Poster Presentations: P1P176

P1-065 GENETIC VARIANTS ASSOCIATEDWITH

Table 1

Results from m

SNPID

chr1:20757596

chr2:12789281

chr6:47473274

chr7:14311791

chr8:27495342

chr11:5985155

chr11:8567327

chr19:966693

chr19:5182813

chr11:1212674

INCIDENCE OF LATE-ONSETALZHEIMER’S

DISEASE IN CAUCASIANS

Seung Hoan Choi1, Christiane Reitz2, Vincent Chouraki3,

Shubhabrata Mukherjee4, Lei Yu5, Mohammad Ikram6, Joshua Bis7,

Carla Ibrahim-Verbaas8, Ren�ee de Bruijn9, Badri Vardarajan2,

Annette Fitzpatrick7, Oscar Lopez10, Lenore Launer11, Benjamin Grenier-

Bolay12, Eric Larson13, C�eline Bellenguez14, Jean-Charles Lambert15,

Richard Mayeux2, Cornelia Van Duijn8, Paul Crane4, Sudha Seshadri3,

Philippe Amouyel16, Anita Destefano17, David Bennett18, 1Boston

University, Boston, Massachusetts, United States; 2Columbia University,

New York, New York, United States; 3Boston University School of Medicine,

Boston, Massachusetts, United States; 4University of Washington School of

Medicine, Seattle, Washington, United States; 5Rush Alzheimer’s Disease

Center, Chicago, Illinois, United States; 6Erasmus MC, Rotterdam,

Netherlands; 7University of Washington, Seattle, Washington, United

States; 8Erasmus University Medical Center, Rotterdam, Netherlands;9Erasmus MC University Medical Center, Rotterdam, Netherlands;10University of Pittsburgh, Pittsburgh, Pennsylvania, United States;11National Institute on Aging, Bethesda, Maryland, United States;12INSERM U744 Institut Pasteur de Lille, Lille, France; 13Group Health

Research Institute, Seattle, Washington, United States; 14INSERM U744,

Lille, France; 15INSERM U744 Institut Pasteur de Lille, Lille, France;16Inserm / University Lille 2 / Institut Pasteur de Lille, Lille, France;17Boston University School of Public Health, Boston, Massachusetts,

United States; 18Rush Medical Center, Chicago, Illinois, United States.

Contact e-mail: seuchoi@bu.edu

Background: A series of large genome wide association studies identi-

fied several variants that affect susceptibility of Alzheimer’s disease in-

cluding CR1, CLU, PICALM, BIN1, CD2AP, CD33, EPHA1,

MS4A6A/MS4E4 and ABCA7 and more recently additional genes

have been identified via a large mega-meta-analysis by the International

Genomics of Alzheimer Project (IGAP). However, a large part of the

genetic contribution to late-onset Alzheimer’s disease (LOAD) remains

to be explained. The aim of the present study was to identify novel ge-

netic loci associated with incidence of LOAD in Caucasians and for

previously identified genes assess their impact on ’when’ in addition

to ’whether’ one developed AD. Methods: IGAP assembled a dataset

of 13,863 subjects (mean age at baseline75.2166.55, 61.1% female,

1,449 incident AD cases) from 6 independent sites representing studies

in the Alzheimer’s Disease Genetic Consortium (Washington Heights-In-

wood Columbia Aging Project and Adult Changes in Thought studies),

Cohorts for Heart and Aging Research for Genomic Epidemiology

(CHARGE) consortium (Framingham Heart Study, Cardiovascular Health

Study, Rotterdam Study) and European Alzheimer’s Disease initiative (3

City study). Additional cohorts (Religious Orders Study, Rush Memory

and Aging Project) will soon be included as well. Genome-wide associa-

tions with AD incidence were assessed using Cox proportional hazards

models adjusted for baseline age, sex and where appropriate study-site

and familial relationships. Inverse-variance meta-analysis was performed

eta analysis of Cox proportional hazard analysis of AD incidence

Coded/non-coded allele Coded allele frequ

7 T/C 0.9911 (0.0019)

0 T/C 0.4024 (0.0378)

A/G 0.2037 (0.0059)

9 A/G 0.5522 (0.0193)

T/C 0.0365 (0.0017)

1 A/G 0.2514 (0.0987)

0 T/C 0.4427 (0.0244)

T/C 0.8432 (0.1526)

5 T/C 0.741 (0.0883)

85 A/T 0.8974 (0.1258)

for SNPs present in at least 5 studies. Results: In preliminary analyses, no

novel genome-wide significant findings were identified. Among previ-

ously implicated AD genes the most significant associations were ob-

served for CD2AP (hazard ratio (hr) ¼ 0.81, p¼1.99xe-5), followed by

SORL1 (hr¼0.78, p¼.0002) and BIN1 (hr¼1.15, p¼ 0.00048) Conclu-

sions: This preliminary study, which examines the impact of previously

identified risk variants on age at onset of disease observed the most signif-

icant associations for genes involved in endocytosis and intracellular traf-

ficking. While these findings need to be confirmed in the larger meta-

analysis of the complete sample, it is biologically plausible to postulate

that genes modulating intracellular enrichment of potentially pathogenic

proteins are related to onset of AD.

P1-066 ORDERED SUBSETANALYSIS CNVASSOCIATION

in 6 cohorts fo

ency(SE)

WITH ALZHEIMER’S DISEASE AAO

PHENOTYPES

Kinga Szigeti1, Brian Trummer2, Deepika Lal2, Li Yan3, Liu Song3,

Rachelle Doody4, 1University at Buffalo, Buffalo, New York, United States;2University at Buffalo, Buffalo, New York, United States; 3Roswell Park

Cancer Institute, Buffalo, New York, United States; 4Baylor College of

Medicine, Houston, Texas, United States. Contact e-mail: szigeti@buffalo.

edu

Background:Alzheimer’s disease (AD) is a devastating neurodegenerative

disorder affecting approximately 5.4 million individuals in the US and is the

most common cause of dementia in North America and Europe. Genetic fac-

tors play an important role in the pathogenesis of AD and contribute to the

variance of age at onset (AAO). APOE4 has a strong effect on AAO and an

additional 4-5 loci may contribute to the heritability of AAO. AAO is an im-

portant endophenotype and a potential therapeutic target for disease modi-

fying therapy. Ordered subset analysis using AAO as a quantitative

endophenotype may identify genetic risk that contributes to a subset of

AD. This CNVassociation study with AAO of AD was designed to examine

the role of these low-frequency variants with intermediate penetrance in re-

lation to AAO of AD. Methods: 781 AD subjects and 200 normal controls

enrolled in the Texas Alzheimer Research Consortium project were pheno-

typed for AAO and genotyped on the Genome-Wide Human SNPArray 6.0

(Affymetrix) array. The PCA corrected logR data was segmented to reduce

the dataset to probes where any event occurred. Ordered subset analysis for

the numeric array data means was performed by adding sequentially groups

of subjects with the same AAO. Results: The ordered subset analysis using

the univariate segmentated logR data identified 28 probes corresponding to

6 chromosomal regions, where the maximum p-value survivedmultiple test-

ing correction. When using the multivariate segmented logR data, 41 probes

survived multiple testing correction. Several of the gains and losses contrib-

uted to the young AAO subset (<60), but some of the dosage differences ap-

peared to be contributing to the older AAO (>80). The APOE4 allele had its

maximum effect at around AAO 74 consistent with previous reports. Repli-

cation on the NIA-LOAD familial dataset is ongoing.Conclusions: The ge-

netic heterogeneity of AD has been demonstrated in multiple GWAS

studies. Ordered subset analysis usingAAO as a quantitative endophenotype

r known AD genes.

Gene Hazard ratio P-value

CR1 0.590727766 0.01423

BIN1 1.147745978 0.0004753

CD2AP 0.80646079 1.99E-05

EPHA1 0.903210176 0.01072

CLU 1.247323431 0.0166

MS4A6A 1.100428886 0.03096

PICALM 1.096474464 0.01043

ABCA7 0.84324309 0.02493

CD33 0.868228804 0.0012

SORL1 0.781296937 0.0002141

Poster Presentations: P1 P177

generates candidate genes that can be tested in case-control association stud-

ies informed by the AAO where the maximum effect is expected.

P1-067 RNA-SEQUENCING EVALUATION OF FRESH

FROZEN AND FORMALDEHYDE-FIXED

PREFRONTAL CORTEX BRAIN TISSUE FOR THE

STUDY OFALZHEIMER’S DISEASE

Alexandra Dumitriu1, Richard H. Myers2, Ann McKee1, Sudha Seshadri2,

Anita Destefano3, 1Boston University, Boston, Massachusetts, United

States; 2Boston University School of Medicine, Boston, Massachusetts,

United States; 3Boston University School of Public Health, Boston,

Massachusetts, United States. Contact e-mail: adumitri@bu.edu

Background: We performed an RNA-Seq pilot study conducted using

brain tissue from the Framingham Heart Study brain bank. Tissue sam-

ples were obtained from the prefrontal cortex (Brodmann area 9) from

two Framingham brain bank donors: 1) non-demented at death with

a 3/3 APOE genotype and 2) AD diagnosed with a 4/4 APOE genotype.

Methods: Both fresh frozen and fixed tissue were obtained for each of the

two brain bank donors, yielding a total of four samples in the pilot study. The

four available samples were processed using the Ovation RNA-Seq FFPE

system for RNA extraction, followed by the Encore NGS Library System

I for library preparation (NuGEN, CA, USA). Single read 40-nucleotide

RNA-sequencing was performed on the Illumina Genome Analyzer II at

the Boston University Medical Center Sequencing Core. Alignment of the

available reads to the hg19 human genome was performed using TopHat

(v2.0.6) and evaluation of differential gene expression was achieved using

the R package DESeq (v1.8.3). Results: Sequencing quality differed be-

tween the fixed and frozen samples and, therefore, analyses were conducted

separately. Multiple differentially expressed genes were identified between

the AD and control frozen tissue samples. Interestingly, among the top re-

sults for the frozen tissue we detected MS4A6A to have increased expres-

sion in the AD sample; this gene was recently identified to be associated

with risk for AD via meta-analysis of GWAS results. Two genes encoding

for heat shock proteins, HSPA1A and HSPA1B, which have been studied

in association with AD, were also among our top results, showing increased

expression in the diseased sample. Conclusions: RNA-based expression

analysis from fixed tissue is a novel technology and the sequencing data ob-

tained was of lower quality than the ones obtained in frozen tissue. Trim-

ming the obtained sequences is one possible way to salvage the available

data. Our preliminary results suggest that RNA-seq can provide key insights

regarding previously implicated and novel genes and is an area that should

be further investigated for the study of AD.

P1-068 ASSOCIATION BETWEEN VITAMIN D BINDING

PROTEIN (VDBP) HAPLOTYPE AND LATE-ONSET

ALZHEIMER’S DISEASE

Omur Selin Araz1, Erdinc Dursun1, Duygu Gezen-Ak2,

Hasmet Hanagasi3, Basar Bilgic4, Ebba Lohmann4, Turan Ertan5,

Hakan Gurvit4, Engin Eker5, Murat Emre4, Selma Yilmazer1, 1Istanbul

University, Cerrahpasa Faculty of Medicine, Department of Medical

Biology, Istanbul, Turkey; 2Istanbul University, Cerrahpasa Faculty of

Medicine, Department of Medical Biology, Istanbul, Turkey; 3Istanbul

University, Istanbul Faculty of Medicine, Behavioral and Movement

Disorders Unit, Istanbul, Turkey; 4Istanbul University, Istanbul Faculty of

Medicine, Department of Neurology, Behavioral and Movement Disorders

Unit, Istanbul, Turkey; 5Istanbul University, Cerrahpasa Faculty of

Medicine, Department of Geropsychiatry, Istanbul, Turkey.

Contact e-mail: selinaraz@yahoo.com

Background: Alzheimer’s disease (AD) is the most common type of de-

mentia and a progressive neurodegenerative disease. Recent studies have re-

ported that vitamin D (1,25 (OH)2 D 3) has protective effects on the nervous

system and it is suggested to exert its neuroprotective effects by modulating

neuronal calcium homeostasis and production of neurotrophins. A recent

genome wide association study (GWAS) which was designed to determine

the genetic factors related to the vitamin D deficiency indicated a significant

association between 18 polymorphic site of different genes and the plasma

levels of vitamin D. One of these genes was GC (Vitamin D Binding Protein

- VDBP) gene is located chromosome 4p12. The aim of this study is to in-

vestigate rs2282679, rs3755967, rs17467825, rs2298850, rs1155563 poly-

morphisms of GC (Vitamin D Binding Protein - VDBP) gene and to

determine whether these polymorphisms contribute to late onset Alzheim-

er’s disease or not. Methods: Sixty one late onset Alzheimer’s patients

and 60 healthy controls were recruited according to the DSM-IV criteria.

DNA isolation were done with Qiaamp DNA mini kit from blood samples.

We used qRT-PCRmethod and LightSNiP assays in order to determine these

polymorphisms of the GC (Vitamin D Binding Protein - VDBP) gene. Hap-

lotype analysis was performed by "Haploview 4.2" software with chromo-

some position of each SNP taken as reference. Results: When patient and

control groups were compared for the frequency of VDBP gene haplotypes,

"ACGCT" haplotype was found to be significantly higher in the control

groups (p¼ 0.0414). Conclusions: Our study indicates that VDBP gene

haplotype is associated with Alzheimer’s disease. In addition to association

of vitamin D receptor haplotype with AD that was demonstrated in our pre-

vious studies, this particular study also provides additional evidence for the

genetic background of vitamin D related mechanisms in AD.

P1-069 GENETIC ANALYSIS OF KOREANS WITH EARLY-

ONSETALZHEIMER’S DISEASE, UPDATED

Eva Bagyinszky1, Young Ho Park2, Jae-Won Jang3, Seung Chan Kim4,

Young Chul Youn5, Seong An6, SangYun Kim7, 1Gachon University

Department of BioNano Technology, Seongnam-si, South Korea;2Department of Neurology, Seoul National University Bundang Hospital,

Seongnam-si, Gyeonggi-do, South Korea; 3Seoul National University

Bundang Hospital, Sungnam, South Korea; 4Gachon University,

Department of BioNano Technology, Seongnam-si, South Korea;5Department of Neurology, College of Medicine, Chung-Ang University,

Seoul, South Korea; 6Gachon University, Department of BioNano

Technology, Sungnam-si, South Korea; 7Department of Neurology, Seoul

National University Budang Hospital, Seongnam-si, South Korea.

Contact e-mail: navigator120@gmail.com

Background: Alzheimer’s Disease (AD) is the most common form of

senile dementia. AD can also occur in the younger ages below 60 or

65 years, which was defined as a cutoff point between early onset

(EOAD) and late onset AD (LOAD). Three genes are involved in the

EOAD: Amyloid precursor protein (APP), presenilin1 (PSEN1) and pre-

senilin2 (PSEN2). In Korea, only a few publications were reported on

EOAD mutations. In total, 89 dementia patients under 60 years of

age were screened for AD mutations: 45 with AD, 8 with FTD, 1

with PD, 1 with CJD, 15 with MCI, 11 with subjective memory impair-

ment, 3 with vascular cognitive impairment, 1 with transient global am-

nesia, 1 with Hashimoto encephalopathy, and 3 normal person (kin to

the dementia patients). Methods: A PCR based genetic analysis was

performed: we amplified the APP exon 16-17; PSEN1 exon 4-8

and 11; PSEN2 exon 4-7 and 12. After purification, two mutation de-

tection methods were used, single strand conformation polymorphism

(SSCP) and heteroduplex analysis with mismatch-specific nuclease.

To identify the specific mutations, the PCR products were sequenced.

Results:No missense mutation was found in the APP gene, but a silent

polymorphism, Gly708Gly, was detected. This SNP might not have

any pathogenic effect. Two pathogenic missense mutations were de-

tected in the PSEN1 gene: A novel His->Pro exchange at codon163

and a Thr->Ile exchange at codon116. A silent mutation was reported

for Pro218. In the PSEN2, a novel Val->Leu exchange was found at

codon 214 in two individual patients. A silent PSEN2 mutation,

His87His, was very common between the Korean demented patients,

and it appeared in both homo-and heterozygous stages. Conclusions:

In Korea, five pathogenic EOAD mutations were reported. In our find-

ings, we detected five different polymorphisms: three silent SNPs, and

three additional pathogenic mutations. PSEN1 His163Pro was found in

one EOAD patient without family history. Thr116Ile is a known patho-

genic EOAD-associated SNP, and this is the first report in Asia. Val214-

Leu is the first Asian PSEN2 mutation. Since APP, PSEN1 and PSEN2