, High-density lipoprotein cholesterol and the role of statins, Circ. J. 66 (2002) 1037–1044. [6] E.H. Corder, A.M. Saunders, W.J. Strittmatter, D.E. Schmechel, P.C. Gaskell, G.W. Small, Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families, Science 261 (1993) 921–923. [7] P. Desai, S.T. DeKosky, M.I. Kamboh, Genetic variation in the cholesterol 24-hydroxylase (CYP46) gene and the risk of Alzheimer’s disease, Neurosci. Lett. 2 (2002) 9–12. [8] R.S. Doody, P. Massman, J.K. Dunn, A method for estimating progression rates in Alzheimer disease, Arch. Neurol. 58 (2001) 449–454. [9] E.R. Frears, D.J. Stephens, C.E. Walters, H. Davies, B.M. Austen, The role of cholesterol in the biosynthesis of beta-amyloid, Neuroreport 10 (1999) 1699–1705. [10] J.L. Galbete, T.R. Martin, E. Peressini, P. Modena, R. Bianchi, G. Forloni, Cholesterol decreases secretion of the secreted form of amyloid precursor protein by interfering with glycosylation in the protein secretory pathway, Biochem. J. 2 (2000) 307–313. [11] J.L. Goldstein, M.S. Brown, Regulation of the mevalonate pathway, Nature 343 (2000) 425–430. [12] L.M. Grimaldi, V.M. Casadei, C. Ferri, F. Licastro, G. Annoni, I. Biunno, G. De Bellis, S. Sorbi, C. Mariani, N. Canal, W.S. Griffin, M. Franceschi, Association of early-onset Alzheimer’s disease with an interleukin-1alpha gene polymorphism, Ann. Neurol. 47 361–365. [13] M.O. Grimm, H.S. Grimm, A.J. Patzold, E.G. Zinser, R. Halonen, M. Duering, J.A. Tschape, B. De Strooper, U. Muller, J. Shen, T. Hartmann, Regulation of cholesterol and sphingomyelin metabolism by amyloid- beta and presenilin, Nat. Cell Biol. 7 (2005) 1118–1123.

70

E. Porcellini et al. / Neuroscience Letters 416 (2007) 66–70

[14] D.S. Howland, S.P. Trusko, M.J. Savage, A.G. Reaume, D.M. Lang, J.D. Hirsch, N. Maeda, R. Siman, B.D. Greenberg, R.W. Scott, D.G. Flood, Modulation of secreted beta-amyloid precursor protein and amyloid beta-peptide in brain by cholesterol, J. Biol. Chem. 273 (1998) 16576– 16582. [15] M.J. Ignatius, E.M. Shooter, R.E. Pitas, R.W. Mahley, Lipoprotein uptake by neuronal growth cones in vitro, Science 236 (1987) 959–962. [16] G.P. Jarivk, E.M. Wijsman, W.A. Kukull, G.D. Schellenberg, C. Yu, E.B. Larson, Interaction of Apolipoprotein E genotype, total cholesterol level, age and sex in prediction of Alzheimer’s disease: a case control study, Neurology 45 (1995) 1092–1096. [17] A. Johansson, H. Katzov, H. Zetterberg, L. Feuk, B. Johansson, N. Bogdanovic, N. Andreasen, B. Lenhard, A.J. Brookes, N.L. Pedersen, K. Blennow, J.A. Prince, Variants of CYP46A1 may interact with age and APOE to influence CSF Abeta42 levels in Alzheimer’s disease, Hum. Genet. 114 (2004) 581–587. [18] H. Kolsch, U. Ptok, I. Mohamed, S. Schmitz, M.L. Rao, W. Maier, R. Heun, Association of the C766T polymorphism of the low-density lipoprotein receptor-related protein gene with Alzheimer’s disease, Am. J. Med. Genet. 121 (2003) 128–130. [19] A.R. Kuodinov, T.T. Berezov, N. Kuodinova, Alzheimer’s amyloid beta and lipid metabolism: a missing link? FASEB 12 (1998) 1097–1099. [20] F. Licastro, S. Pedrini, M. Govoni, A. Pession, C. Ferri, G. Annoni, V. Casadei, F. Veglia, S. Bertolini, L.M. Grimaldi, Apolipoprotein E and alpha-1-antichymotrypsin allele polymorphism in sporadic and familial Alzheimer’s disease, Neurosci. Lett. 270 (1999) 129–132. [21] G. McKhann, D. Drachman, M. Folstein, R. Katzman, D. Price, E.M. Stadlan, Clinical diagnosis of Alzheimer’s disease: report of the NINCDSADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease, Neurology 34 (1984) 939–944. [22] K. Morgan, F. Licastro, L. Tilley, A. Ritchie, L. Morgan, S. Pedrini, N. Kalsheker, Polymorphism in the alpha(1)-antichymotrypsin (ACT) gene promoter: effect on expression in transfected glial and liver cell lines and plasma ACT concentrations, Hum. Genet. 109 (2001) 303–310.

[23] A. Papassotiropoulos, J.R. Steffer, M. Tlolaki, S. Schmid, D. Thal, F.I. Nicosia, Increased brain beta amyloid load, phosphorilated tau and risk of Alzheimer disease associated with an intronic CYP46 polymorphism, Arch. Neurol. 60 (2003) 16–18. [24] I.H. Park, E.M. Hwang, H.S. Hong, J.H. Boo, S.S. Oh, J. Lee, M.W. Jung, O.Y. Bang, S.U. Kim, I. Mook-Jung, Lovastatin enhances Abeta production and senile plaque deposition in female Tg2576 mice, Neurobiol. Aging 24 (2003) 637–643. [25] L.M. Refolo, B. Malester, J. LaFrancois, T. Bryant-Thomas, R. Wang, G.S. Tint, K. Sambamurti, K. Duff, M.A. Pappolla, Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model, Neurobiol. Dis. 7 (2000) 321–331. [26] K. Rockwood, S. Kirkland, D.B. Hogan, C. MacKnight, H. Merry, R. Verreault, C. Wolfson, I. McDowell, Use of lipid-lowering agents, indication bias, and the risk of dementia in community-dwelling elderly people, Arch. Neurol. 59 (2002) 223–227. [27] D.C. Rubinsztein, The genetics of Alzheimer’s disease, Prog. Neurobiol. 52 (1997) 447–454. [28] J.G. Schulz, J. Bosel, M. Stoeckel, D. Megow, U. Dirnagl, M. Endres, HMG-CoA reductase inhibition causes neurite loss by interfering with geranylgeranylpyrophosphate synthesis, J. Neurochem. 89 (2004) 24–32. [29] M. Simons, P. Keller, B. De Strooper, K. Beyreuther, C.G. Dotti, K. Simons, Cholesterol depletion inhibits the generation of beta amyloid in hyppocampal neurons, Proc. Nat. Acad. Sci. USA 95 (1998) 6460–6464. [30] M.A. Wollmer, J.R. Streffer, D. Lutjohann, M. Tsolaki, V. Iakovidou, T. Hegi, T. Pasch, H.H. Jung, K. Bergmann, R.M. Nitsch, C. Hock, A. Papassotiropoulos, ABCA1 modulates CSF cholesterol levels and influences the age at onset of Alzheimer’s disease, Neurobiol. Aging 24 (2003) 421–426. [31] M.A. Wollmer, J.R. Streffer, M. Tsolaki, L.M. Grimaldi, D. Lutjohann, D. Thal, K. Von Bergmann, R.M. Nitsch, C. Hock, A. Papassotiropoulos, Genetic association of acyl-coenzyme A: cholesterol acyl-transferase with cerebrospinal fluid cholesterol levels, brain amyloid load, and risk for Alzheimer’s disease, Mol. Psychiatry 8 (2003) 635–638. [32] F.L. Zhang, P.J. Casey, Protein prenylation: molecular mechanisms and functional consequences, Annu. Rev. Biochem. 65 (1996) 241–269.