- Email: [email protected]
An age effect on the association of common variants of ACE with Alzheimer's disease
Neuroscience Letters 461 (2009) 181–184
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
An age effect on the association of common variants of ACE with Alzheimer’s disease Nicole Helbecque a,b,c,d , Valerie Codron a,b,c,d , Dominique Cottel a,b,c,d , Philippe Amouyel a,b,c,d,∗ a
Univ Lille Nord de France, F-59000 Lille, France INSERM, U744, F-59000 Lille, France UDSL, F-59000 Lille, France d Institut Pasteur de Lille, F-59000 Lille, France b c
a r t i c l e
i n f o
Article history: Received 28 April 2009 Received in revised form 2 June 2009 Accepted 11 June 2009 Keywords: Alzheimer’s disease Association study Genetic polymorphism Angiotensin-I converting enzyme
a b s t r a c t It is now well established that vascular risk factors are associated with cognitive performances. The renin–angiotensin system (RAS) components, major determinants of the cardiovascular system, are expressed in the brain and were shown to play a role on amyloid metabolism, learning and memory. The angiotensin-converting enzyme (ACE), a pivotal RAS protein, is encoded by a huge gene containing many variants, one of them, the I/D variant (rs1799752), being associated with Alzheimer’s disease (AD). Other variants, such as SNPs rs4291A>T located −240 bp from the initiation codon, and rs4343G>A encoding a silent mutation in exon 16, were inconsistently associated with the risk of AD. In a case–control study including 376 late-onset AD patients and 444 control subjects, we showed a statistically significant effect on the risk of AD of two variants (rs4343 and rs1799752) and of the haplotype ATI (rs4343/rs4291/rs1799752) in subjects aged 73 years and above. © 2009 Elsevier Ireland Ltd. All rights reserved.
Alzheimer’s disease (AD) is the most common form of dementia, contributing to about two thirds of all dementias [19], and the greatest risk factor for the development of AD is advancing age. AD is a heterogeneous disorder with both familial (about 1% of cases) and sporadic forms. This disease is mainly characterized by the presence of senile plaques and neurofibrillary tangles in the medial temporal lobe structures and cortical areas of the brain. The major constituent of the plaques is the beta amyloid peptide (A) generated by enzymatic cleavage from its precursor, the amyloid precursor protein (APP). Mutations in the genes encoding APP and presenilins-1 and -2 account for most cases of familial AD, that is less than 1% of AD cases. Up to now only the apolipoprotein E (APOE) 4 allele has been reported as a genetic risk factor for sporadic cases. All the genome-wide association studies (GWAS) published these last 3 years [2,3,10,29,33] confirm this result and lead to the inconsistent identification of various other genes as risk factors for AD. Up to date no consensus for one (or more) specific gene(s) has been found. In the past we investigated a possible impact of the gene encoding the angiotensin-converting enzyme (ACE) on the risk of cognitive impairment [1] or of dementia of the AD type [5,23,24]. This work was sustained by epidemiological studies pointing to a link between hypertension or risk factors for atherosclerotic vascu-
∗ Corresponding author at: INSERM, U744, Institut Pasteur de Lille, 1 rue Calmette, BP 245, 59019 Lille Cedex, France. Tel.: +33 3 20 87 77 10; fax: +33 3 20 87 78 94. E-mail address: [email protected] (P. Amouyel). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.06.006
lar disease and AD [27], possibly involving mechanisms associated with hypertension-related pathologies, such as cerebrovascular disease, atherosclerosis, and arteriosclerosis [4]. Interestingly previous large epidemiological studies suggested that brainpenetrating angiotensin-converting enzyme (ACE) inhibitors had a beneficial effect on the rate of cognitive decline [11,22,31]. Several biological studies suggest that the renin–angiotensin system (RAS), and most specifically ACE, could play a role in AD pathology by acting on amyloid -protein metabolism. This hypothesis was reinforced by experiments in rats models of AD, showing an increase of A deposition and an increase of ACE activity in the brains of AD rats, whereas treatment by perindropril, an ACE inhibitor, correlated with a decrease of ACE activity and of A deposition, leading to delayed onset of AD-like pathology [14]. This study was in accordance with a previous work showing an important role of ACE in converting the highly amyloidogenic A(1–42) to the less neurotoxic A(1–40) in mouse and human brain homogenates, whereas the use of captopril, another ACE inhibitor, impeded this process in a mouse model of AD [34]. Various mechanisms could explain the observed effect of ACE inhibitors on AD. For instance the ACE inhibitors have been reported to increase brain substance P (which is normally decreased by ACE), which in turn increases the activity of neprilysin, a well known A degrading enzyme [7]. On the other hand, a reduced availability of ACE could lead to an increase of the concentrations of the amyloid peptide A, a hallmark of AD, as shown by several in vitro studies ([12,28] and references therein), supporting the role of ACE as a possible A degrading
0.79 0.064 0.072 The bold values correspond to significant or borderline p values.
P I
342 (45.5) 120 (48.8) 222 (43.9) 410 (54.5) 126 (51.2) 284 (56.1) 398 (44.8) 239 (41.8) 159 (50.3) 490 (55.2) 333 (58.2) 157 (49.7) 0.26 0.15 0.0075 67 (17.8) 29 (23.6) 38 (15.0) 208 (55.3) 62 (50.4) 146 (57.7) 101 (26.9) 32 (26.0) 69 (27.3) 89 (20.0) 46 (16.1) 43 (27.2)
P II ID II ID
220 (49.6) 147 (51.4) 73 (46.2) 135 (30.4) 93 (32.5) 42 (26.6)
Patients
D DD DD
I Control subjects Patients Control subjects
D
Alleles, n (%) Genotypes, n (%) ACE 1799752
All (444/376) Age <73 Age ≥73
0.21 0.14 0.033 275 (36.6) 81 (32.9) 194 (38.3) 477 (63.4) 165 (67.1) 312 (61.7) 264 (41.0) 219 (38.3) 145 (45.9) 524 (59.0) 353 (61.7) 171 (54.1) 0.16 0.14 0.06 44 (11.7) 14 (11.4) 30 (11.8) 187 (49.7) 53 (43.1) 134 (53.0) 145 (38.6) 56 (45.5) 89 (35.2) 65 (14.6) 34 (11.9) 31 (19.6) 234 (52.7) 151 (52.8) 83 (52.5)
P TT AT TT AT
145 (32.7) 101 (35.3) 44 (27.9)
T Patients
A T
Control subjects
AA AA
A
Alleles, n (%)
Patients Genotypes, n (%)
Control subjects
All (444/376) Age <73 Age ≥73
P
0.88 0.10 0.033 ACE 4291
G
406 (54.0) 124 (50.4) 282 (55.7) 346 (46.0) 122 (49.6) 224 (44.3)
A G
476 (53.6) 324 (56.6) 152 (48.1) 412 (46.4) 248 (43.4) 164 (51.9)
A P
0.96 0.10 0.03 105 (27.9) 33 (26.8) 72 (28.5)
GG AG
196 (52.1) 58 (47.2) 138 (54.5) 75 (20.0) 32 (26.0) 43 (17.0)
AA
124 (27.9) 86 (30.1) 38 (24.1)
GG AG
228 (51.4) 152 (53.1) 76 (48.1)
AA
92 (20.7) 48 (16.8) 44 (27.8)
Control subjects
All (444/376) Age <73 Age ≥73
Patients Alleles, n (%)
Patients Genotypes, n (%)
Control subjects
ACE 4343
enzyme. These studies showed that ACE exerts an endopeptidolytic attack on the N-terminal 4–15 fragment of human A as well as on human A(1–40), these effects being blocked by ACE inhibitors ([12,28] and references therein). However no change in brain A levels according to ACE availability was observed either in mice [6,13,32] or in human [18,21], suggesting that ACE did not play any role in A degradation in vivo. Moreover these mechanisms seem to disagree with previous studies reporting that reduced concentrations of plasma ACE, contributed by a common insertion/deletion (I/D) polymorphism in the ACE gene (rs1799752), were associated with an increased risk of AD [9,17]. The location of the ACE I/D polymorphism in a noncoding region of the gene makes it unlikely to be a functional variant. Several studies suggested that at least two single nucleotide polymorphisms (SNPs), one in the 5 part of the gene, the other most likely located between intron 18 and the 3 end of the gene, are necessary to explain functionality [25]. Indeed two SNPs (SNPs rs4291A>T located −240 bp from the initiation codon, and rs4343G>A encoding a silent mutation in exon 16) were suggested to influence A42 levels in the cerebrospinal fluid (CSF) [16]. Both were reported as statistically significantly associated with the risk of developing AD in independent Caucasian population samples [16]. One of these SNPs (rs4343G>A) was also associated with AD in an Israeli Arab community [20]. We previously tested for a possible impact of these two SNPs (rs4291A>T and rs4343G>A) on the risk of AD in a large population-based cohort of 9294 subjects [5], 141 of them developing AD during follow-up. No effect could be set in evidence, probably due to a lack of statistical power. In the present study we examined the I/D polymorphism (rs1799752) together with rs4291 and rs4343 in a European Caucasian population sample including 444 controls and 376 late-onset AD patients for which we previously showed an impact of the ACE D allele on the risk of AD [24]. Patients (35.9% men, mean age ± SD: 75 ± 5 years) recruited in hospitals in France (70%), UK (9%), Netherlands (5%), Spain (9%) and Italy (7%) were thoroughly investigated clinically. The diagnosis of AD was made according to the DSM-III-R and NINCDS-ADRDA criteria. None of the subjects had a family history of AD consistent with Mendelian inheritance. Caucasian controls (35.8% men, mean age ± SD: 74 ± 8 years) were recruited in France and defined as subjects without DSM-III-R dementia criteria and with integrity of their cognitive functions. The study was approved by the Ethics Committee of the Lille University Hospital. Each individual or their relatives signed an informed consent. For each participant, genomic DNA was prepared from white blood cells using a salt-out procedure. The ACE rs1799752 polymorphism was previously genotyped in this population [24], whereas genotyping of SNPs rs4291A>T and rs4343G>A was performed as previously described [5]. The laboratory personnel were blind to clinical information concerning the subjects tested. 2 analysis was used to compare genotype and allele frequencies and to assess deviation from Hardy–Weinberg equilibrium. Statistical analyses were performed with the SAS software release 8.2 (SAS Institute Inc., Cary, NC, USA). A statistically significant interaction with APOE status was systematically looked for. Effects of haplotypes were tested using the THESIAS (Testing Haplotype Effects In Association Studies) software. This program is based on a maximum likelihood model and is linked to the SEM algorithm [30]. Statistical significance was considered at the P < 0.05 level. The ACE SNPs rs4343, rs4291 and rs1799752 genotype distributions in control subjects were in accordance with the Hardy–Weinberg equilibrium (rs4343: 2 = 0.466, 2 df, P = 0.49; rs4291: 2 = 3.550, 2 df, P = 0.06; rs1799752: 2 = 0.001, 2 df, P = 0.97). In AD cases, there were no statistically significant differ-
P
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
Table 1 Haplotype frequencies and ORs of AD risk in the studied population sample.
182
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
183
Table 2 Haplotype frequencies and ORs of AD risk in the studied population sample. Haplotype ACE 4343/4291/1799752
Haplotype frequency
Haplotypic OR [95% CI]
Population-adjusteda haplotypic OR [95% CI], P
Control subjects
Cases
(A) The whole population sample AAI AAD ATI GAD GTD
0.367 0.015 0.079 0.207 0.327
0.425 0.023 0.013 0.178 0.345
1.17 [0.53–2.58] 0.14 [0.06–0.32] 0.72 [0.55–0.96] 0.84 [0.66–1.07]
Reference 1.34 [0.58–3.08], P = 0.49 0.09 [0.04–0.21], P < 10−4 0.67 [0.48–0.92], P = 0.015 0.81 [0.62–1.07], P = 0.13
(B) Subjects aged above 73 years AAI AAD ATI GAD GTD
0.033 0.015 0.177 0.200 0.281
0.402 0.015 0.021 0.185 0.359
0.59 [0.16–2.11] 0.08 [0.03–0.20] 0.70 [0.45–1.10] 0.89 [0.61–1.29]
Reference 0.51 [0.11–2.45], P = 0.40 0.08 [0.03–0.24], P = 10−5 0.63 [0.35–1.13], P = 0.12 1.23 [0.78–1.95], P = 0.38
a
Adjustment for age, sex and APOE status.
ences for ACE SNPs rs4343, rs4291 and rs1799752 genotype and allele distributions between North European and South European subjects (rs4343: 2 = 2.770, 2 df, P = 0.25; rs4291: 2 = 0.012, 2 df, P = 0.99; rs1799752: 2 = 2.004, 2 df, P = 0.37). For the whole population sample, the genotype and allele distributions did not differ according to cognitive status (Table 1). Since age is a well known risk factor for the development of AD, we stratified the population sample on the median age of the AD sample (under or above 73 years old). The oldest patients bearing either the ACE rs4343G or the ACE rs1799752D allele were at reduced risk of AD (Table 1; for rs4343 AA vs AG + GG: OR = 0.49, 95% CI [0.27–0.88], P = 0.02; for rs4291 AA vs AT + TT: OR = 1.20, 95% CI [0.72–2.02], P = 0.48; for ACE rs1799752 II vs ID + DD: OR = 0.43, 95% CI [0.24–0.79], P = 0.007, adjusted for age, sex, APOE status). We then performed a haplotype analysis in the whole population sample, including the three polymorphisms in the following order: rs4343, rs4291 and rs1799752. A highly significant global effect (P = 1.3 × 10−8 after adjustment) of the haplotypes on the risk of AD was observed. The frequency of the ATI combination was lower in cases than in control subjects (OR = 0.09, 95% CI [0.04–0.21], P < 10−4 , adjusted for age, sex, APOE status) (Table 2) opposite to our findings in the 3C cohort. When dividing the population sample according to median age, the protective haplotype effect remains statistically significant only for the oldest (age ≥73 years: for ATI vs AAI OR = 0.08, 95% CI [0.03–0.24], P = 10−5 , adjusted on age, sex, APOE status) (Table 2). A similar effect was also found for the AT haplotype when considering only rs4343 and rs4291 (OR = 0.10, 95% CI [0.03–0.31], P = 6 × 10−5 , adjusted for age, sex, APOE status). The main result of this study is the presence of a statistically significant effect of the common ACE polymorphisms rs4343, rs4291, and rs1799752 on the risk of AD in Caucasian subjects aged 73 years and above. A haplotypic effect could also be set in evidence in this population sample. Although no significant effect on the risk of AD was reported for either ACE rs4343 (6 studies), rs4291 (5 studies), or rs1799752 (30 studies) in various separate studies, meta-analyses in Caucasians (http://www.alzforum.org/res/com/gen/alzgene/default.asp) showed borderline effects on the risk of AD for each of them (rs4343G allele: OR = 0.91, 95% [0.82–1.00]; rs4291T allele: OR = 0.87, 95% CI [0.76–0.99]; rs1799752I allele: OR = 1.05, 95% CI [0.99–1.12]). Although a small, significantly underpowered, GWAS conducted on AD patients with a mean age of 72.2 years did not show any association, either direct or indirect, of rs4343 or rs4291 with AD [10], many recent association studies [8,15,26] and a GWAS [29] are in favour of a role of ACE haplotypes in AD, in accordance with our results.
In conclusion, it seems that ACE play a modest role in the development of AD, at least in the oldest patients. Acknowledgements We thank all the patients and their families. This study was supported by the Institut National de la Santé et de la Recherche Médicale, by the Centre Hospitalier et Universitaire de Lille, and by the Institut Pasteur de Lille. References [1] P. Amouyel, F. Richard, D. Cottel, C. Amant, V. Codron, N. Helbecque, The deletion allele of the angiotensin I converting enzyme gene as a genetic susceptibility factor for cognitive impairment, Neurosci. Lett. 217 (1996) 203–205. [2] G.W. Beecham, E.R. Martin, Y.J. Li, M.A. Slifer, J.R. Gilbert, J.L. Haines, M.A. Pericak-Vance, Genome-wide association study implicates a chromosome 12 risk locus for late-onset Alzheimer disease, Am. J. Hum. Genet. 84 (2009) 35–43. [3] L. Bertram, C. Lange, K. Mullin, M. Parkinson, M. Hsiao, M.F. Hogan, B.M. Schjeide, B. Hooli, J. Divito, I. Ionita, H. Jiang, N. Laird, T. Moscarillo, K.L. Ohlsen, K. Elliott, X. Wang, D. Hu-Lince, M. Ryder, A. Murphy, S.L. Wagner, D. Blacker, K.D. Becker, R.E. Tanzi, Genome-wide association analysis reveals putative Alzheimer’s disease susceptibility loci in addition to APOE, Am. J. Hum. Genet. 83 (2008) 623–632. [4] K. Blennow, M.J. de Leon, H. Zetterberg, Alzheimer’s disease, Lancet 368 (2006) 387–403. [5] A. Bruandet, F. Richard, C. Tzourio, C. Berr, J.F. Dartigues, A. Alperovitch, P. Amouyel, N. Helbecque, Haplotypes across ACE and the risk of Alzheimer’s disease: the three-city study, J. Alzheimers Dis. 13 (2008) 333–339. [6] E.A. Eckman, S.K. Adams, F.J. Troendle, B.A. Stodola, M.A. Kahn, A.H. Fauq, H.D. Xiao, K.E. Bernstein, C.B. Eckman, Regulation of steady-state beta-amyloid levels in the brain by neprilysin and endothelin-converting enzyme but not angiotensin-converting enzyme, J. Biol. Chem. 281 (2006) 30471–30478. [7] E.A. Eckman, C.B. Eckman, Abeta-degrading enzymes: modulators of Alzheimer’s disease pathogenesis and targets for therapeutic intervention, Biochem. Soc. Trans. 33 (2005) 1101–1105. [8] T.L. Edwards, M. Pericak-Vance, J.R. Gilbert, J.L. Haines, E.R. Martin, M.D. Ritchie, An association analysis of Alzheimer disease candidate genes detects an ancestral risk haplotype clade in ACE and putative multilocus association between ACE, A2M, and LRRTM3, Am. J. Med. Genet. B: Neuropsychiatr. Genet., doi:10.1002/ajmg.b.30899. [9] J.S. Elkins, V.C. Douglas, S.C. Johnston, Alzheimer disease risk and genetic variation in ACE: a meta-analysis, Neurology 62 (2004) 363–368. [10] T.M. Feulner, S.M. Laws, P. Friedrich, S. Wagenpfeil, S.H. Wurst, C. Riehle, K.A. Kuhn, M. Krawczak, S. Schreiber, S. Nikolaus, H. Forstl, A. Kurz, M. Riemenschneider, Examination of the current top candidate genes for AD in a genome-wide association study, Mol. Psychiatry (2009), doi:10.1038/mp.2008.141. [11] O. Hanon, F. Forette, Prevention of dementia: lessons from SYST-EUR and PROGRESS, J. Neurol. Sci. 226 (2004) 71–74. [12] M.L. Hemming, D.J. Selkoe, Amyloid beta-protein is degraded by cellular angiotensin-converting enzyme (ACE) and elevated by an ACE inhibitor, J. Biol. Chem. 280 (2005) 37644–37650. [13] M.L. Hemming, D.J. Selkoe, W. Farris, Effects of prolonged angiotensinconverting enzyme inhibitor treatment on amyloid beta-protein metabolism in mouse models of Alzheimer disease, Neurobiol. Dis. 26 (2007) 273–281. [14] D.R. Hou, Y. Wang, L. Zhou, K. Chen, Y. Tian, Z. Song, J. Bao, Q.D. Yang, Altered angiotensin-converting enzyme and its effects on the brain in a rat model of Alzheimer disease, Chin. Med. J. (Engl.) 121 (2008) 2320–2323.
184
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
[15] J.S. Kauwe, J. Wang, K. Mayo, J.C. Morris, A.M. Fagan, D.M. Holtzman, A.M. Goate, Alzheimer’s disease risk variants show association with cerebrospinal fluid amyloid beta, Neurogenetics 10 (2009) 13–17. [16] P.G. Kehoe, H. Katzov, L. Feuk, A.M. Bennet, B. Johansson, B. Wiman, U. de Faire, N.J. Cairns, G.K. Wilcock, A.J. Brookes, K. Blennow, J.A. Prince, Haplotypes extending across ACE are associated with Alzheimer’s disease, Hum. Mol. Genet. 12 (2003) 859–867. [17] D.J. Lehmann, M. Cortina-Borja, D.R. Warden, A.D. Smith, K. Sleegers, J.A. Prince, C.M. van Duijn, P.G. Kehoe, Large meta-analysis establishes the ACE insertion–deletion polymorphism as a marker of Alzheimer’s disease, Am. J. Epidemiol. 162 (2005) 305–317. [18] C.L. Lendon, U. Thaker, J.M. Harris, A.M. McDonagh, J.C. Lambert, M.C. Chartier-Harlin, T. Iwatsubo, S.M. Pickering-Brown, D.M. Mann, The angiotensin 1-converting enzyme insertion (I)/deletion (D) polymorphism does not influence the extent of amyloid or tau pathology in patients with sporadic Alzheimer’s disease, Neurosci. Lett. 328 (2002) 314–318. [19] J.A. Luchsinger, R. Mayeux, Cardiovascular risk factors and Alzheimer’s disease, Curr. Atheroscler. Rep. 6 (2004) 261–266. [20] Y. Meng, C.T. Baldwin, A. Bowirrat, K. Waraska, R. Inzelberg, R.P. Friedland, L.A. Farrer, Association of polymorphisms in the angiotensin-converting enzyme gene with Alzheimer disease in an Israeli Arab community, Am. J. Hum. Genet. 78 (2006) 871–877. [21] S. Miners, E. Ashby, S. Baig, R. Harrison, H. Tayler, E. Speedy, J.A. Prince, S. Love, P.G. Kehoe, Angiotensin-converting enzyme levels and activity in Alzheimer’s disease: differences in brain and CSF ACE and association with ACE1 genotypes, Am. J. Transl. Res. 1 (2009) 163–177. [22] T. Ohrui, T. Matsui, M. Yamaya, H. Arai, S. Ebihara, M. Maruyama, H. Sasaki, Angiotensin-converting enzyme inhibitors and incidence of Alzheimer’s disease in Japan, J. Am. Geriatr. Soc. 52 (2004) 649–650. [23] F. Richard, C. Berr, C. Amant, N. Helbecque, P. Amouyel, A. Alperovitch, Effect of the angiotensin I-converting enzyme I/D polymorphism on cognitive decline. The EVA Study Group, Neurobiol. Aging 21 (2000) 75–80. [24] F. Richard, I. Fromentin-David, F. Ricolfi, P. Ducimetiere, C. Di Menza, P. Amouyel, N. Helbecque, The angiotensin I converting enzyme gene as a susceptibility factor for dementia, Neurology 56 (2001) 1593–1595.
[25] F.A. Sayed-Tabatabaei, B.A. Oostra, A. Isaacs, C.M. van Duijn, J.C. Witteman, ACE polymorphisms, Circ. Res. 98 (2006) 1123–1133. [26] B.M. Schjeide, M.B. McQueen, K. Mullin, J. DiVito, M.F. Hogan, M. Parkinson, B. Hooli, C. Lange, D. Blacker, R.E. Tanzi, L. Bertram, Assessment of Alzheimer’s disease case-control associations using family-based methods, Neurogenetics 10 (2009) 19–25. [27] J.A. Staessen, T. Richart, W.H. Birkenhager, Less atherosclerosis and lower blood pressure for a meaningful life perspective with more brain, Hypertension 49 (2007) 389–400. [28] X. Sun, M. Becker, K. Pankow, E. Krause, M. Ringling, M. Beyermann, B. Maul, T. Walther, W.E. Siems, Catabolic attacks of membrane-bound angiotensinconverting enzyme on the N-terminal part of species-specific amyloid-beta peptides, Eur. J. Pharmacol. 588 (2008) 18–25. [29] T.A. Thornton-Wells, J.H. Moore, E.R. Martin, M.A. Pericak-Vance, J.L. Haines, Confronting complexity in late-onset Alzheimer disease: application of two-stage analysis approach addressing heterogeneity and epistasis, Genet. Epidemiol. 32 (2008) 187–203. [30] D.A. Tregouet, S. Barbaux, S. Escolano, N. Tahri, J.L. Golmard, L. Tiret, F. Cambien, Specific haplotypes of the P-selectin gene are associated with myocardial infarction, Hum. Mol. Genet. 11 (2002) 2015–2023. [31] C. Tzourio, C. Anderson, N. Chapman, M. Woodward, B. Neal, S. MacMahon, J. Chalmers, Effects of blood pressure lowering with perindopril and indapamide therapy on dementia and cognitive decline in patients with cerebrovascular disease, Arch. Intern. Med. 163 (2003) 1069– 1075. [32] J. Wang, L. Ho, L. Chen, Z. Zhao, W. Zhao, X. Qian, N. Humala, I. Seror, S. Bartholomew, C. Rosendorff, G.M. Pasinetti, Valsartan lowers brain betaamyloid protein levels and improves spatial learning in a mouse model of Alzheimer disease, J. Clin. Invest. 117 (2007) 3393–3402. [33] S.C. Waring, R.N. Rosenberg, Genome-wide association studies in Alzheimer disease, Arch. Neurol. 65 (2008) 329–334. [34] K. Zou, H. Yamaguchi, H. Akatsu, T. Sakamoto, M. Ko, K. Mizoguchi, J.S. Gong, W. Yu, T. Yamamoto, K. Kosaka, K. Yanagisawa, M. Michikawa, Angiotensinconverting enzyme converts amyloid beta-protein 1–42 (Abeta(1–42)) to Abeta(1–40), and its inhibition enhances brain Abeta deposition, J. Neurosci. 27 (2007) 8628–8635.
Contents lists available at ScienceDirect
Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet
An age effect on the association of common variants of ACE with Alzheimer’s disease Nicole Helbecque a,b,c,d , Valerie Codron a,b,c,d , Dominique Cottel a,b,c,d , Philippe Amouyel a,b,c,d,∗ a
Univ Lille Nord de France, F-59000 Lille, France INSERM, U744, F-59000 Lille, France UDSL, F-59000 Lille, France d Institut Pasteur de Lille, F-59000 Lille, France b c
a r t i c l e
i n f o
Article history: Received 28 April 2009 Received in revised form 2 June 2009 Accepted 11 June 2009 Keywords: Alzheimer’s disease Association study Genetic polymorphism Angiotensin-I converting enzyme
a b s t r a c t It is now well established that vascular risk factors are associated with cognitive performances. The renin–angiotensin system (RAS) components, major determinants of the cardiovascular system, are expressed in the brain and were shown to play a role on amyloid metabolism, learning and memory. The angiotensin-converting enzyme (ACE), a pivotal RAS protein, is encoded by a huge gene containing many variants, one of them, the I/D variant (rs1799752), being associated with Alzheimer’s disease (AD). Other variants, such as SNPs rs4291A>T located −240 bp from the initiation codon, and rs4343G>A encoding a silent mutation in exon 16, were inconsistently associated with the risk of AD. In a case–control study including 376 late-onset AD patients and 444 control subjects, we showed a statistically significant effect on the risk of AD of two variants (rs4343 and rs1799752) and of the haplotype ATI (rs4343/rs4291/rs1799752) in subjects aged 73 years and above. © 2009 Elsevier Ireland Ltd. All rights reserved.
Alzheimer’s disease (AD) is the most common form of dementia, contributing to about two thirds of all dementias [19], and the greatest risk factor for the development of AD is advancing age. AD is a heterogeneous disorder with both familial (about 1% of cases) and sporadic forms. This disease is mainly characterized by the presence of senile plaques and neurofibrillary tangles in the medial temporal lobe structures and cortical areas of the brain. The major constituent of the plaques is the beta amyloid peptide (A) generated by enzymatic cleavage from its precursor, the amyloid precursor protein (APP). Mutations in the genes encoding APP and presenilins-1 and -2 account for most cases of familial AD, that is less than 1% of AD cases. Up to now only the apolipoprotein E (APOE) 4 allele has been reported as a genetic risk factor for sporadic cases. All the genome-wide association studies (GWAS) published these last 3 years [2,3,10,29,33] confirm this result and lead to the inconsistent identification of various other genes as risk factors for AD. Up to date no consensus for one (or more) specific gene(s) has been found. In the past we investigated a possible impact of the gene encoding the angiotensin-converting enzyme (ACE) on the risk of cognitive impairment [1] or of dementia of the AD type [5,23,24]. This work was sustained by epidemiological studies pointing to a link between hypertension or risk factors for atherosclerotic vascu-
∗ Corresponding author at: INSERM, U744, Institut Pasteur de Lille, 1 rue Calmette, BP 245, 59019 Lille Cedex, France. Tel.: +33 3 20 87 77 10; fax: +33 3 20 87 78 94. E-mail address: [email protected] (P. Amouyel). 0304-3940/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2009.06.006
lar disease and AD [27], possibly involving mechanisms associated with hypertension-related pathologies, such as cerebrovascular disease, atherosclerosis, and arteriosclerosis [4]. Interestingly previous large epidemiological studies suggested that brainpenetrating angiotensin-converting enzyme (ACE) inhibitors had a beneficial effect on the rate of cognitive decline [11,22,31]. Several biological studies suggest that the renin–angiotensin system (RAS), and most specifically ACE, could play a role in AD pathology by acting on amyloid -protein metabolism. This hypothesis was reinforced by experiments in rats models of AD, showing an increase of A deposition and an increase of ACE activity in the brains of AD rats, whereas treatment by perindropril, an ACE inhibitor, correlated with a decrease of ACE activity and of A deposition, leading to delayed onset of AD-like pathology [14]. This study was in accordance with a previous work showing an important role of ACE in converting the highly amyloidogenic A(1–42) to the less neurotoxic A(1–40) in mouse and human brain homogenates, whereas the use of captopril, another ACE inhibitor, impeded this process in a mouse model of AD [34]. Various mechanisms could explain the observed effect of ACE inhibitors on AD. For instance the ACE inhibitors have been reported to increase brain substance P (which is normally decreased by ACE), which in turn increases the activity of neprilysin, a well known A degrading enzyme [7]. On the other hand, a reduced availability of ACE could lead to an increase of the concentrations of the amyloid peptide A, a hallmark of AD, as shown by several in vitro studies ([12,28] and references therein), supporting the role of ACE as a possible A degrading
0.79 0.064 0.072 The bold values correspond to significant or borderline p values.
P I
342 (45.5) 120 (48.8) 222 (43.9) 410 (54.5) 126 (51.2) 284 (56.1) 398 (44.8) 239 (41.8) 159 (50.3) 490 (55.2) 333 (58.2) 157 (49.7) 0.26 0.15 0.0075 67 (17.8) 29 (23.6) 38 (15.0) 208 (55.3) 62 (50.4) 146 (57.7) 101 (26.9) 32 (26.0) 69 (27.3) 89 (20.0) 46 (16.1) 43 (27.2)
P II ID II ID
220 (49.6) 147 (51.4) 73 (46.2) 135 (30.4) 93 (32.5) 42 (26.6)
Patients
D DD DD
I Control subjects Patients Control subjects
D
Alleles, n (%) Genotypes, n (%) ACE 1799752
All (444/376) Age <73 Age ≥73
0.21 0.14 0.033 275 (36.6) 81 (32.9) 194 (38.3) 477 (63.4) 165 (67.1) 312 (61.7) 264 (41.0) 219 (38.3) 145 (45.9) 524 (59.0) 353 (61.7) 171 (54.1) 0.16 0.14 0.06 44 (11.7) 14 (11.4) 30 (11.8) 187 (49.7) 53 (43.1) 134 (53.0) 145 (38.6) 56 (45.5) 89 (35.2) 65 (14.6) 34 (11.9) 31 (19.6) 234 (52.7) 151 (52.8) 83 (52.5)
P TT AT TT AT
145 (32.7) 101 (35.3) 44 (27.9)
T Patients
A T
Control subjects
AA AA
A
Alleles, n (%)
Patients Genotypes, n (%)
Control subjects
All (444/376) Age <73 Age ≥73
P
0.88 0.10 0.033 ACE 4291
G
406 (54.0) 124 (50.4) 282 (55.7) 346 (46.0) 122 (49.6) 224 (44.3)
A G
476 (53.6) 324 (56.6) 152 (48.1) 412 (46.4) 248 (43.4) 164 (51.9)
A P
0.96 0.10 0.03 105 (27.9) 33 (26.8) 72 (28.5)
GG AG
196 (52.1) 58 (47.2) 138 (54.5) 75 (20.0) 32 (26.0) 43 (17.0)
AA
124 (27.9) 86 (30.1) 38 (24.1)
GG AG
228 (51.4) 152 (53.1) 76 (48.1)
AA
92 (20.7) 48 (16.8) 44 (27.8)
Control subjects
All (444/376) Age <73 Age ≥73
Patients Alleles, n (%)
Patients Genotypes, n (%)
Control subjects
ACE 4343
enzyme. These studies showed that ACE exerts an endopeptidolytic attack on the N-terminal 4–15 fragment of human A as well as on human A(1–40), these effects being blocked by ACE inhibitors ([12,28] and references therein). However no change in brain A levels according to ACE availability was observed either in mice [6,13,32] or in human [18,21], suggesting that ACE did not play any role in A degradation in vivo. Moreover these mechanisms seem to disagree with previous studies reporting that reduced concentrations of plasma ACE, contributed by a common insertion/deletion (I/D) polymorphism in the ACE gene (rs1799752), were associated with an increased risk of AD [9,17]. The location of the ACE I/D polymorphism in a noncoding region of the gene makes it unlikely to be a functional variant. Several studies suggested that at least two single nucleotide polymorphisms (SNPs), one in the 5 part of the gene, the other most likely located between intron 18 and the 3 end of the gene, are necessary to explain functionality [25]. Indeed two SNPs (SNPs rs4291A>T located −240 bp from the initiation codon, and rs4343G>A encoding a silent mutation in exon 16) were suggested to influence A42 levels in the cerebrospinal fluid (CSF) [16]. Both were reported as statistically significantly associated with the risk of developing AD in independent Caucasian population samples [16]. One of these SNPs (rs4343G>A) was also associated with AD in an Israeli Arab community [20]. We previously tested for a possible impact of these two SNPs (rs4291A>T and rs4343G>A) on the risk of AD in a large population-based cohort of 9294 subjects [5], 141 of them developing AD during follow-up. No effect could be set in evidence, probably due to a lack of statistical power. In the present study we examined the I/D polymorphism (rs1799752) together with rs4291 and rs4343 in a European Caucasian population sample including 444 controls and 376 late-onset AD patients for which we previously showed an impact of the ACE D allele on the risk of AD [24]. Patients (35.9% men, mean age ± SD: 75 ± 5 years) recruited in hospitals in France (70%), UK (9%), Netherlands (5%), Spain (9%) and Italy (7%) were thoroughly investigated clinically. The diagnosis of AD was made according to the DSM-III-R and NINCDS-ADRDA criteria. None of the subjects had a family history of AD consistent with Mendelian inheritance. Caucasian controls (35.8% men, mean age ± SD: 74 ± 8 years) were recruited in France and defined as subjects without DSM-III-R dementia criteria and with integrity of their cognitive functions. The study was approved by the Ethics Committee of the Lille University Hospital. Each individual or their relatives signed an informed consent. For each participant, genomic DNA was prepared from white blood cells using a salt-out procedure. The ACE rs1799752 polymorphism was previously genotyped in this population [24], whereas genotyping of SNPs rs4291A>T and rs4343G>A was performed as previously described [5]. The laboratory personnel were blind to clinical information concerning the subjects tested. 2 analysis was used to compare genotype and allele frequencies and to assess deviation from Hardy–Weinberg equilibrium. Statistical analyses were performed with the SAS software release 8.2 (SAS Institute Inc., Cary, NC, USA). A statistically significant interaction with APOE status was systematically looked for. Effects of haplotypes were tested using the THESIAS (Testing Haplotype Effects In Association Studies) software. This program is based on a maximum likelihood model and is linked to the SEM algorithm [30]. Statistical significance was considered at the P < 0.05 level. The ACE SNPs rs4343, rs4291 and rs1799752 genotype distributions in control subjects were in accordance with the Hardy–Weinberg equilibrium (rs4343: 2 = 0.466, 2 df, P = 0.49; rs4291: 2 = 3.550, 2 df, P = 0.06; rs1799752: 2 = 0.001, 2 df, P = 0.97). In AD cases, there were no statistically significant differ-
P
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
Table 1 Haplotype frequencies and ORs of AD risk in the studied population sample.
182
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
183
Table 2 Haplotype frequencies and ORs of AD risk in the studied population sample. Haplotype ACE 4343/4291/1799752
Haplotype frequency
Haplotypic OR [95% CI]
Population-adjusteda haplotypic OR [95% CI], P
Control subjects
Cases
(A) The whole population sample AAI AAD ATI GAD GTD
0.367 0.015 0.079 0.207 0.327
0.425 0.023 0.013 0.178 0.345
1.17 [0.53–2.58] 0.14 [0.06–0.32] 0.72 [0.55–0.96] 0.84 [0.66–1.07]
Reference 1.34 [0.58–3.08], P = 0.49 0.09 [0.04–0.21], P < 10−4 0.67 [0.48–0.92], P = 0.015 0.81 [0.62–1.07], P = 0.13
(B) Subjects aged above 73 years AAI AAD ATI GAD GTD
0.033 0.015 0.177 0.200 0.281
0.402 0.015 0.021 0.185 0.359
0.59 [0.16–2.11] 0.08 [0.03–0.20] 0.70 [0.45–1.10] 0.89 [0.61–1.29]
Reference 0.51 [0.11–2.45], P = 0.40 0.08 [0.03–0.24], P = 10−5 0.63 [0.35–1.13], P = 0.12 1.23 [0.78–1.95], P = 0.38
a
Adjustment for age, sex and APOE status.
ences for ACE SNPs rs4343, rs4291 and rs1799752 genotype and allele distributions between North European and South European subjects (rs4343: 2 = 2.770, 2 df, P = 0.25; rs4291: 2 = 0.012, 2 df, P = 0.99; rs1799752: 2 = 2.004, 2 df, P = 0.37). For the whole population sample, the genotype and allele distributions did not differ according to cognitive status (Table 1). Since age is a well known risk factor for the development of AD, we stratified the population sample on the median age of the AD sample (under or above 73 years old). The oldest patients bearing either the ACE rs4343G or the ACE rs1799752D allele were at reduced risk of AD (Table 1; for rs4343 AA vs AG + GG: OR = 0.49, 95% CI [0.27–0.88], P = 0.02; for rs4291 AA vs AT + TT: OR = 1.20, 95% CI [0.72–2.02], P = 0.48; for ACE rs1799752 II vs ID + DD: OR = 0.43, 95% CI [0.24–0.79], P = 0.007, adjusted for age, sex, APOE status). We then performed a haplotype analysis in the whole population sample, including the three polymorphisms in the following order: rs4343, rs4291 and rs1799752. A highly significant global effect (P = 1.3 × 10−8 after adjustment) of the haplotypes on the risk of AD was observed. The frequency of the ATI combination was lower in cases than in control subjects (OR = 0.09, 95% CI [0.04–0.21], P < 10−4 , adjusted for age, sex, APOE status) (Table 2) opposite to our findings in the 3C cohort. When dividing the population sample according to median age, the protective haplotype effect remains statistically significant only for the oldest (age ≥73 years: for ATI vs AAI OR = 0.08, 95% CI [0.03–0.24], P = 10−5 , adjusted on age, sex, APOE status) (Table 2). A similar effect was also found for the AT haplotype when considering only rs4343 and rs4291 (OR = 0.10, 95% CI [0.03–0.31], P = 6 × 10−5 , adjusted for age, sex, APOE status). The main result of this study is the presence of a statistically significant effect of the common ACE polymorphisms rs4343, rs4291, and rs1799752 on the risk of AD in Caucasian subjects aged 73 years and above. A haplotypic effect could also be set in evidence in this population sample. Although no significant effect on the risk of AD was reported for either ACE rs4343 (6 studies), rs4291 (5 studies), or rs1799752 (30 studies) in various separate studies, meta-analyses in Caucasians (http://www.alzforum.org/res/com/gen/alzgene/default.asp) showed borderline effects on the risk of AD for each of them (rs4343G allele: OR = 0.91, 95% [0.82–1.00]; rs4291T allele: OR = 0.87, 95% CI [0.76–0.99]; rs1799752I allele: OR = 1.05, 95% CI [0.99–1.12]). Although a small, significantly underpowered, GWAS conducted on AD patients with a mean age of 72.2 years did not show any association, either direct or indirect, of rs4343 or rs4291 with AD [10], many recent association studies [8,15,26] and a GWAS [29] are in favour of a role of ACE haplotypes in AD, in accordance with our results.
In conclusion, it seems that ACE play a modest role in the development of AD, at least in the oldest patients. Acknowledgements We thank all the patients and their families. This study was supported by the Institut National de la Santé et de la Recherche Médicale, by the Centre Hospitalier et Universitaire de Lille, and by the Institut Pasteur de Lille. References [1] P. Amouyel, F. Richard, D. Cottel, C. Amant, V. Codron, N. Helbecque, The deletion allele of the angiotensin I converting enzyme gene as a genetic susceptibility factor for cognitive impairment, Neurosci. Lett. 217 (1996) 203–205. [2] G.W. Beecham, E.R. Martin, Y.J. Li, M.A. Slifer, J.R. Gilbert, J.L. Haines, M.A. Pericak-Vance, Genome-wide association study implicates a chromosome 12 risk locus for late-onset Alzheimer disease, Am. J. Hum. Genet. 84 (2009) 35–43. [3] L. Bertram, C. Lange, K. Mullin, M. Parkinson, M. Hsiao, M.F. Hogan, B.M. Schjeide, B. Hooli, J. Divito, I. Ionita, H. Jiang, N. Laird, T. Moscarillo, K.L. Ohlsen, K. Elliott, X. Wang, D. Hu-Lince, M. Ryder, A. Murphy, S.L. Wagner, D. Blacker, K.D. Becker, R.E. Tanzi, Genome-wide association analysis reveals putative Alzheimer’s disease susceptibility loci in addition to APOE, Am. J. Hum. Genet. 83 (2008) 623–632. [4] K. Blennow, M.J. de Leon, H. Zetterberg, Alzheimer’s disease, Lancet 368 (2006) 387–403. [5] A. Bruandet, F. Richard, C. Tzourio, C. Berr, J.F. Dartigues, A. Alperovitch, P. Amouyel, N. Helbecque, Haplotypes across ACE and the risk of Alzheimer’s disease: the three-city study, J. Alzheimers Dis. 13 (2008) 333–339. [6] E.A. Eckman, S.K. Adams, F.J. Troendle, B.A. Stodola, M.A. Kahn, A.H. Fauq, H.D. Xiao, K.E. Bernstein, C.B. Eckman, Regulation of steady-state beta-amyloid levels in the brain by neprilysin and endothelin-converting enzyme but not angiotensin-converting enzyme, J. Biol. Chem. 281 (2006) 30471–30478. [7] E.A. Eckman, C.B. Eckman, Abeta-degrading enzymes: modulators of Alzheimer’s disease pathogenesis and targets for therapeutic intervention, Biochem. Soc. Trans. 33 (2005) 1101–1105. [8] T.L. Edwards, M. Pericak-Vance, J.R. Gilbert, J.L. Haines, E.R. Martin, M.D. Ritchie, An association analysis of Alzheimer disease candidate genes detects an ancestral risk haplotype clade in ACE and putative multilocus association between ACE, A2M, and LRRTM3, Am. J. Med. Genet. B: Neuropsychiatr. Genet., doi:10.1002/ajmg.b.30899. [9] J.S. Elkins, V.C. Douglas, S.C. Johnston, Alzheimer disease risk and genetic variation in ACE: a meta-analysis, Neurology 62 (2004) 363–368. [10] T.M. Feulner, S.M. Laws, P. Friedrich, S. Wagenpfeil, S.H. Wurst, C. Riehle, K.A. Kuhn, M. Krawczak, S. Schreiber, S. Nikolaus, H. Forstl, A. Kurz, M. Riemenschneider, Examination of the current top candidate genes for AD in a genome-wide association study, Mol. Psychiatry (2009), doi:10.1038/mp.2008.141. [11] O. Hanon, F. Forette, Prevention of dementia: lessons from SYST-EUR and PROGRESS, J. Neurol. Sci. 226 (2004) 71–74. [12] M.L. Hemming, D.J. Selkoe, Amyloid beta-protein is degraded by cellular angiotensin-converting enzyme (ACE) and elevated by an ACE inhibitor, J. Biol. Chem. 280 (2005) 37644–37650. [13] M.L. Hemming, D.J. Selkoe, W. Farris, Effects of prolonged angiotensinconverting enzyme inhibitor treatment on amyloid beta-protein metabolism in mouse models of Alzheimer disease, Neurobiol. Dis. 26 (2007) 273–281. [14] D.R. Hou, Y. Wang, L. Zhou, K. Chen, Y. Tian, Z. Song, J. Bao, Q.D. Yang, Altered angiotensin-converting enzyme and its effects on the brain in a rat model of Alzheimer disease, Chin. Med. J. (Engl.) 121 (2008) 2320–2323.
184
N. Helbecque et al. / Neuroscience Letters 461 (2009) 181–184
[15] J.S. Kauwe, J. Wang, K. Mayo, J.C. Morris, A.M. Fagan, D.M. Holtzman, A.M. Goate, Alzheimer’s disease risk variants show association with cerebrospinal fluid amyloid beta, Neurogenetics 10 (2009) 13–17. [16] P.G. Kehoe, H. Katzov, L. Feuk, A.M. Bennet, B. Johansson, B. Wiman, U. de Faire, N.J. Cairns, G.K. Wilcock, A.J. Brookes, K. Blennow, J.A. Prince, Haplotypes extending across ACE are associated with Alzheimer’s disease, Hum. Mol. Genet. 12 (2003) 859–867. [17] D.J. Lehmann, M. Cortina-Borja, D.R. Warden, A.D. Smith, K. Sleegers, J.A. Prince, C.M. van Duijn, P.G. Kehoe, Large meta-analysis establishes the ACE insertion–deletion polymorphism as a marker of Alzheimer’s disease, Am. J. Epidemiol. 162 (2005) 305–317. [18] C.L. Lendon, U. Thaker, J.M. Harris, A.M. McDonagh, J.C. Lambert, M.C. Chartier-Harlin, T. Iwatsubo, S.M. Pickering-Brown, D.M. Mann, The angiotensin 1-converting enzyme insertion (I)/deletion (D) polymorphism does not influence the extent of amyloid or tau pathology in patients with sporadic Alzheimer’s disease, Neurosci. Lett. 328 (2002) 314–318. [19] J.A. Luchsinger, R. Mayeux, Cardiovascular risk factors and Alzheimer’s disease, Curr. Atheroscler. Rep. 6 (2004) 261–266. [20] Y. Meng, C.T. Baldwin, A. Bowirrat, K. Waraska, R. Inzelberg, R.P. Friedland, L.A. Farrer, Association of polymorphisms in the angiotensin-converting enzyme gene with Alzheimer disease in an Israeli Arab community, Am. J. Hum. Genet. 78 (2006) 871–877. [21] S. Miners, E. Ashby, S. Baig, R. Harrison, H. Tayler, E. Speedy, J.A. Prince, S. Love, P.G. Kehoe, Angiotensin-converting enzyme levels and activity in Alzheimer’s disease: differences in brain and CSF ACE and association with ACE1 genotypes, Am. J. Transl. Res. 1 (2009) 163–177. [22] T. Ohrui, T. Matsui, M. Yamaya, H. Arai, S. Ebihara, M. Maruyama, H. Sasaki, Angiotensin-converting enzyme inhibitors and incidence of Alzheimer’s disease in Japan, J. Am. Geriatr. Soc. 52 (2004) 649–650. [23] F. Richard, C. Berr, C. Amant, N. Helbecque, P. Amouyel, A. Alperovitch, Effect of the angiotensin I-converting enzyme I/D polymorphism on cognitive decline. The EVA Study Group, Neurobiol. Aging 21 (2000) 75–80. [24] F. Richard, I. Fromentin-David, F. Ricolfi, P. Ducimetiere, C. Di Menza, P. Amouyel, N. Helbecque, The angiotensin I converting enzyme gene as a susceptibility factor for dementia, Neurology 56 (2001) 1593–1595.
[25] F.A. Sayed-Tabatabaei, B.A. Oostra, A. Isaacs, C.M. van Duijn, J.C. Witteman, ACE polymorphisms, Circ. Res. 98 (2006) 1123–1133. [26] B.M. Schjeide, M.B. McQueen, K. Mullin, J. DiVito, M.F. Hogan, M. Parkinson, B. Hooli, C. Lange, D. Blacker, R.E. Tanzi, L. Bertram, Assessment of Alzheimer’s disease case-control associations using family-based methods, Neurogenetics 10 (2009) 19–25. [27] J.A. Staessen, T. Richart, W.H. Birkenhager, Less atherosclerosis and lower blood pressure for a meaningful life perspective with more brain, Hypertension 49 (2007) 389–400. [28] X. Sun, M. Becker, K. Pankow, E. Krause, M. Ringling, M. Beyermann, B. Maul, T. Walther, W.E. Siems, Catabolic attacks of membrane-bound angiotensinconverting enzyme on the N-terminal part of species-specific amyloid-beta peptides, Eur. J. Pharmacol. 588 (2008) 18–25. [29] T.A. Thornton-Wells, J.H. Moore, E.R. Martin, M.A. Pericak-Vance, J.L. Haines, Confronting complexity in late-onset Alzheimer disease: application of two-stage analysis approach addressing heterogeneity and epistasis, Genet. Epidemiol. 32 (2008) 187–203. [30] D.A. Tregouet, S. Barbaux, S. Escolano, N. Tahri, J.L. Golmard, L. Tiret, F. Cambien, Specific haplotypes of the P-selectin gene are associated with myocardial infarction, Hum. Mol. Genet. 11 (2002) 2015–2023. [31] C. Tzourio, C. Anderson, N. Chapman, M. Woodward, B. Neal, S. MacMahon, J. Chalmers, Effects of blood pressure lowering with perindopril and indapamide therapy on dementia and cognitive decline in patients with cerebrovascular disease, Arch. Intern. Med. 163 (2003) 1069– 1075. [32] J. Wang, L. Ho, L. Chen, Z. Zhao, W. Zhao, X. Qian, N. Humala, I. Seror, S. Bartholomew, C. Rosendorff, G.M. Pasinetti, Valsartan lowers brain betaamyloid protein levels and improves spatial learning in a mouse model of Alzheimer disease, J. Clin. Invest. 117 (2007) 3393–3402. [33] S.C. Waring, R.N. Rosenberg, Genome-wide association studies in Alzheimer disease, Arch. Neurol. 65 (2008) 329–334. [34] K. Zou, H. Yamaguchi, H. Akatsu, T. Sakamoto, M. Ko, K. Mizoguchi, J.S. Gong, W. Yu, T. Yamamoto, K. Kosaka, K. Yanagisawa, M. Michikawa, Angiotensinconverting enzyme converts amyloid beta-protein 1–42 (Abeta(1–42)) to Abeta(1–40), and its inhibition enhances brain Abeta deposition, J. Neurosci. 27 (2007) 8628–8635.