METHODS OF DIAGNOSING ALZHEIMER'S DISEASE AND ASSOCIATED MARKERS

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application Serial No. 60/933,582, filed June 6, 2007, and U.S. Provisional Application Serial No. 60/925,638, filed April 20, 2007, both of which are hereby incorporated by reference in their entireties.

BACKGROUND

1. Field

[0002] The present disclosure relates to genetic markers, compositions for the detection of genetic markers and methods of diagnosing and screening for Alzheimer's disease (AD). Further encompassed are therapeutic methods and agents for decreasing the deterioration of cells associated with AD and methods of screening such agents.

2. Related Art

[0003] Disorders of the brain are serious medical conditions causing disability and diminished quality of life. Neurological damage is largely irreversible and thus early diagnosis and close monitoring are critical to the successful treatment of patients. Alzheimer's disease (AD) is a neurodegenerative disease associated with progressive memory loss and cognitive dysfunction. It is associated with abnormal clumps (amyloid plaques) and tangled bundles of fibers (neurofibrillary tangles) in the brain which are considered signs of AD. An estimated 4 million Americans have AD. By the year 2030 approximately 1 in every 80 persons in the U.S. will have AD.

[0004] Familial Alzheimer's disease (FAD) is known to be inherited. In affected families, members of at least two generations have had the disease. FAD is rare, accounting for less than 1%

of all cases of AD. FAD has an earlier onset, i.e., about 40 years of age and can be observed to run in families.

[0005] Early -onset Alzheimer's disease (EOAD) is a rare form of Alzheimer's disease in which individuals are diagnosed with the disease before age 65. Less than 10% of all Alzheimer's disease patients have EOAD, Younger individuals who develop Alzheimer's disease exhibit more of the brain abnormalities that are normally associated with Alzheimer's disease. EOAD is usually familial and follows an autosomal dominant inheritance pattern. To date, mutations in three genes including amyloid precursor protein (APP) on chromosome 21, presenilin 1 (PSENl) on chromosome 14 and presenilin 2 (PSEN2) on chromosome 1 have been identified in families with EOAD. Mutations in the APP, PSENl and PSEN2 genes account for about 50% of the disease. Most of the pathogenic mutations in the APP and presenilin genes are associated with abnormal processing of APP, which leads to the overproduction of toxic Aβ-1-42. Down syndrome patients, who have three copies of chromosome 21 which includes the APP gene, begin to develop the characteristic senile plaques and tau tangles at the ages of 30 and 40 (M. Ilyas Kamboh (2004), Molecular Genetics of Late-Onset Alzheimer's Disease, Annals of Human Genetics 68(4):381—404).

[0006] Late-onset Alzheimer's disease (LOAD) is the most common form of Alzheimer's disease, accounting for about 90% of cases and usually occurring after age 65. LOAD strikes almost half of all individuals over the age of 85 and may or may not be hereditary. It is a complex and multifactorial disease with the possible involvement of several genes. Genome-wide linkage or linkage disequilibrium studies on LOAD have provided informative data for the existence of multiple putative genes for AD on several chromosomes, with the strongest evidence on chromosomes 12, 10, 9 and 6. LOAD cases tend to be sporadic, wherein there is no family history of the disease. Genetic susceptibility at multiple genes and interaction between these genes as well as environmental factors are most likely responsible for the etiology of LOAD. Twin data on incident cases indicates that almost 80% of the LOAD risk is attributable to genetic factors. The Apolipoprotein E (APOE) gene on chromosome 19ql3 has been identified as a strong risk factor for LOAD. In fact, the APOE-ε4 allele has been established as a strong susceptibility marker that accounts for nearly 30% of the risk in late-onset AD. More specifically, three variants of APOE, encoded by codons 112 and 158, have been found to modify the risk of LOAD. As compared to the

common AP0E-ε3 allele (codon 112 = Cys and codon 158 = Arg), the AP0E-ε4 allele (codon 112 = Arg and codon 158 = Arg) increases the risk of AD, while the AP0E-ε2 allele (codon 112 = Cys and codon 158 = Cys) decreases the risk of AD. The effect of the APOE-ε4 allele is dose related, wherein one or two copies of the APOE-ε4 allele are associated with 3 -fold or 15-fold risk, respectively. However, the effect of the APOE-ε4 allele on AD risk appears to decline with increasing age (M. Ilyas Kamboh (2004), supra).

[0007] From the time of diagnosis, people with AD survive about half as long as those of similar age without dementia. Medicare costs for beneficiaries with AD were $91 billion in 2005 and may increase to as much as $160 billion in 2010. Finding a treatment that could delay the onset by five years could reduce the number of individuals with AD by nearly 50 percent after 50 years. Drug development for AD is very active and sensitive diagnostic and screening technologies could identify patients for therapy and monitor their response. Improved diagnostic tools for AD would thus be a significant advancement to drug development for this disease and would also provide a means to guide therapeutic decision making thus improving outcomes and reducing unnecessary exposure of patients to costly medications with unwanted side effects.

SUMMARY

[0008] The present disclosure relates to genetic markers and methods of diagnosing and screening for late-onset Alzheimer's disease (LOAD). As such, the disclosure encompasses a whole-genome association analysis of single nucleotide polymorphisms (SNPs) of which a number are located within the GRB2-associated binding protein 2 (GAB2) gene and other genes as set forth in Tables 7A-C and Tables 8A-C. The disclosure identifies two novel haplotypes within the GAB2 gene, i.e., a LOAD risk-enhancing and a LOAD risk-decreasing haplotype as well as other markers disclosed on Tables 7A-C and Tables 8A-C associated with LOAD risk. These haplotypes and some of the markers modify LOAD risk differentially in combination with APOE alleles. Further encompassed are therapeutic methods and agents of decreasing the deterioration of cells associated with LOAD.

[0009] One aspect of the disclosure provides a method of assigning a subject to a late onset Alzheimer's disease (LOAD) risk group wherein the method includes providing a biological sample from the subject, detecting a marker in the biological sample selected from a group comprising: a

marker associated with a haplotype of GAB2 associated with LOAD, a marker in Table 7A, Table 7B, Table 7C, Table 8A, Table 8B, and Table 8C; and assigning the subject to the late onset Alzheimer's disease (LOAD) risk group based upon the presence or absence of the marker. The method involves directly or indirectly detecting the presence or absence of the marker. In one embodiment, the marker is associated with increased risk of LOAD and the subject is assigned to a high LOAD risk group if the subject has the marker. Thus, the presence of this haplotype indicates that the subject is at an increased risk of developing LOAD. As such, the haplotype of GAB2 has the SNP haploblock identified by SEQ ID NO: 2. In another embodiment, the marker is associated with a decreased risk of LOAD and the subject is assigned to a low LOAD risk group if the subject has the marker. Thus, the presence of this haplotype indicates that the subject is at a decreased risk of developing LOAD. As such, the haplotype of GAB 2 has the SNP haploblock identified by SEQ ID NO: 1. In addition, the subject usually carries an apolipoprotein E allele associated with increased LOAD risk regardless of the haplotype. Most often, the subject will carry APOE-ε4, an allele commonly associated with LOAD. In certain instances, the subject will carry the APOC-Ia SNP rs4420638.

[0010] The disclosure contemplates that the marker is detected by a method including but not limited to nucleic acid hybridization, antibody binding, activity assay, polymerase chain reaction (PCR), Sl nuclease assay and via gene chip. For example, the marker can be detected by nucleic acid hybridization using at least one hybridization probe specific to a SNP specific, by way of example, to the SNP haploblock identified by SEQ ID NO: 1 or SEQ ID NO: 2.

[0011] Another aspect of the disclosure provides for an isolated and/or recombinant nucleic acid encoding the amino acid sequence of GAB2-R2, wherein the amino acid sequence codes for a GAB2 protein that is associated with increased risk of LOAD in individuals that carry the APOE-ε4 allele. The corresponding nucleotide sequence is provided by GAB2-R1. The nucleic acid can be provided in combination with an expression vector which can be transformed into a host cell such that the host cell expresses the GAB2 protein that is associated with increased risk of LOAD.

[0012] Another aspect of the disclosure provides for an isolated and/or recombinant nucleic acid encoding the amino acid sequence of GAB2-P2, wherein the amino acid sequence codes for a

GAB2 protein that is associated with decreased risk of LOAD in individuals that carry the APOE-ε4 allele. The corresponding nucleotide sequence is provided by GAB2-P1. The nucleic acid can be provided in combination with an expression vector which can be transformed into a host cell such that the host cell expresses the GAB2 protein that is associated with decreased risk of LOAD. This protein has a protective effect and thus, it decreases the likelihood of developing LOAD.

[0013] The disclosure further encompasses a set of molecular probes for detection, including at least two probes capable of detecting, directly or indirectly, at least two markers disclosed herein associated with increased or decreased risk of LOAD, wherein the molecular probes are not associated with a microarray of greater than 1000 elements. In certain embodiments, at least one of the probes is specific for a marker associated with increased risk of LOAD and at least one other probe is specific for a marker associated with decreased risk of LOAD, Thus, the probes are able to detect directly or indirectly the presence or absence of the marker and are useful for diagnosis, monitoring and risk assessment of LOAD.

[0014] Another aspect of the disclosure provides a method of decreasing neuronal cell deterioration in a subject with dementia {e.g., LOAD), wherein the method includes administering a construct with a nucleic acid of SEQ ID NO: 4 or a fragment thereof to the subject. The nucleic acid codes for GAB2 and the construct includes a recombinant vector in order to transform the construct into cells or tissues of the subject such that the protein can be expressed. In one embodiment, the construct is provided to the subject in form a pharmaceutical composition. Thus, the construct or pharmaceutical composition thereof can be used to treat LOAD. Alternatively, the disclosure provides for a method of decreasing neuronal cell deterioration in a subject with dementia, wherein the method includes administering a polypeptide of SEQ ID NO: 5 or a fragment thereof to the subject. The polypeptide codes for GAB2 which can be administered in form of a pharmaceutical composition, Thus, the protein or pharmaceutical composition thereof can be used to treat LOAD. The subject to be treated for LOAD usually carries an apolipoprotein E allele such as APOE-ε4. In a preferred embodiment, the subject is a human or other mammal.

[0015] The disclosure further includes a method of decreasing or preventing neuronal cell deterioration in a subject at risk for LOAD, wherein the method includes administering a compound

that modulates PI3K or ERK activity. Such a compound can mimic the function of GAB2 which is encoded by the nucleic acid sequence of SEQ ID NO: 4. The corresponding amino acid sequence is encoded by SEQ ID NO: 5.

[0016] The disclosure also encompasses a method of decreasing or preventing neuronal cell deterioration in a subject at risk for LOAD, wherein the method includes administering an siRNA specific for the haplotype of GAB2 with the SNP haploblock identified by SEQ ID NO: 2. Herein, the subject is heterozygous for the haplotype of GAB2 with the SNP haploblock identified by SEQ ID NO: 2. In a preferred embodiment, the subject is a human or other mammal.

[0017] Yet, another aspect of the disclosure provides a method of treating or preventing neuronal cell deterioration in a subject at risk for LOAD, wherein the method includes isolating stem cells from the subject; contacting the stem cells with a nucleic acid encoding GAB2 or a fragment thereof which is not the haplotype of GAB2 with the SNP haploblock identified by SEQ ID NO: 2, wherein the nucleic acid homologously recombines with at least one genomic copy of GAB2 in the stem cells; and contacting the subject with the stem cells to treat or prevent neuronal cell deterioration. Alternatively, the disclosure provides a method of treating or preventing neuronal cell deterioration in a subject at risk for LOAD, wherein the method includes isolating stem cells from the subject; contacting the stem cells with a nucleic acid encoding GAB2 or a fragment thereof which is the haplotype of GAB2 with the SNP haploblock identified by SEQ ID NO: 1, wherein the nucleic acid homologously recombines with at least one genomic copy of GAB2 in the stem cells; and contacting the subject with the stem cells to treat or prevent neuronal cell deterioration. The subject usually carries an apolipoprotein E allele associated with increased LOAD risk such as the APOE- ε4 allele.

[0018] Still, another aspect of the disclosure provides a method for screening for a therapeutic agent effective to inhibit development of cell deterioration associated with dementia, wherein the method includes incubating neuronal cells in vitro with a mixture of PDK and an agent to be tested; measuring PI3K activity; comparing the activity to PI3K activity in the absence of the agent to be tested; and identifying the therapeutic agent by indication of PI3K activity in the presence of the

agent. The therapeutic agent can inhibit the cell deterioration associated with dementia such as LOAD.

[0019] The disclosure further contemplates a non-human transgenic animal with a genome including SEQ ID NO: 1 or SEQ ID NO: 2 in association with a gene for GAB2. The non-human animal can be used as a disease model for screening therapeutic compounds for treatment of LOAD.

[0020] In yet another aspect, the disclosure provides a method for screening for a therapeutic agent effective to inhibit development of cell deterioration associated with dementia, wherein the method includes providing a transgenic animal with a genome including SEQ ID NO: 1 or SEQ ID NO: 2 in association with a gene for GAB2; administering to the transgenic animal an agent to be tested for its effectiveness in treating or preventing the cell deterioration associated with dementia; and assessing the effectiveness of the agent. This method can be used to screen for therapeutic agents that are effective in the treatment and prevention of LOAD.

BRIEF DESCRIPTION OF THE DRAWING FIGURES

[0021] The present disclosure is best understood when read in conjunction with the accompanying figures, which serve to illustrate the preferred embodiments. It is understood, however, that the disclosure is not limited to the specific embodiments disclosed in the Figures.

[0022] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawings will be provided by the Office upon request and payment of the necessary fee.

[0023] Figure 1 depicts the Linkage Disequilibrium (LD) structure, haplotype significance levels (p-values), and Odds Ratios (ORs) 95% CIs for the region encompassing GAB2 on chromosome 1 Iql4.1. LD mapping of the 300 kb surrounding the APOE locus was performed importing genotypes into the HaploView program version 3.33. A strong LD (measured by D' statistic) represents the likelihood that two genetic markers are inherited together. Pair-wise LD values (as measured by D "), reflecting the likehihood that two genetic markers are inherited together, were calculated for each pair of SNPs across the 300 kb interval using the Haploview software and

plotted. Following the standard Haploview color scheme, the gradation bar at the bottom of the figure provides a key for D' versus color. SNP numbers correspond to the following dbSNP ID identification numbers: 1:18579711, 2:18977978, 3:18637149, 4:18977977, 5:18901104, 6:181385600, 7:rsl 1237419, 8:rsl007837, 9:rs2450130, 10:rs2510054, l l :rsl 1237429, 12:rs2510038, 13:rs2511170, 14:rs4945261, 15:rs7101429, 16:rsl0793294, 17:rs4291702, 18:rsl 1602622, 19:rsl0899467, 20:rs2458640, 21:rsl0793302, 22:rs2373115, 23:rsl2280198, 24 :rs 12287010, 25:rsl7136630, 26:rs4945276, 27:rsl996172, 28:rs7395344, 29:rsl 1237522, and 30:rs7950813. In all three cohorts, there was a GAB2 haplotype associated with higher LOAD risk, a haplotype associated with lower risk, and a haplotype unrelated to LOAD risk in APOE ε4 carriers.

[0024] Figure 2 shows the following:

A - Kaplan-Meier plot for carriers of the risk-modifying haplotypes of GAB2.

B - Signaling Cascasde for GAB2: GAB2 is an important scaffolding molecule that mediates signaling from growth factor and cytokine receptors to tau phosphorylation, ApoE synthesis, mast cell degranulation and learning and memory.

[0025] Figure 3 illustrates a map that depicts the various SNPs: PupaSNP was used to identify known SNPs that were likely to play a significant physiological role due to their location in exonic splicing enhancers (ESEs) or triplex forming regions (TRPs), both of which provide for sequence specific regulation of expression and splicing. The four non-synonymous coding SNPs are the most likely candidates for causing functional variance.

[0026] Figure 4A-D depicts GAB2 immunoreactivity in LOAD hippocampus and posterior cingulate cortex using an affinity purified goat polyclonal antibody directed against a C-terminal epitope of GAB2 (Santa Cruz Biotechnology, Santa Cruz, CA). Blocks were obtained from rapid autopsy LOAD cases (< 3 hours postmortem) (N = 5). Hippocampus sections derived from blocks that were fixed for 24 hours in 4% paraformaldehyde and sectioned at 40μm on a freezing microtome, Posterior cingulate sections derived from snap-frozen blocks that were sectioned at 6 μm on a cryostat. Immunohistochemical protocols were as described by Li et al. (Neurobiol Aging (2004) 25:991-999). Immunoreactivity was visualized with nickel-intensified diaminobenzidine.

A) LOAD hippocampus (neutral red counterstain) (4OX objective). The arrow indicates a highly dystrophic cell with the size and morphology of a cortical pyramidal neuron, the cell type selected by laser capture for genomic analyses. Arrowheads point to one of many structures in the sections that resemble dystrophic neurites or neuropil threads.

B) LOAD hippocampus (neutral red counterstain) (40X). The arrow denotes a putative neurofibrillary tangle containing neuron. Arrowheads again indicate a dystrophic neurite. Immunohistochemistry with a rabbit polyclonal antibody directed at GAB2 residues 285-676 (Upstate Cell Signaling Solutions, Lake Placid, NY) gave similar but weaker staining (not shown).

C) AD posterior cingulate gyrus (40X). Filled arrows point to relatively normal appearing, putative neurons. The open arrow points to a cell with the features of a neurofibrillary tangle bearing neuron. Immunoreactive structures clearly resembling pyramidal cell apical dendrites were also observed ascending through the cortical layers (arrowheads).

D) AD posterior cingulate cortex (10OX objective). GAB2 immunoreactive cell with the flame-shaped cytoplasmic inclusion typical of the neurofibrillary tangle.

[0027] Figure 5 shows the results of siRNA treatment. In comparison with vehicle treatment (red, upper left), in GAB2 siRNA treatment there was a 1.70-fold increase in tau phosphorylation at the Serene-262 residue (red, lower left), which is phosphorylated in LOAD neurofibrillary tangle- bearing neurons. This fold-change was not attributable to an increase in total tau (upper and lower right, green),

[0028] Figure 6 shows gene expression data, wherein the raw expression level is shown on the y- axis and the haplotype is shown on the x-axis (right side has gene expression in samples from LOAD patients homozygous for the risk increasing GAB2 haplotype (RR) and left side has gene expression in samples from LOAD patients not homozygous for the risk increasing GAB2 haplotype (RP - heterozygous for the risk increasing and the risk decreasing (P) GAB2 haplotype; NN - homozygous for the risk non-altering GAB2 haplotype).

[0029] Figure 7 is a table providing additional details on each SNPs 5-22 in each of the GAB2 haplotypes including location and strand.

[0030] Figure 8 is a table providing LOAD-association significance levels and additional data of 10 SNPs located in the GRB-associated binding protein 2 (GAB2 ) gene on chromosome 11 q 14.1.

BRIEF DESCRIPTION OF THE SEQUENCES

[0031] The present disclosure encompasses the following sequences with assigned sequence identifiers:

[0032] SEQ ID NO: 1 - TC-GCA-TGAGGTGTCTT - Haploblock of SNPs that identify the haplotype of GAB2 that decreases LOAD risk.

[0033] SEQ ID NO: 2 - CT-AAG-CAGATCAGACG - Haploblock of SNPs that identify the Haplotype of GAB2 that increases LOAD risk.

[0034] SEQ ID NO: 3 - CT- AAG-C AG AGC AGCCG - Haploblock of SNPs that identify the Haplotype of GAB2 that does not alter risk for LOAD.

[0035] SEQ ID NO: 4 - cDNA for GAB2 (from a fourth haplotype).

[0036] SEQ ID NO: 5 - protein sequence for GAB2 (from a fourth haplotype).

[0037] GAB2-R1 - cDNA sequence for LOAD risk increasing allele of human GAB2.

[0038] GAB2-R2 - polypeptide sequence for LOAD risk increasing allele of human GAB2.

[0039] GAB2-P1 - cDNA sequence for LOAD risk decreasing (protective) allele of human GAB2.

[0040] GAB2-P2 - polypeptide sequence for LOAD risk decreasing (protective) allele of human GAB2.

[0041] GAB2-R1 -Genomic - genomic sequence for LOAD risk increasing allele of human GAB2.

[0042] GAB2 -Pl -Genomic - genomic sequence for LOAD risk decreasing (protective) allele of human GAB2.

[0043] GAB2-Neutral-Genomic - genomic sequence for non-LOAD risk altering allele of human GAB2.

DETAILED DESCRIPTION

i.) General Overview

[0044] Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by memory and cognitive impairments and other non-cognitive behavioral symptoms. Age is the strongest risk factor, wherein almost 50% of people over the age of 85 are affected. Early-onset AD (EOAD) is associated with genetic mutations in amyloid precursor protein (APP), presenilin 1 (PSENl) and presenilin 2 (PSEN2). However, sporadic or late-onset AD (LOAD) is multi-factorial and genetically more complex. In addition, genetic factors may account for as much as 80% of the disease risk associated with LOAD (Gatz et al. (2006) Arch. Gen. Psychiatry 63(2): 168-174). While monogenic mutations cause EOAD, the only extensively validated susceptibility gene for LOAD is the apolipoprotein E (APOE-ε4) allele (Saunders et al. (1993) Neurology 43(8): 1467-1472 and Farrer et al. (1997) JAMA 278(16):1349-1356). But alleles of the APOE gene do not account for all of the genetic load responsible for LOAD predisposition. Stratification by APOE-ε4 carrier status allows for the detection of association signals that are normally overwhelmed and thus, masked by the signal of APOE alleles in a non-stratified study design. Specifically, it allowed the present inventors to detect a locus that modifies the LOAD risk of individuals who carry the APOE- ε4 allele.

[0045] Late-onset Alzheimer's Disease (LOAD) is the most common age-related dementia, and is caused by neuronal death of poorly understood etiology. The most accurate method of diagnosis is the presence of characteristic plaques and tangles in the brain which can be revealed through

histological examination. The apolipoprotein E (APOE-ε4) allele (supra) accounts for approximately 65% of the total risk. The disclosure encompasses a new whole-genome association analysis of single nucleotide polymorphisms (SNPs) using 502,627 SNPs in 760 histopathologically verified Alzheimer's disease (AD) cases and controls. Of the 25 SNPs with the most significant p- values in the APOE-ε4 carrier group, 10 SNPs are found to be located within the GRB2-associated binding protein 2 (GAB2) gene. There is a single haplotype block encompassing the entire gene and which contains four potentially functional non-synonymous nucleotide changes. The disclosure identifies a risk-enhancing and a risk-decreasing haplotype within the GAB2 gene. These haplotypes modify risk differentially in combination with APOE alleles. In addition, the present disclosure includes additional markers associated with increased or decreased risk of LOAD as shown in Tables 7A-C and 8A-C.

H.) Definitions

[0046] The following definitions are set forth to illustrate and define the meaning and scope of the various terms used to describe the present disclosure.

[0047] The term "single polynucleotide polymorphism (SNP)" refers to a DNA sequence variation that involves a change in a single nucleotide. They present in humans with a frequency of about once in every 1000 bases and contribute to differences among individuals. The majority of SNPs have no effect. However, some affect the risk for certain diseases.

[0048] The term "haplotype" means one or more closely linked alleles, i.e., genes or DNA polymorphisms (e.g., SNPs) inherited as a unit. Different combinations of polymorphisms are known as haplotypes. The difference of a single genetic marker can delineate a distinct haplotype. Alternatively, the results from several loci could be referred to as a haplotype. For example, a haplotype can be a set of SNPs on a single chromatid that are statistically associated (i.e., inherited as a unit). It is thought that these associations, and the identification of a few alleles of a "haplotype block", can help to identify all other polymorphic sites in its region. With the identification of a haplotype block associated with a particular haplotype, one of skill in the art may readily identify all other DNA polymorphisms associated with the particular haplotype by routine sequencing of the

genomic DNA of an individual having such haplotype, preferably homozygous for such haplotype. Thus, such information is valuable for investigating the genetics behind common diseases.

[0049] A "SNP haploblock identified by SEQ ID NO:" means set of at least two SNPs that are associated with an allele of a gene such as GAB2, wherein said SNPs are grouped together in form of a synthetic nucleotide sequence. The SNPs in a given haploblock may be associated with a risk increasing or risk decreasing haplotype for LOAD. Thus, when a nucleic acid is "specific to a SNP haploblock" or "specific to a SNP within a haploblock" then that nucleic acid (e.g., nucleic acid probe) is complementary to at least one SNP that is grouped within that haploblock, which may include one of the SNPs that are associated with the synthetic nucleotide sequence or any other SNP associated with the haplotype identified by the SNP haploblock. A "SNP that is specific to a SNP haploblock" is a SNP where the DNA polymorphism is found only in the particular haplotype and no other. By way of example, considering the three GAB2 haplotypes identified by the haploblocks in Table 1 below, SNP 5 is specific to the SNP haploblock identified by SEQ ID NO: 1 since the SNP haploblock identified by SEQ ID NO: 1 is the risk decreasing haplotype and this haplotype has a T at SNP 5 while the other two haplotypes have a C.

[0050] The term "HapMap" refers to a catalog of common genetic variants that occur in human beings. It describes what these variants are, where they occur in the DNA, and how they are distributed among individuals within populations and among populations in different parts of the world (A haplotype map of the human genome (2005) Nature 437:1299-1320).

Ui.) Haplotypes ofGAB2 as Genetic Markers for AD

[0051] The genetic sequences of different individuals are remarkably similar. When the chromosomes of two humans are compared, their DNA sequences can be identical for hundreds of bases. But at about one in every 1000 to 1,200 bases, on average, the sequences will differ. As such, one individual might have an A at that location, while another individual has a G, or a person might have extra bases at a given location or a missing segment of DNA. Differences in individual bases are the most common type of genetic variation. These genetic differences are known as single nucleotide polymorphisms (SNPs) (supra). SNPs act as markers to locate genes in DNA. Given the relatively close spacing between these SNPs, SNPs are typically inherited in blocks.

[0052] GAB2 is scaffolding protein involved in multiple signaling pathways that impact AD pathology (see Figure 2). GAB2 acts downstream of growth factor and cytokine receptors and is the primary activator of the phosphatidylrnositol 3 -kinase (PI3K) signaling pathway. PI3K activates Akt, which in turn inhibits Gsk3, which has been shown to inhibit phosphorylation of Tau. Activated Akt also inhibits Bad and the FOXO family proteins, both of which cause neuron death. Inhibition of PI3K signaling decreases the efficacy of the Acetylcholinesterase inhibitors donepezil and galanthamine (i.e., treatments for AD) which act, at least in part, through a pathway whose main activator is GAB2. Furthermore, Gab2 associates with Grb2 and Sos which then activate the Erk 1/2 signaling cascade. The Erk signaling cascade influences learning and memory through Creb. In addition, GAB2 plays a role in mast cell degranualtion through FCεRI signaling. GAB2 complexes with Fyn to initiate a degranulation cascade through Pkc. For example, GAB2 knockout mice are viable, but show profound mast cell deficiencies and reduced degranulation.

[0053] Despite the postulation of many genes as possible susceptibility factors in LOAD, the only well-established genetic link is the APOE-ε4 allele. Association studies traditionally fail to replicate due to underlying genetic structure, low sample number, poor genomic coverage and incorrect diagnosis. However, the inventors have addressed these issues by controlling for underlying genetic structure and interrogating 502,627 SNPs in 1469 samples of which 1086 samples are histopathologically confirmed based on the presence or absence of plaques and tangles in the postmortem samples. The inventors have further enhanced the analysis of this complex disorder by stratifying the data for APOE-ε4 carrier status in order to detect signals that would normally be masked in a non-stratified design. Thus, the disclosure identifies two GAB2 haplotype blocks wherein one is associated with increased risk for LOAD (SEQ ID NO: 1) and the other is associated with decreased risk for LOAD (SEQ ID NO: 2), wherein the latter exhibits a strong protective effect. These haplotypes modify risk differentially in combination with APOE alleles.

[0054] There is a single haplotype block encompassing the entire GAB2 gene and which contains four potentially functional non-synonymous nucleotide changes. The haplotype block is shown in Figure 1. SNPS 5-22 are depicted horizontally within the haploblock as shown in Table 1 below.

* Does not increase or decrease risk.

As shown in Table 1, SNPS 7 and 11 were insignificant while all others are considered significant. The four potentially functional non-synonymous nucleotide changes are believed to be among the significant SNPS within this haplotype. As depicted, the risk-increasing and risk-decreasing haplotype are within the GAB2 gene. These haplotypes modify risk differentially in combination with APOE alleles.

[0055] Determining the functional variant in the GAB2 haplotype block is challenging because the block spans 189kb and includes at least 614 known SNPs, each of which reflects a significant association with AD due to the significant haplotype block. Of the 614 described SNPs, 4 are non- synonymous and are the best candidates for causing the disease risk-modifying effects of the haplotype. It is unlikely that a complex, partially penetrant disease like LOAD would have a truncated protein or other catastrophic risk or protective allele. However, it is likely that subtle genetic differences lead to altered protein expression or phosphorylation. PupaSNP is used to identify known SNPs that are likely to play a significant physiological role due to their location in exonic splicing enhancers (ESEs) or triplex forming regions (TRPs), both of which provide for sequence specific regulation of expression and splicing (see Figure 3). The present disclosure encompasses a high density whole genome association study of LOAD, utilizing a large set of highly accurate clinically and pathologically characterized samples. Notably, replication of six SNPs in three independent study populations reveal a haplotype block encompassing GAB2 on chromosome 11 ql4.1 that is protective against LOAD in APOE-ε4 carriers. This finding is significant since GAB2 is integral to pathways affecting Tau phosphorylation, Creb activation, mast cell degranulation, and APOE protein synthesis.

[0056] An array platform (e.g., AFFYMETRIX 500K array) can be used to genotype all three populations that are utilized. However, any other platform can be employed as well. There are two reasons for confidence that the genotype calls are correct and that the association effect observed is indeed accurate (and not due to a bias of any single platform used). First, the observation is not a single SNP association, but rather the association of a large haplotype block which is in agreement with the LD structure of the HapMap CEPH population. Second, one benefit of using SNiPer-HD for genotype calls is the generation of a quality score measuring how well the clustering algorithm is able to separate the chip output into three distinct Gaussians. All six SNPs showing replication in the region have high quality scores (>0.45), indicating that the data for each SNP clustered into three distinct Gaussians corresponding to the three possible genotype calls.

[0057] The whole genome association study uses an evenly distributed panel of SNPs across the genome with an average inter-marker distance of 6 kb in a sample of 1469 well characterized AD cases and age-matched unaffected controls. As such, it identified risk-modifying haplotypes of GAB2 for APOE-ε4 carriers. Ultra-high density SNP genotyping is successfully employed to identify risk and protective loci in complex diseases because it is rigorously controlled and performed using a sufficiently reliable and dense set of markers. Further efforts are in progress to identify the functional variant(s) present in GAB2 in order to include pre-symptomatic risk assessment strategies and actual drug targets.

[0058] Several additional markers were identified in or near other genes that have utility in diagnosing and monitoring late onset Alzheimer's disease which may be used alone or in combination with the markers disclosed herein or otherwise known in the art such as the APOE-ε4 allele.

iv.) Methods of Diagnosing AD

[0059] The disclosure provides a method of assigning a subject to a late onset Alzheimer's disease (LOAD) risk group in order to assess the likelihood of the subject being afflicted with the disease. This method can be employed to assess the risk at early stages of disease progression. The method includes providing a biological sample from the subject, detecting a marker in a biological sample, which can be a haplotype of GAB2 associated with LOAD or any of the markers on Tables

7A-C and Tables 8A-C, and assigning the subject to the late onset Alzheimer's disease (LOAD) risk group based upon the presence or absence of the haplotype. The method involves directly or indirectly detecting the presence or absence of the markers. For example, the haplotype of GAB2 may be associated with increased risk of LOAD and the subject is then assigned to a high LOAD risk group if the subject has the haplotype of GAB2. As an example of such is the haplotype of GAB2 identified by the SNP haploblock identified by SEQ ID NO: 2. Thus, the presence of haploblock identified by SEQ ID NO: 2 indicates that the subject is at increased risk for developing the disease. Alternatively, the marker may be associated with decreased risk of LOAD and the subject is assigned to a low LOAD risk group if the subject has the marker. An example of such is the haplotype of GAB2 identified by the SNP haploblock identified by SEQ ID NO: 1. Thus, the presence of haploblock identified by SEQ ID NO: 1 indicates that the subject is at a decreased risk for developing the disease. In addition, the subject may be further stratified by LOAD risk group based upon whether the subject carries an apolipoprotein E allele associated with increased or altered LOAD risk. By way of example, APOE-ε4, is commonly associated with LOAD. In addition, various haplotypes of APOE have been associated with LOAD risk groups as set forth in U.S. Patent Publ. 2005/0277129. Finally multiple markers disclosed herein may be used in combination to improve the accuracy, preferably two or more, three or more, four or more, five or more, or ten or more of the markers may be used.

[0060] The disclosure contemplates that the markers may be detected by a variety of methodologies or procedures that are well know in the art including, but not limited to, nucleic acid hybridization, antibody binding, activity assay, polymerase chain reaction (PCR), Sl nuclease assay and via gene chip or microarray as well as any other assay known in the art that may be used to detect the SNPs associated with a haplotype or the gene product produced from the gene of the haplotype including mRNA and protein. For example, the haplotype of GAB2 can be detected by nucleic acid hybridization using at least one hybridization probe specific to a SNP specific to the SNP haploblock identified by SEQ ID NO: 1 or SEQ ID NO: 2 (see Figure 1 for SNPS 5-22). Hybridization of a SNP-specific oligonucleotide to a target polynucleotide may be performed with both entities in solution, or such hybridization may be performed when either the oligonucleotide or the target polynucleotide is covalently or noncovalently affixed to a solid support. Attachment may

be mediated, for example, by antibody-antigen interactions, poly-L-Lys, streptavidin or avidin- biotin, salt bridges, hydrophobic interactions, chemical linkages, UV cross-linking baking, etc. SNP-specific oligonucleotides may be synthesized directly on the solid support or attached to the solid support subsequent to synthesis. Solid-supports suitable for use in detection methods of the disclosure include substrates made of silicon, glass, plastic, paper and the like, which may be formed, for example, into wells (as in 96-well plates), slides, sheets, membranes, fibers, chips, dishes, and beads. The solid support may be treated, coated or derivatized to facilitate the immobilization of the SNP-specific oligonucleotide or target nucleic acid. Detecting the nucleotide or nucleotide pair of interest may also be determined using a mismatch detection technique, including but not limited to the RNase protection method using riboprobes (Winter et al (1985) Proc. Natl. Acad. ScL USA 82:7575; Meyers et al. (1985) Science 230:1242) and proteins which recognize nucleotide mismatches, such as the E. coli mutS protein (Modrich (1991) Ann. Rev. Genet. 25:229-53). Alternatively, variant SNPs or variant alleles can be identified by single strand conformation polymorphism (SSCP) analysis (Orita et al (1989) Genomics 5:874-9); Humphries et al (1996) in MOLECULAR DIAGNOSIS OF GENETIC DISEASES, Elles, ed., pp. 321-340) or denaturing gradient gel electrophoresis (DGGE) (Wartell et al. (1990) Nucl. Acids Res. 18:2699- 706); Sheffield et al. (1989) Proc. Natl. Acad. ScL USA 86:232-6). A polymerase-mediated primer extension method may also be used to identify the polymorphism(s). Several such methods have been described in the patent and scientific literature and include the "Genetic Bit Analysis" method (WO 92/15712) and the ligase/polymerase mediated genetic bit analysis (U.S. Pat. No. 5,679,524. Related methods are disclosed in WO 91/02087, WO 90/09455, WO 95/17676, and U.S. Pat. Nos. 5,302,509 and 5,945,283. Extended primers containing the complement of the polymorphism may be detected by mass spectrometry as described in U.S. Pat. No. 5,605,798. Another primer extension method is allele-specific PCR (Ruano et al. (1989) Nucl. Acids Res. 17:8392; Ruano et al. (1991) Nucl Acids Res. 19:6877-82); WO 93/22456; Turki et al. (1995) J. CHn. Invest. 95:1635-41). The haplotype for a gene of an individual may also be determined by hybridization of a nucleic acid sample containing one or both copies of the gene, mRNA, cDNA or fragment(s) thereof, to nucleic acid arrays and sub-arrays such as described in WO 95/112995. The arrays would contain a battery of SNP-specific or allele-specific oligonucleotides representing each of the polymorphic sites to be included in the haplotype.

[0061] The step of detecting the presence or absence of a marker disclosed herein or a close isoform thereof may be carried out either directly or indirectly by any suitable means. A variety of techniques are known to those skilled in the art (supra). All generally involve the step of collecting a biological sample containing DNA or protein from the subject, and then detecting whether or not the subject possesses the marker or a close isoform thereof. For example, the detecting step may be carried out by collecting a biological sample from the subject (any fluid or tissue containing the markers associated with the haplotype), and then determining the presence or absence of the haplotype in the sample {e.g., by nucleic acid hybridization, immunoassay such as antibody binding, activity assay, polymerase chain reaction (PCR), Sl nuclease assay, via gene chip or microarray, via isoelectric focusing, etc.). Other markers may also be used that are associated with the markers disclosed herein such as SNPs or other polymorphic markers that are in close enough proximity to have a statistically significant association with the marker disclosed herein (i.e., other markers in linkage disequilibrium with a marker disclosed herein). It will be readily appreciated by those of skill in the art that the detecting steps described herein may be carried out directly or indirectly. Thus, for example, if a marker or a close isoform thereof is detected in the subject, then it is determined that the subject is either at increased or decreased risk for LOAD depending on which marker or close isoform thereof is identified {i.e., a significant enough number of markers associated with a haplotype). If, for example, the subject is identified to carry the risk increasing GAB2 haplotype in combination with the APOE-ε4 allele then the subject is very likely to be at increased risk of developing LOAD. If, for example, the subject is identified to carry the risk decreasing GAB2 haplotype in combination with the APOE-ε4 allele then the subject is very likely not at increased risk of developing LOAD or even at a lower risk of developing LOAD compared to APOE-ε4 carriers that do not carry any GAB2 haplotype associated with either SEQ ID NO: 1 or SEQ ID NO: 2. Other means of indirectly determining a risk for LOAD could be by measuring other polymorphic markers that are in statistically significant proximity to any of the SNPs of the GAB2 haplotypes.

[0062] The disclosure also provides set of molecular probes for detection, including at least two probes capable of detecting, directly or indirectly, a marker disclosed herein associated with increased or decreased risk of LOAD, wherein the molecular probes are not associated with a

microarray of greater than 1000 elements, a microarray with greater than 500 elements, a microarray with greater than 100 elements a microarray with greater than 50 elements, or are not associated with a microarray. In preferred embodiments, at least one of the probes is capable of detecting, directly or indirectly, a marker disclosed herein associated with increased risk of LOAD and at least one other probe is capable of detecting, directly or indirectly, a marker disclosed herein associated with decreased risk of LOAD. Thus, the probes are able to detect directly or indirectly the presence or absence of such marker disclosed herein.

[0063] Detecting the presence or absence of a marker disclosed herein or a close isoform thereof may be carried out either directly or indirectly. For example, SNP testing technologies can be employed to detect the haplotype of GAB2. Based upon a patient's specific genotype (SNP pattern) an estimate of the patient's relative risk for LOAD could be provided. This would include genotyping first for ApoE 2/3/4 status, and then further genotyping for the GAB2 haplotype status. This includes Odds Ratio estimates based on carrier status of these two genes.

[0064] Detection of the disease also includes detection of the haplotype by any SNPs/markers within the haplotype, but also indirectly through SNPs/markers outside the haplotype and leveraging linkage disequilibrium to identify carriers of the haplotype.

[0065] In addition to determining a patient's relative risk for LOAD, the diagnosis may include prescribing therapeutic regimens to treat, prevent or delay onset of LOAD.

[0066] In certain preferred embodiments, the method of diagnosis will further include direct or indirect detection of APOE alleles associated with LOAD, preferably the APOE-ε4 allele. Such detection may be performed using any of the detection methods available to one of skill in the art and the markers disclosed herein and APOE alleles may be detected using the same or different methods and may be detected at the same or different times. Further, the method of diagnosis may rely upon the information regarding the APOE alleles of the subject that had been previously determined. With information regarding the marker disclosed herein and APOE alleles of a subject, the diagnosis of risk may be determined and present in the form of Odds Ratio estimates for the set of alleles the subject has.

v.) Screening Methods

[0067] The disclosure also provides a method for screening for a therapeutic agent effective to inhibit development of cell deterioration associated with dementia. This method includes incubating neuronal cells in vitro with an agent to be tested, measuring PI3K activity, comparing the activity to PI3K activity in the absence of the agent to be tested, and identifying the therapeutic agent by indication of PI3K. activity in the presence of the agent. The therapeutic agent can inhibit the cell deterioration associated with dementia such as LOAD. These methods also includes incubating neuronal cells in vitro with an agent to be tested, measuring GAB2 activity directly or indirectly, comparing the activity to GAB2 activity in the absence of the agent to be tested, and identifying the therapeutic agent by detection of modulation of GAB2 activity in the presence of the agent. Examples of measuring GAB2 activity include measuring mRNA level, mRNA transcription rate, protein level, assaying the activity of the GAB2 protein or one of the proteins that it regulates such as PI3K, and measuring activity of a reporter gene operably linked to a GAB2 regulatory element which can include the GAB2 promoter and/or one or more GAB2 enhancer or repressor elements. The therapeutic agent can inhibit the cell deterioration associated with dementia such as LOAD. Thus, the therapeutic agent will act as an agonist to activate PI3K. For the purpose of screening for such agents, the disclosure also includes a non-human transgenic animal or mammalian cell based assay (preferably a human cell based assay) with a genome including the human GAB2 gene associated with the haplotype identified by the SNP haploblock of SEQ ID NO: 1 or SEQ ID NO: 2. The non-human animal or mammalian cell based assay can be used as a disease model for screening therapeutic compounds for treatment of LOAD. Thus, the disclosure provides a method for screening for a therapeutic agent effective to inhibit development of cell deterioration associated with dementia, wherein the method includes providing a transgenic animal or mammalian cell based assay with a genome including the human GAB2 gene associated with the haplotype identified by the SNP haploblock of SEQ ID NO: 1 or SEQ ID NO: 2. The method further includes administering to the transgenic animal an agent to be tested for its effectiveness in treating or preventing the cell deterioration associated with dementia, and assessing the effectiveness of the agent.

vi.) Therapeutic Methods and Formulations

[0068] In people with AD, changes in the brain may begin 10 to 20 years before any visible signs or symptoms appear. Some regions of the brain may begin to shrink, resulting in memory loss, the first visible sign of AD. Over time, AD progresses through three main stages: mild, moderate, and severe. These stages are characterized by a collection of signs and symptoms and behaviors that individuals with AD experience. People with mild symptoms of AD often seem healthy, but they are actually having difficulty making sense of the world around them. Initial symptoms are often confused with changes that take place in normal aging. Symptoms and early signs of AD may include difficulty learning and remembering new information, difficulty managing finances planning meals, taking medication on schedule, depression symptoms (sadness, decreased interest in usual activities, loss of energy), getting lost in familiar places, etc. In moderate AD, the damaging processes occurring in the brain worsen and spread to other areas that control language, reasoning, sensory processing, and thought. In this stage, symptoms and signs of AD become more pronounced and behavioral problems may become more obvious. Signs and symptoms of moderate AD may include forgetting old facts, repeating stories and/or questions over and over, making up stories to fill gaps, difficulty performing tasks, following written notes, agitation, restlessness, repetitive movements, wandering, paranoia, delusions, hallucinations, deficits in intellect and reasoning, lack of concern for appearance, hygiene, and sleep, etc. In the advanced stage of AD, damage to the brain's nerve cells is widespread. At this point, full-time care is typically required. People with severe AD may have difficulty walking, and they often suffer complications from other illnesses, such as pneumonia. Signs of severe AD may include screaming, mumbling, speaking gibberish, refusing to eat, failing to recognize family or faces, and difficulty with all essential activities of daily living.

[0069] The disclosure provides a method of decreasing or preventing neuronal cell deterioration in a subject at risk for LOAD, wherein the method includes administering a compound that mimics the function of GAB2. By way of example, such a compound may modulate PBK or ERK activity. Such a compound is provided by the instant disclosure and is encoded by the amino acid sequence of SEQ ID NO: 5. The corresponding nucleic acid sequence is encoded by SEQ ID NO: 4. Additional compounds and agents may be readily identified by the screening methods discussed above. Such modulating may include, for example, turning on PI3K activity which will in turn

activate the protein AKT which will lead to inhibition of TAU which is associated with increased neurofibrillary tangles and plaques in the brain that are characteristic of LOAD, and to inhibition of the FOXO proteins and BAD protein which increase cell death. Since PI3K is a known cancer- related gene, many compounds that modulate PI3K activity have been identified and either may be used directly in the methods herein or may be used to design libraries for further screening of PI3K modulators. Examples of such PI3K modulators may be found in U.S. Patent Application Nos: 2006005832, 20060128732, 20060106038, and 20020151549. Such modulating may also include, for example, turning on the ERK 1 and 2 cascades will lead to activated CREB which enhances learning and memory. Thus, a compound that mimics GAB2 can be used as a therapeutic agent or agonist that is expected to decrease or prevent neuronal cell deterioration associated with LOAD.

[0070] GAB2 agonists, mimics, and modulatory agents (e.g., drugs based on SEQ ID NO: 5) are formulated as pharmaceuticals to be used in the methods of the disclosure such as for the treatment and prevention of AD, particularly LOAD. Any composition or compound that can stimulate a biological response associated with the binding of a ligand analogue (i.e., an agonist or modulator) to GAB2 or that can mimic GAB2 can be used as a pharmaceutical in the disclosure. General details on techniques for formulation and administration are well described in the scientific literature (see "Remington's Pharmaceutical Sciences", Maack Publishing Co, Easton Pa.). GAB2 agonist, mimic, and modulatory agent pharmaceutical formulations can be prepared according to any method known in the art for the manufacture of pharmaceuticals. The GAB2 agonists, mimics, and modulatory agents used in the methods of the disclosure can be formulated for administration in any conventionally acceptable way including via intravenous injection, IM, IP, orally, topically, vaginally or rectally. Oral administration is preferred. Illustrative examples are set forth below.

[0071] Pharmaceutical formulations for oral administration can be formulated using pharmaceutically acceptable carriers well known in the art in dosages suitable for oral administration. Such carriers enable the pharmaceutical formulations to be formulated in unit dosage forms as tablets, pills, powder, capsules, liquids, lozenges, gels, syrups, slurries, suspensions, etc., suitable for ingestion by the patient. Pharmaceutical preparations for oral use can be obtained through combination of GAB2 agonist compounds with a solid excipient, optionally grinding a resulting mixture, and processing the mixture of granules, after adding suitable additional

compounds, if desired, to obtain tablets or pills. Suitable solid excipients are carbohydrate or protein fillers which include, but are not limited to, sugars, including lactose, sucrose, mannitol, or sorbitol; starch from corn, wheat, rice, potato, or other plants; cellulose such as methyl cellulose, hydroxypropylmethyl-cellulose, or sodium carboxymethylcellulose; and gums including arabic and tragacanth; as well as proteins such as gelatin and collagen. If desired, disintegrating or solubilizing agents may be added, such as the cross-linked polyvinyl pyrrolidone, agar, alginic acid, or a salt thereof, such as sodium alginate. Pharmaceutical preparations of the disclosure that can also be used orally are, for example, push-fit capsules made of gelatin, as well as soft, sealed capsules made of gelatin and a coating such as glycerol or sorbitol. Push-fit capsules can contain GAB2 agonists, mimics, and modulatory agents mixed with a filler or binders such as lactose or starches, lubricants such as talc or magnesium stearate, and, optionally, stabilizers. In soft capsules, the GAB2 agonist, mimic, and modulatory agent compounds may be dissolved or suspended in suitable liquids, such as fatty oils, liquid paraffin, or liquid polyethylene glycol with or without stabilizers,

[0072] Aqueous suspensions of the disclosure contain a GAB2 agonist, mimic, and modulatory agent in admixture with excipients suitable for the manufacture of aqueous suspensions. Such excipients include a suspending agent, such as sodium carboxymethylcellulose, methylcellulose, hydroxypropylnethylcellulose, sodium alginate, polyvinylpyrrolidone, gum tragacanth and gum acacia, and dispersing or wetting agents such as a naturally occurring phosphatide (e.g., lecithin), a condensation product of an alkylene oxide with a fatty acid (e.g., polyoxyethylene stearate), a condensation product of ethylene oxide with a long chain aliphatic alcohol (e.g., heptadecaethylene oxycetanol), a condensation product of ethylene oxide with a partial ester derived from a fatty acid and a hexitol (e.g., polyoxyethylene sorbitol mono-oleate), or a condensation product of ethylene oxide with a partial ester derived from fatty acid and a hexitol anhydride (e.g., polyoxyethylene sorbitan monooleate). The aqueous suspension can also contain one or more preservatives such as ethyl or n-propyl p- hydroxybenzoate, one or more coloring agents, one or more flavoring agents and one or more sweetening agents, such as sucrose, aspartame or saccharin. Formulations can be adjusted for osmolarity.

[0073] Oil suspensions can be formulated by suspending a GAB 2 agonist, mimic, and modulatory agent in a vegetable oil, such as arachis oil, olive oil, sesame oil or coconut oil, or in a mineral oil

such as liquid paraffin. The oil suspensions can contain a thickening agent, such as beeswax, hard paraffin or acetyl alcohol. Sweetening agents can be added to provide a palatable oral preparation. These formulations can be preserved by the addition of an antioxidant such as ascorbic acid.

[0074] Dispersible powders and granules of the disclosure suitable for preparation of an aqueous suspension by the addition of water can be formulated from a GAB2 agonist, mimic, and modulatory agent in admixture with a dispersing, suspending and/or wetting agent, and one or more preservatives. Suitable dispersing or wetting agents and suspending agents are exemplified by those disclosed above. Additional excipients, for example sweetening, flavoring and coloring agents, can also be present.

[0075] The pharmaceutical formulations of the disclosure can also be in the form of oil-in-water emulsions. The oily phase can be a vegetable oil, such as olive oil or arachis oil, a mineral oil, such as liquid paraffin, or a mixture of these. Suitable emulsifying agents include naturally-occurring gums, such as gum acacia and gum tragacanth, naturally occurring phosphatides, such as soybean lecithin, esters or partial esters derived from fatty acids and hexitol anhydrides, such as sorbitan mono-oleate, and condensation products of these partial esters with ethylene oxide, such as polyoxyethylene sorbitan mono-oleate. The emulsion can also contain sweetening and flavoring agents. Syrups and elixirs can be formulated with sweetening agents, such as glycerol, sorbitol or sucrose. Such formulations can also contain a demulcent, a preservative, a flavoring or a coloring agent.

[0076] When the drugs are delivered by intravenous injection, the GAB2 agonist, mimic, and modulatory agent pharmaceutical formulations of the disclosure can be in the form of a sterile injectable preparation, such as a sterile injectable aqueous or oleaginous suspension. This suspension can be formulated according to the known art using those suitable dispersing or wetting agents and suspending agents which have been mentioned above. The sterile injectable preparation can also be a sterile injectable solution or suspension in a nontoxic parenterally-acceptable diluent or solvent. Among the acceptable vehicles and solvents that can be employed are water and Ringer's solution, an isotonic sodium chloride. In addition, sterile fixed oils can conventionally be employed as a solvent or suspending medium. For this purpose any bland fixed oil can be

employed including synthetic mono- or diglycerides. In addition, fatty acids such as oleic acid can likewise be used in the preparation of injectables.

vii.) Administration and Dosing Regimen of G AB2 Agonists, Mimics, and Modulatory Agents

[0077] The GAB2 agonists, mimics, and modulatory agents used in the methods of the disclosure can be administered in any conventionally acceptable way including via intravenous injection, IM, IP, orally, topically, vaginally or rectally. Oral administration is preferred. Administration will vary with the pharmacokinetics and other properties of the drugs and the patients' condition of health. The methods of the disclosure reduce cellular deterioration in patients who suffer from LOAD. The amount of GAB2 agonist, mimic, and modulatory agent that is adequate to accomplish this is considered the therapeutically effective dose. Precise dose schedules cannot be stated. The dosage schedule and amounts effective for this use, i.e., the "dosing regimen," will depend upon a variety of factors, including the stage of LOAD, the severity of LOAD, the severity of the adverse side effects, the general state of the patient's health, the patient's physical status, age and the like. In calculating the dosage regimen for a patient, the mode of administration is also taken into consideration. The dosage regimen must also take into consideration the pharmacokinetics, i.e., the agonist, mimic, and modulatory agent rate of absorption, bioavailability, metabolism, clearance, and the like.

[0078] The state of the art allows the clinician to determine the dosage regimen for each individual patient who suffers from LOAD (mild, moderate and severe stages). The formulations should provide a sufficient quantity of GAB2 agonist, mimic, and modulatory agent to effectively ameliorate LOAD which is characterized by symptoms such as memory loss, depression, restlessness, etc. A typical pharmaceutical formulation for oral administration of GAB2 agonist, mimic, and modulatory agent would depend on the stage of LOAD and condition of the patient. For example, a GAB2 agonist, mimic, and modulatory agent may be administered to a patient through mono therapy {i.e., with no other Alzheimer's medications) or in combination therapy with other Alzheimer's drugs including cholinesterase inhibitors such as DONEPEZIL (Aricept), GALANTAMINE (Razadyne), RIVASTIGMINE (Exelon), and TACRINE (Cognex), drugs that generally used for treatment of mild to moderate AD. NAMENDA (memantine HCl), a drug used

for treatment of moderate to severe AD may also be used in combination with GAB2 agonists, mimics, and modulatory agents. Thus, the dosages of selective GAB2 agonists, mimics, and modulatory agents administered to a patient may vary depending on age, degree of illness, drug tolerance, and concomitant medications and conditions.

viii.) Examples

[0079] The following specific examples are intended to illustrate the disclosure and should not be construed as limiting the scope of the claims.

[0080] In ε4 carriers from neuropathologically verified discovery, neuropathologically verified replication, and clinically characterized replication cohorts of 1425 cases and controls, LOAD was associated with six SNPs from the GRB -associated binding protein 2 (GAB2) gene and with a common haplotype encompassing the entire GAB2 gene. Of those homoaygous for both the ε4 allele and the GAB2 risk haplotype, 98.7% had LOAD. In addition, the GAB2 gene was over- expressed in pathologically vulnerable neurons, the GAB 2 protein was detected in neurons, tangle- bearing neurons and dystrophic neurites, and interference with GAB2 gene expression increased Tau phosphorylation. Collectively, the inventors have shown that GAB2 modifies LOAD risk in APOE ε4 carriers and influences Alzheimer's neuropathology.

[0081] More specifically, whole-genome association analysis of single nucleotide polymorphisms (SNPs) was performed using 502,627 SNPs in 760 histopathologically verified AD cases and controls (discovery cohort). AU samples were stratified for APOE-ε4 carrier status prior to analysis to identify additional LOAD risk loci. Of the 25 SNPs with the most significant p- values in the APOE-ε4 carrier group, 10 were located within the GRB2-associated binding protein 2 (GAB2) gene (p-valmax = 5.9x10-7; OR = 0.2571 (0.1173-0.5636). These results were replicated in an additional 326 histopathologically diagnosed cases and controls, also stratified for APOE-ε4 carrier status (replication cohort 1). An additional cohort of 383 clinically diagnosed LOAD samples provided further validation for the GAB2 association (replication cohort 2). There is a single haplotype block encompassing the entire gene and which contains four potentially functional non- synonymous nucleotide changes. This analysis identified a risk-enhancing and a risk-decreasing

haplotype within the GAB2 gene which modify risk differentially in combination with APOE alleles in LOAD.

[0082] Thus, the inventors conducted individualized genome-wide surveys of 502,627 single nucleotide polymorphisms (SNPs) in an effort to characterize and confirm novel LOAD susceptibility genes in three separate cohorts of LOAD cases and controls, including a discovery cohort of clinically and neuropathologically characterized expired brain donors, a replication cohort of similarly characterized expired brain donors, and a replication cohort of clinically characterized living subjects. The expired brain donor cohorts were selected to exclude clinically misdiagnosed cases and cognitively normal but neuropathologically affected elderly controls; the clinical cohort was selected to confirm genetic associations independent of any brain donor selection bias. Within each cohort, LOAD cases and controls were stratified into sub-groups of APOE ε4 carriers and non- carriers, permitting the investigation of genes that modify LOAD risk in the ε4 carriers and genes that might otherwise be masked by disproportionately large ε4 effects in the non-carriers. The inventors discovered associations between a common gene and LOAD in APOE ε4 carriers in the three neuropathologie and clinical cohorts and showed that the implicated gene is associated with AD neuropathology in neuronal microarray and immunohistochemical studies. Further provided is a possible mechanism by which GAB2 modifies AD risk in a preliminary small-interfering RNA (siRNA) study.

Example 1 : Design

[0083] The software program STRUCTURE (Pritchard et al. (2000) Genetics 155:945-59) was used to test for underlying genetic stratification, using 5000 randomly selected SNPs with at least 100 SNPs per chromosome. Initial analysis yielded empirical evidence of three populations (K=3). One, which contained fourteen samples and was far removed from the rest of the study population, was eliminated from further analyses. STRUCTURE was re-run with K=2, and each sample was assigned an admixture of the resulting populations. Comparison of the resulting case and control populations defined by these admixture vectors yielded a silhouette score of 0.06, indicating that while the samples in this study are the likely result of admixture of two populations, the distribution

of those populations is equivalent in cases and controls (Pritchard et al. (2000) Genetics 155(2):945- 59).

Example 2: Introduction to High-Density Genome-Wide Association Study

[0084] The genome-wide association study was performed using 502,627 SNPs on DNA samples extracted from brain tissue or blood of donors who were at least 65 years of age at the time of their death. The donors included a postmortem cohort of 664 patients who satisfied clinical and neuropathological criteria for the diagnosis of AD and 422 persons that did not meet clinical or neuropathological criteria for AD as controls (broken up into a 70% "discovery cohort" and a 30% "replication cohort 1" design). Tissue and neuropathological diagnoses were supplied by investigators from twenty National Institute on Aging Alzheimer's Disease Centers (ADCs) in accordance with an agreement with these Centers, the National Institute on Aging, and the National Alzheimer's Coordinating Center (NACC). Additional post-mortem samples were received from Sun Health Research Institute and the Netherlands Brain Bank. An additional sample series including 228 clinically assessed antemortem AD patients and 155 clinically assessed antemortem controls ("replication cohort 2") was provided by the Mayo Clinic Rochester. APOE genotypes were obtained by either pyrosequencing (Ahmadian et al. (2000) Anal. Biochem. 280(l):103-l 10) or restriction fragment length polymorphism (RFLP) analysis (Lai et al. (1998) Genomics 54(1):31- 38).

[0085] A three-tiered approach was used for the analysis of the genome-wide association study. The postmortem set was split into two cohorts matched for sex and age, a discovery cohort containing 70% of the postmortem sample ("discovery cohort") and a replication cohort containing 30% of the postmortem samples ("replication cohort 1"). An ante-mortem series from the Mayo Clinic Rochester was used as a second replication cohort ("replication cohort 2"). SNiPer-HD (Hua et al. (2007) Bioinformatics 23(l):57-63) and BRLMM were used to call the genotypes and the results were filtered to exclude any SNPs with SNiPer-HD quality score less than 0.45 (excludes monomorphic loci, and loci that do not cluster into three distinct Gaussians), those with Hardy Weinberg equilibrium p-value less than 0.01 or minor allele frequency less than 3%, and those where concordance between SNiPer-HD calls and BRLMM calls is less than 98%. After filtering,

321,953 SNPs remained. Allelic chi squares were computed for each SNP in the discovery cohort (70%) in 3 classes: all samples, APOE-ε4 carriers and APOE-ε4 non-carriers.

[0086] There were no loci in the ApoE-ε4 non-carriers that had a significant p-value after Bonnferroni correction for multiple testing (321,953 independent tests). Only one SNP, rs4420638, the only SNP on the AFFYMETRIX 500K Mapping Array in linkage disequilibrium (LD) with the ApoE-ε4 locus, was significant in the entire sample after multiple testing correction. Two SNPs in the ApoE-ε4 carrier population were significant after Bonferroni multiple testing correction. One of these SNPs, rs2373115, is located in an intron of GRB2-associated binding protein 2 (GAB2). Visual inspection revealed that ten of the top twenty-five SNPs in the ApoE-ε4 carriers stratified analysis were located in the introns or coding sequence of GAB2. These ten SNPs were then interrogated in the 30% replication cohort and the Mayo Clinic Jacksonville replication cohort (see Table 2 below). All ten SNPs examined were significant (p-value < 0.05) in the 30% cohort, and six SNPs were significant in the Mayo Rochester ante-mortem cohort. The most significant p- values and lowest odds ratios (indicating a protective allele) were found to be located toward the distal, 5' end of GAB2 (see Table 2 below).

TABLE 2

DISCOVERY COHORT

REPLICATION 1 COHORT

REPLICATION 2 COHORT

Example 3: High-Density Genome-Wide Association Study Specifics

[0087] Individualized genome-wide surveys were performed in a "neuropathological discovery cohort" of 749 expired brain donors, a "neuropathological replication cohort" of 307 expired brain donors, and an additional "clinical replication cohort" of 369 living subjects who were at least 65 years old at the time of their death or last clinical assessment and who were independently assessed for their APOE genotype. For the two neuropathological cohorts, brain tissue for DNA extraction, neuropathological diagnoses and data were supplied by investigators from twenty of the National Institute on Aging (NIA) sponsored Alzheimer's Disease Centers (ADCs) (in accordance with agreements with the NIA, the ADCs and the National Alzheimer's Coordinating Center) and from the Netherlands Brain Bank. 70% of the 1 ,056 expired LOAD cases and controls were randomly assigned to the hypothesis-generating neuropathological discovery cohort and the remaining 30% were assigned to the hypothesis-testing neuropathological replication cohort. For the hypothesis- testing clinical replication cohort, DNA extracted from blood, clinical diagnoses and data from subjects assessed in Rochester, Minnesota, were supplied by investigators from the Mayo Clinic.

[0088] The neuropathological discovery cohort included 454 LOAD cases (274 ε4 carriers and 180 ε4 non-carriers) and 295 controls (105 ε4 carriers and 190 ε4 non-carriers); the neuropathological replication cohort included 193 LOAD cases (102 ε4 carriers and 91 ε4 non- carriers) and 114 controls (27 ε4 carriers and 87 ε4 non-carriers); and the clinical replication cohort included 224 LOAD cases (117 ε4 carriers and 107 ε4 non-carriers) and 145 controls (28 ε4 carriers and 117 ε4 non-carriers). The expired brain donor cases satisfied clinical and neuropathological

criteria for LOAD, and were age 73.5±6.2 at the time of death. The expired brain donor controls did not have significant cognitive impairment or significant neuropathological features of AD, and were age 75.8±7.5 at the time of death. The clinical cases satisfied criteria for probable AD, and were ages 78.9±7.8 at their last clinical visit. The clinical controls did not have clinically significant cognitive impairment and were age 81.7±6.6 at their last clinical assessment. APOE genotypes were obtained in each subject by either pyrosequencing (Ahmadian, supra) or restriction fragment length polymorphism (RFLP) analysis (Lai, supra).

[0089] The 500K GeneChip (Affymetrix, Santa Clara, CA) was used to survey 502,267 SNPs in each subject. Genotypes were extracted using both SNiPer-HD (Hua et al (2007) Bioinformatics 23:57-63) and BRLMM (Affymetrix, Santa Clara, CA) software, and 321,953 SNPs were analyzed after excluding those that were monomorphic, did not cluster into three distinct Gaussian distributions, were associated with Hardy Weinberg equilibrium significance levels less than 0.01, had minor allele frequencies less than 3%, or exhibited less than 98% concordance between the SNiPer-HD and BRLMM calls. The software program STRUCTURE {supra) was employed to test for underlying genetic stratification, using 5,000 randomly selected SNPs and including at least 100 SNPs per chromosome. The initial analysis yielded empirical evidence of three populations. Since fourteen subjects belonged to a population far removed from the rest of the study population, they were eliminated from farther analyses. STRUCTURE was used to demonstrate a comparable admixture of the two populations in the cases and controls. After stratifying the LOAD cases and controls for presence or absence of the APOE ε4 allele, allelic chi squares were computed for each SNP.

[0090] The inventors initially surveyed SNPs in the neuropathological cohort to explore LOAD associations in the ε4 carrier and non-carrier sub-groups. Within the discovery sub-group of APOE ε4 carriers, ten of the twenty- five SNPs with the most significant LOAD-association significance levels (p=2xlθ ^"4 to 6x10 ^~7 ) were located in the GRB-associated binding protein 2 (GAB2) gene on chromosome 1 Iql4.1 (see Figure 8). LOAD associations in six of these SNPs were confirmed in both the neuropathological replication and clinical replication cohorts (see Figure 8). These ten SNPs were not significantly associated with LOAD in the APOE ε4 non-carrier group (p=0.08 to

0.97). Combining data from all 653 APOE ε4-carrying cases and controls, we found highly significant associations between LOAD and all 10 GAB2 alleles (p-values 1.19 x 10 ^"5 to 9.66 x 10 ^{" π} ), with 5 of the 6 consistently implicated alleles surviving the highly conservative Bonferroni correction for 321,953 independent comparisons (see Figure 8).

[0091] Haploview v3.32 was used to determine the linkage disequilibrium (LD) structure of the chromosome l lql4.I region surrounding GAB2 in each of the three APOE ε4-stratified cohorts (Figure 1). Three haplotype blocks are present in this region: one block upstream of GAB2, roughly corresponding to the ALG8 locus; one 189 kilobase-pair block encompassing most of the GAB2 locus; and one downstream block corresponding to the NARS2 locus. These blocks were consistent with the LD structure of the HapMap CEPH populations. The GAB2 gene is completely encompassed by this single haplotype block, which has three major haplotypes: an extremely common "GAB2 risk haplotype," a common "GAB2 protective haplotype," and a relatively uncommon GAB2 haplotype unrelated to LOAD risk in APOE ε4 carriers (Figure 1). In all three cohorts, the GAB2 CT-AAG-CAGATCAGACG haplotype was associated with higher LOAD risk, the GAB2 TC-GCA-TGAGGTGTCTT haplotype was associated with a lower LOAD risk, and the TA-T-GGA was unrelated to LOAD risk in the APOE ε4 carriers (Figure 8). The association between the NARS2 haplotype and decreased LOAD risk in two of the cohorts appears to be attributable to moderate LD between the NARS2 and GAB2 blocks.

[0092] Data from the 1425 subjects (including 653 APOE ε4 carriers and 772 non-carriers) in all three cohorts were combined to characterize odds ratios (ORs) and 95% confidence intervals (CIs) for GAB2 risk haplotype homozygotes (see Table 6 below) and GAB2 protective haplotype carriers.

TABLE 6

APOE e4 APOE ε4 GAB2 risk Control Cases % OR ¹ OR ² OR ³

Group Group OR haplotype dose S (N) LOAD (95% Cl) (95% Cl) (95% Cl)

(N) GAB2 vs GAB2 HM vs entire ε4 Non-HM in group the ε4 group ¹

0.99 0.99

0-1 130 110 45.8%

Non-Carriers (0.74-1.34)

1.00 (0.74-1.34) 1.01 2 1.00 222 189 1.00 (073-1.38) 46.0% (0.78-1.2B) (0.78-1.28)

0 57 3 45

„ „ 0-1 53 157 748% (039-0 84)

Carπers (1 81-6.68) 2.70

(46 ⁶ 3-- ⁰ 7 ⁷ 95) 1 55 9 41 (1.72-4.73)

41 328 88 9% (1 04-2 30) (4.82-1829)

' OR in comparison with all of the subjects from the same APOE ε4 carrier or non-earner group

² OR in comparison subjects carrying no copies of the APOE e4 allele and fewer than 2 copies of the GAB2 risk haplotype

³ OR of GAB2 risk haplotype homozygotes (HM) in comparison with the non-homozygotes from the same APOE ε4 earner or non-carrier group

[0093] In ε4 carπers, 67% of the LOAD cases and 43% of the controls were homozygous for the GAB2 risk haplotype. In comparison with the other ε4 carriers, GAB 2 risk haplotype homozygotes had a significantly higher risk of LOAD (OR 2.70, 95% CI 1.72-4.73, P=9.4 x 10 ^"6 ). Indeed, 78 out of 79 persons (98.7%) homozygous for both the ε4 allele and the GAB2 risk haplotype had LOAD, a significantly higher proportion than the 35 out of 39 APOE ε4 homozygotes (89.7%) with fewer than 2 copies of the GAB2 risk haplotype (P=0.023). Within the same APOE ε4 carrier group, 22% of the LOAD cases and 49% of the controls had at least one copy of the GAB2 protective haplotype In comparison with the other ε4 carriers, carriers of the GAB2 protective haplotype had a significantly lower risk of LOAD (OR 0.37, 95% CI 0.23-0.59, P=I 5 x 10 ^"6 ). In contrast, APOE ε4 non-carriers homozygous for the GAB2 risk haplotype did not have a higher LOAD risk than the other ε4 non-carriers (OR 1.01, 95% CI 0.73-1.38, P=0.97) and ε4 non-carriers with the GAB2 protective haplotype did not have a lower LOAD risk than the other ε4 non-carriers (OR 1.05, 95% CI 0.74-1.50, P=0.79). Whereas the inventors confirmed a younger age at dementia onset in the APOE ε4 carriers than in non-carriers (age 80.8±7.7 versus 86.0±7.5, P=3.8 x 1O ^'S ), there was no significant GAB2 haplotype effect on age at dementia onset in either the ε4 carriers (P=0.32) or non-carriers (P=O.84) Hence, the effects of GAB2 on LOAD risk most likely depends on both the ε4 allele and a person's age.

Example 4: Identification of a Protective GAB2 Haplotype in LOAD

[0094] Haploview v3.32 was used to determine the linkage disequilibrium (LD) structure of the region on chromosome 11 q 14.1 surrounding GAB2 for each of the three APOE-ε4 stratified cohorts (discovery, replication 1, and replication 2; see Figure 1). Three haplotype blocks are present in this region, one upstream of GAB2, roughly corresponding to the ALG8 locus, one 189 kilo-base pairs (kb) block encompassing most of the GAB2 locus and one downstream block corresponding to the

NARS2 locus. These blocks are consistent with the LD structure of the HapMap CEPH populations. The GAB2 haplotype CT-AAG-CAGATCAGACG increases LOAD risk in all populations, although most strongly in the "replication 2" cohort (see Table 3 below). This is a surprising finding since this haplotype is common, being present in 67-78% of the individuals across our three separate study populations. The haplotype defined by TC-GCA-TGAGGTGTCTT shows a strong protective effect in the "discovery" and "replication 1" cohorts and is trending toward a protective effect in the "replication 2" cohort (see Table 3 below). Moderate LD exists between the GAB2 block and the NARS2 block in all three populations, but is greatest in the "replication 2" cohort. This LD accounts for the significance of the TATGGA NARS2 haplotype in the "discovery" cohort, and the trend toward significance in the 'replication 1 ' cohort.

Example 5: Arrangement of SNPs

[0095] Figure 7 is a table which provides additional details on each SNPs 5-22 in each of the GAB2 haplotypes including location and strand.

Example 6: Neuronal Microarray Studies

[0096] In order to provide converging evidence that GAB2 is biologically relevant to AD neuropathology, the inventors analyzed data from a neuronal microarray study of LOAD cases and

controls. Expression profiling using the Affymetrix Human Genome U 133 Plus 2.0 array was used to characterize and compare GAB2 expression in laser-capture microdissected non-tangle bearing neurons of cases and controls in six brain regions differentially affected by AD. LOAD cases had significantly greater neuronal GAB2 expression in the posterior cingulate cortex (9 cases, 13 controls, 4.50-fold change, P=O.00039) and hippocampus (9 cases, 13 controls, 2.94-fold change, P=0.00085), and no significant differences in the entorhinal cortex (10 cases, 13 controls, 1.20-fold change, P=0.46), middle temporal gyrus (13 cases, 12 controls, 1.44-fold change, P=0.14), superior frontal gyrus (22 cases, 11 controls, 1.25-fold change, P=0.47), or primary visual cortex (17 cases, 12 controls, 1.53-fold change, P=0.14).

[0097] The hippocampus is especially vulnerable to AD-related neurofibrillary tangles (Braak et al. (1991) Acta. Neuropathol. 82:239-259), neuronal loss and brain atrophy (Bobinski et al. (2000) Neuroscience 95:721-725). It is preferentially involved in AD-related memory impairment (Jack et al. (1999) Neurology 52:1397-1403) and is associated with the highest cerebral GAB2 expression in the rodent brain (Lein et al. (2007) Nature 445: 160-161). The posterior cingulate cortex is vulnerable to AD-related hypometabolic abnormalities and fibrillar amyloid deposition, and is also involved in AD-related memory impairment. While the entorhinal cortex, temporal and prefrontal regions are also affected by AD neuropathology, the visual cortex is relatively spared. Using a repeated measures analysis of variance to analyze neuronal GAB2 gene expression data from the same 8 LOAD cases and 10 controls, there was a significant group-by-region interaction (P=OOl 1), with LO AD-related increases in neuronal GAB2 gene expression that were greater in the posterior cingulate cortex and hippocampus than in the visual cortex.

Example 7: Immunohistochemical Validation

[0098] In order to show neuronal expression of GAB2 at the protein level, GAB2 immunohistochemistry was assessed in LOAD hippocampus and posterior cingulate cortex in LOAD cases. In hippocampus, GAB2 immunoreactivity was observed in structures with the morphology of dystrophic neurites or neuropil threads, neurons, and corpora amylacea. The putative neurons were almost entirely dystrophic in appearance (Figure 4A) or had cytoplasmic inclusions resembling neurofibrillary tangles (Figure 4B). Dystrophic neurons and neurites (Figure

4C) and neurofibrillary tangle bearing cells (Figure 4D) were also revealed by the GAB2 antibody in posterior cingulate. Here, however, many relatively normal neurons were observed as well, with long stretches of immunoreactive apical dendrites ascending through the cortical layers (Figures 4C and 4D).

Example 8: siRNA Study

[0099] In addition to its other properties, GAB2 is the principal activator of the phosphatidylinositol 3 -kinase (Pϊ3K) signaling pathway (Pratt et al. (2001) J. Immunol. 165:4158- 4163). PI3K activates Akt, which in turn promotes glycogen synthase kinase-3 (GSK-3) phosphorylation/inactivation, suppressing GSK-3-dependent phosphorylation of tau at AD-related hyperphosphorylated tau residues, the principal component of neurofibrillary tangles, and preventing apoptosis of confluent cells (Baki et al. (2004) EMBO J. 23:2586-2596; Kang et al. (2005) J. Biol. Chem. 208:31537-31547). Based on this relationship, GAB2 could function to protect cells from neuronal tangle formation and cell death and a loss-of-function GAB2 haplotype would lose some of this protection function. Thus, interference with GAB2 expression using siRNA treatment would increase tau expression at the serine-262 residue known to be hyperphosphorylated in AD.

[00100] Neuroglioma cells overexpressing tau protein were grown in 96-well plates and transfected with siRNA directed at GAB2 mRNA. Following four days of transfection, cells were fixed, permeablized, and immunostained with antibodies against total tau protein and tau protein phosphorylated on serine-262. A FITC- and Cy5-conjugated secondary antibody cocktail was then applied. After incubation and washing, images were captured and quantitated using the InCeIl imager 3000 (General Electric). The fold increase in serine-262 phosphorylated tau levels was calculated against control samples that had been transfected with a scrambled siRNA sequence. As shown in Figure 5, GAB2 siRNA treatment was associated with a 1.70-fold increase in serine-262 phophorylated tau, and the apparent increase did not appear to be attributable to an increase in total tau levels. Additional siRNA and protein validation studies are performed to confirm this finding (data not shown).

Example 9: Summary of Findings

[00101] The present inventors were able to use their genome-wide survey of more than 500,000 SNPs, two unusually large clinically characterized and neuropathologically verified cohorts of AD cases and controls, a third cohort of clinically well characterized subjects, and stratification of the samples carriers and non-carriers of a major susceptibility gene to characterize and confirm associations between the GAB2 gene and LOAD risk in APOE ε4 carriers. Six SNPs and a common haplotype block encompassing the entire GAB2 gene were implicated in three independent cohorts. The data from the microarray study of laser-capture microdissected neurons in LOAD cases and controls, immunohistochemistry, and a siRNA study provided converging evidence for the relevance of GAB2 to the neuropathology of LOAD, This allows the inventors to link to a mechanism by which GAB2 can modify LOAD risk in ε4 carriers and ,thus, provide targets at which to aim new treatments.

[00102] Although only one genotyping platform was used in all three cohorts, the inventor's findings are unlikely to be attributable to any platform-related bias in genotyping calls since the observed association was not limited to a single SNP but was related to a large haplotype block in agreement with the LD structure of the HapMap CEPH population. Further, all six of the implicated scores had high-quality SNiPer-HD scores (greater than 0.45), indicating that the data for each SNP clustered into three distinct Gaussian distributions corresponding to the three possible genotype calls. While GAB2 haplotype s were characterized in only a few of the subjects from the microarray studies, inspection of posterior cingulate GAB2 gene expression in the three GAB2- characterized controls homozygous for the risk haplotype (10.8±10.1 raw expression units) and the three GAB2-characterized non-homozygous controls (32.8±28.6 raw expression units) support that GAB2 is under-expressed in GAB2 risk haplotype homozygotes (See also, Figure 6).

[00103] GAB2 is a scaffolding protein involved in multiple signaling pathways, which can affect AD-related tau, amyloid, metabolic, or other aspects of AD pathology and cell survival. Thus, discovery of this novel LOAD susceptibility gene provides new opportunities to investigate LOAD pathogenesis, predisposition, treatment and prevention.

[00104] Various modifications and variations of the present disclosure will be apparent to those skilled in the art without departing from the scope and spirit of the disclosure. Although the

disclosure has been described in connection with specific preferred embodiments, it should be understood that the disclosure as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the disclosure which are understood by those skilled in the art are intended to be within the scope of the claims.

Example 10; Additional Markers

[00105] To identify additional markers that may be used in the diagnosis and monitoring of Alzheimer's disease, the data generated from the previous examples was split into three cohorts: a neuropath discovery cohort, a neuropath replication cohort and a clinical replication cohort. Significant SNPs were selected because they were significant (p<0.05) and displayed the same direction of effect (the risk allele was the same) for all three cohorts. This analysis yielded three sets of SNPs: the SNPs that were significant for all samples (Table 7A), the SNPs that were significant in APOE ε4 carriers (Table 7B) and the SNPs that were significant in APOE ε4 noncarriers (Table 7C).

[00106] The data was then combined and split by autopsy status and country of origin into three cohorts: a US postmortem cohort, a Netherlands postmortem cohort and a US antemortem cohort. Significant SNPs were selected because they were significant (p<0.05) and displayed the same direction of effect (the risk allele was the same) for all three cohorts. This analysis yielded three sets of SNPs: the SNPs that were significant for all samples (Table 8A), the SNPs that were significant in APOE ε4 carriers (Table 8B) and the SNPs that were significant in APOE ε4 noncarriers (Table 8C). The markers identified in Tables 7 and 8 may be used in a similar manner as the GAB2 associated markers.

ADDITIONAL SEQUENCES

SEQ ID NO: 4 - nucleic acid sequence for human GAB2: 6006 bp

1 atgagcggcg gcggcgacgt ggtgtgcacc ggctggctga ggaaatcgcc tcccgagaag 61 aagttgaggc gctatgcctg gaagaaacgc tggtttatcc tgcggagtgg ccggatgagc 121 ggtgacccag atgttctgga atactacaag aacgatcact ccaagaagcc tctgcggatc 181 atcaacctga acttctgtga gcaggtagat gcaggcctga cctttaacaa gaaggagctg 241 caggatagtt ttgtgtttga catcaagacc agtgaacgca ccttttacct ggtggctgag 301 acagaagagg acatgaataa gtgggtccag agcatctgcc agatctgtgg cttcaatcag 361 gctgaggaga gcacagactc cctgagaaat gtttcctcag ccggtcatgg cccccgctct

421 tctccagctg agctcagcag ctctagccag caccttctcc gagagcgcaa gtcctcagcc 481 ccatcacact ccagccagcc aactctgttc acgtttgaac cccctgtgtc aaaccacatg 541 cagcccacct tgtccaccag cgcacctcag gagtatctct acttgcacca gtgcataagc 601 cgaagagcag aaaatgcaag gagtgccagc ttctctcagg gcaccagagc ctcttttctc 661 atgaggagtg acacagctgt acaaaaactt gcccagggca atggacactg tgtcaacggg 721 atcagtggtc aagtccatgg cttctatagc cttcccaagc cgagccggca caatacagaa 781 ttcagagaca gtacctacga cctcccccgc agcctggcct cccatggcca caccaagggc 841 agcctcacag gctccgagac agataatgag gatgtgtaca ccttcaagac gcccagcaac 901 accctgtgca gggagttcgg ggacctcctg gtagacaata tggatgttcc ggccacccca 961 ctctcagcct accagatccc taggacattc actctggaca aaaaccacaa tgccatgaca 1021 gtggccactc ctggggactc agccatagct cccccacccc gcccccccaa gccaagtcag 1081 gcagaaacac ctcgatgggg cagtcctcag cagagaccgc caatcagtga aaatagcaga 1141 tctgtcgctg ccaccatccc cagacgcaac accctccctg caatggacaa cagccgactt 1201 caccgagctt cttcctgtga gacctacgag tacccacagc gtggtggaga gagtgcaggc 1261 cggtctgctg aatccatgag tgatggagtt ggctctttcc tgccagggaa aatgattgtg 1321 ggccgatcgg acagcaccaa ttctgaagac aactatgtgc ccatgaatcc aggttcttcc 1381 accctgttgg ccatggaacg agcaggtgat aattcccaga gcgtctacat cccaatgagc 1441 ccaggggccc atcactttga ctcacttggc tacccatcaa caacccttcc tgtgcaccga 1501 ggccccagca gaggaagtga gattcagcca ccccctgtca accgcaacct caaacctgat 1561 cggaaagcaa agccaacacc acttgacctg aggaacaaca ccgtcatcga tgaactcccc 1621 ttcaagtcac ctatcaccaa gtcttggtct agggccaacc acaccttcaa ctccagctcc 1681 tcccagtact gccgccccat ctccacccag agcatcacca gcacagactc aggagacagc 1741 gaagagaact atgtccctat gcaaaaccca gtgtctgcat ctcccgttcc cagtggcacg 1801 aacagtcctg cccctaagaa gagcaccggc agcgttgatt atctggccct ggacttccag 1861 ccgagctccc caagccccca ccgcaagcca tctacttcat ccgtcacctc tgatgagaag 1921 gtggactacg ttcaggtgga caaggagaag acccaggccc tgcagaacac catgcaggag 1981 tggacagacg tgcggcagtc ctcagagcct tccaagggtg ccaagctgtg atgagagggc 2041 caccgcagag cccaggaggc agcatctcca gagctggccc ttcccatctc ccctctcccc 2101 tctcccgttc ttcctcccat ccacctcctc tctactctgc cagtctcagc cttcaaagca 2161 cttgacatca gggaccctga acccttcccc tgggaggtga gggcctgatc aaggcacctc 2221 ctctgcccac tcggggccca gctgtgattt ttatcagtaa tggccatgcc tccacccacc 2281 ttagttagga gctacttcca aaaagcatcc ttcagcctct tcctgtcctt tagacctgac 2341 tctctaccag atgtttggag ggaagggctg gggctctgag ccagattcca cacctcacgt 2401 tcagtcacag ccctcagcta tcttccctcc ggccactggg ctacctctcc ttcagtccca 2461 gaagacaagt ctcaccaacG cagggagtca aggaccagca aaccaaagtg gataatggac 2521 tttttcattc ctgtttttct tggcaggaga gaagcaaggc cactaaaaga ggagatggtg 2581 gagacggagg ctcagcagtg gtcttgaggg gtaaaggact tagatgccca gatgaagagg 2641 gaaagctgac atctgcaggg aacccacttt gaggctgagg ccatggcagg acagctgctg 2701 tggggtgcag aggcagaaga tgaaggacaa aaggaaggga aaactgatgg ccaacctaga 2761 gcagcaagga gcagggcttg gagctcgggt ggtggagatg acaaggacac tgtggggtct 2821 gggtccccag aactctggag ctacaggcca ctctaggccc aagggctagt cctcttcccc 2881 agtcccctca gaggcccccg ccagccccac cttgaaagca gcatacaggg gaaggcttgg 2941 accaagctgg gcgaccaagc acatggggca ggaacacatg gtaaaggggt ggggaatatg 3001 ggagggagtg tggtgtggat gggggtgatg cagggactga ggggaaccct gggacaggca 3061 caggctgggc agaggcacag ggcagtgcag gggactctgc agtggggtcg ggaagtgagt 3121 ttctttgcag tgagcagtgc agtggaagtc gggcaoagag gtagcagaca gatgtgaagc 3181 agtggtgaag gccatgtagc aagtggggaa atacatccaa agggcctggg agttgggggg 3241 tgcccaacgc aatccttggg ggtgcagggt ggagcagaaa gtgaaggagg gacacgtgca 3301 agagtggtgt gcatggtggt gtgacatgag gaccgttcct aggatgggac agtgggtcag 3361 gcaggacaag gagaaagcag ggcagaatga tgcctagagg accacatcag gcatggctga 3421 cagcttgtgc ccatgggctg tggcgtatgt cagatcgcag ggtaggaacg agtctggcct 3481 ggtgccggcc cagtgtttcc tcagctcatc cgccctctgt tgctccctag cattccagga 3541 gccatcttgg actctcctcc ccaggtttga aaggccatca gattagcagg gacggggtgt 3601 agggcatoac ccaaggttcc ttctcttaaa ctaagggtgg gggatctgaa tgtttttatg

3661 ttgactgttc ttgactaaat tttcaagagt ttcagaagca acaggacaga ccagacgttt 3721 cattctaccc tggggcgaac agaacttctt cctcccaaac aatgacttcc tgccatgttt 3781 gatggggaca gctaccactg tcctctgccc ccattcccct ttcagctccc atgagcatgc 3841 atagttcacc agaccaatgg cctagccatt ctctaagtcc catcctggaa gaagttattt 3901 cttcaagagc tgcacctctc ctcctagcat tagtttagat caactcaagg agtatttatt 3961 aatggctgct gtctccagtt tctggggtta agcactaagg acacaagaat caatcagacc 4021 ttctccctga acttaagata gccacaatca gaaaaaggac aaggacatga gacagtggtg 4081 atggccatca gacagagact tcaaatgctg atggagggca gaggaagtac ttagggaggt 4141 tggtgtcaga ggcaggagtg ggggatcagg gaaggtggat tctaggaaaa gggagtgcct 4201 gaggtaggcc ttagaagggg atgagtcaga tttttacaga ggaggagggc agggcttggg 4261 tccagtggag gaagaaggaa ggagaggctt ggaaagcctt gtgtcttggg aaaaaaaggc 4321 ctttgagcat atgggtccag ccactcagaa gtgcaggggc catgccttgg tgttccaata 4381 agtgaatgga agcagtggtg gtagctacac tgggcagagt tggcagggtg ctggttcact 4441 ctgcccagcc ctgaatgtgt gccttaaagg ccccctacaa ggggccccat acgacagagc 4501 ttttaactgg tgccttccct gtacccgcag cagccacaag tgggcccaga ctattgcagc 4561 ctcccataaa catgtgagca tgttctgagt gtgccatgat gtgagtggac ctggctggaa 4621 tcttcggaga gcgactgagg tgttcaaatc gaatctccca ggaggcttcc ttccagcccc 4681 ctattctggt aactaccagg aggcttcctt ccagccccct attctggtaa ctaccaaaat 4741 ccctcgggtg caagtgtagg ggtagagatg gaaggatgag aggtgaaatt gaccttttga 4801 aagcaaagct ctggctcaca ggccccaaac taccagccgt atctagcata tccccaccct 4861 ccacccacta cctcctccaa caaaggagtc aactcagttg aaaaaactgg tcctttggcc 4921 tatccatggg tcaaagtcca cctctcctgg gggcctggag aggactgagc ctacggaaag 4981 gggatacctt cccactcagc actgcttcac acaggccccc tgcctggggc tctccaagga 5041 gccttcttca cccacttcca gctccacttc tgcaaggtta agtcaagtga gaacgatgag 5101 aaatagggag atggtgtctc cttaagtcct tgatctgcct gtctgtggaa tgggaggttg 5161 gattagctgc gctgaggtcc catccacagc tggtgctcag ctgcttgaag gggagactcc 5221 ctcctctgta acttctttct gggggattgg ggtgggcagt acctatcccc agtcccctcc 5281 tagcttgact ttagtggttt ccaatgtaga agttaacaaa gtatgcccca ttcctgtgac 5341 aaaagcacaa ccattctgaa gttactggag catgggctca gctcatcctc cctctggccc 5401 cttctcccat ggggacatct cggccoagca cccctatccc atttccagag ttcttccttc 5461 cccatctggg ccttcataaa atgcagggga agccagactg gtctcaggag cgctaaagcc 5521 cttccgtggg gggtcgtctt tctgggacta gccctgctgt ttaggacctg ggaccacaat 5581 ggggtacctg ccgagggggt ccccaagaga tccaggctgt catgtgattt atggtggcat 5641 gtgttgtgta tttgttggct acttgtgtct tgaaatctag aattatttca cgcagaattg 5701 tcactgtttg tcaggaagag aaaatgggct agtggaagcc cagtcttgag ttcttgtctt 5761 gttaccattt aaaattgaca tttaattttc aaatcactgt tggtgcctaa tcacttaagt 5821 tattaattta ttctgttgta ttcttttttt taaattgtaa catatttatc cggtgggtgg 5831 gacaggagtg tgttcaagtg ggtcatgttt ttgctgtggt gacacatggt acaggcttgg 5941 agcttgcagg tccctttcta ctgtggtgtt ggagcaggac aataaagtcc actagaaatg 6001 cacccc

SEQ ID NO: S - amino acid sequence for human GAB2: 1-676 amino acids

1 msgggdvvct gwlrksppek klrryawkkr wfilrsgrms gdpdvleyyk ndhskkplri

61 inlnfceqvd agltfnkkel qdsfvfdikt sertfylvae teedmnkwvq sicqicgfnq

121 aeestdslrn vssaghgprs spaelssssq hllrerkssa pshssqptlf tfeppvsnhm

181 qptlstsapq eylylhqcis rraenarsas fsqgtrasf1 mrsdtavqkl aqgnghcvng

241 isgqvhgfys lpkpsrhnte frdstydlpr slashghtkg sltgsetdne dvytfktpsn

301 tlcrefgdll vdnmdvpatp lsayqiprtf tldknhnamt vatpgdsaia ppprppkpsq

361 aetprwgspq qrppisensr svaatiprrn tlpamdnsrl hrasscetye ypqrggesag

421 rsaesmsdgv gsflpgkmiv grsdstnsed nyvpmnpgss tllameragd nsqsvyipms

481 pgahhfdslg ypsttlpvhr gpsrgseiqp ppvnrnlkpd rkakptpldl rnntvidelp

541 fkspitksws ranhtfnsss sqycrpistq sitstdsgds eenyvpmqnp vsaspvpsgt

601 nspapkkstg svdylaldfq psspsphrkp stssvtsdek vdyvqvdkek tqalqntmqe

661 wtdvrqssep skgakl

TABLE 7A (part 1)

TABLE 7A (part 2)

NEUROPATH REPLICATION COHORT

No. Mr Mj MAF Affected MAF Unaffected CHISQ P OR L95 U95 1 T C 0.5134 0.4167 5.307 0.02124 1.477 1.059 206 2 C T 0.08743 0.03913 5.144 0.02333 2.353 1.102 5.024 3 T C 0.3968 0.4828 4.333 0.03739 0.7049 0.5069 0.9803 4 C T 0.2884 0.3707 4.481 0.03428 0.6879 0.4861 0.9734 5 C A 0.3235 0.4035 3.963 0.04651 0.707 0.5022 0.9952 6 C T 0.02941 0.07018 5.494 0.01908 0.4015 0.1829 0.8813 7 G T 0.117 0.1739 3.868 0.0492 0.6295 0.3959 1.001 8 C T 0.3803 0.4655 4.296 0.03819 0.7047 0.5058 0.9816 9 A C 0.0788 0.1697 11.32 0.0007654 0.4185 0.2492 0.7028 1O T A 0.2219 0.3 4.61 0.03179 0.6655 0.4584 0.9662

11 A T 0.2364 0.3174 4.736 0.02954 0.6659 0.4612 0.9614

12 T C 0.1905 0.2759 6.05 0.01391 0.6176 0.42 0.9084

13 G A 0.2326 0.1577 4.809 0.02832 1.62 1.05 2.498

14 T C 0.03514 0.07456 4.602 0.03194 0.452 0.2152 0.9491

15 A G 0.1481 0.2155 4.545 0.03302 0.633 0.4149 0.9658

16 C T 02354 0.1509 6.351 0.01173 1.733 1.127 2.667

17 C T 0.3862 0.2241 17.23 0.00003308 2.178 1.502 3.159

TABLE 7B

SNP INFORMATION

No. CHR Physical Position Affymetrix ID dbsnp RS ID Gene Gene Relationship

1 1 61709161 SNP_A-1898422 rs4501865 NFIA downstream

2 2 202181323 SNP_A-2129507 rs12464623 ALS2CR11 intron

3 3 143828507 SNP_A-2216781 rs6800573 PLS1 intron

4 4 4864223 SNP_A-2223920 rs1907991 STX18 upstream

5 10 98112342 SNP_A-4276326 rs10748673 TMEM10 upstream

6 11 5529769 SNP_A-2007311 rs11038071 OR52H1 upstream

7 11 77880036 SN P_A-1867409 rs7395344 NARS2 intron

8 13 74601462 SNP_A-4247955 rs7985837 TBC1D4 downstream

9 13 76269462 SNP_A-2084510 rs7317065 KCTD12 downstream 10 15 93465782 SNP A-2302677 rs1458253

NEUROPATH REPLICATION COHORT

No. Mr Mj MAF Aff d MAF Unaffected CHISQ P OR L95 U95

1 C T 0.1157 0.2222 4.142 0.04183 0.4581 0.2132 0.9846

2 C T 0.04167 0.1481 8.302 0.00396 0.25 0.0916 0.6827

3 C T 0.2009 0.07407 4.798 0.02848 3.143 1.076 9.181

4 G A 0.1574 0.3148 6.986 0.008214 0.4066 0.2058 0.8034

5 T A 0.1376 0.2593 4.723 0.02976 0.4559 0.2218 0.937

6 T A 0.004673 0.03704 4.08 0.04338 0.1221 0.0109 1.372

7 A G 0.1651 0.2963 4.815 0.02821 0.4698 0.2369 0.9318

8 T G 0.3255 0.54 8.02 0.004625 0.411 0.2198 0.7687

9 T C 0.1606 0.2963 5.235 0.02214 0.4542 0.2285 0.9029

10 C G 0.4815 0.3148 4.852 0.02761 2.021 1.073 3.807

Table 7C (part 1)

SNP INFORMATION

No. CHR Physical Position Affymetπx ID dbsnp RS ID Gene Gene Relationship 1 3 137529879 SNP_A-2211252 rs696081 PCCB intron

2 3 137643280 SNP_A-2280398 rs6791611 STAG1 intron 3 3 173887236 SNP_A-2075581 rs494694 AADACL1 intron

4 3 173888897 SNP_A-2285449 rs569097 AADACL 1 intron 5 4 158859117 SNP_A-1979572 — GRIA2 downstream 6 6 124371195 SNP_A-1824132 rs2136699 TCBA1 intron

7 8 58529428 SNP_A-4274680 rs 12679977

8 9 26246238 SNP_A-1996325 rs473367 C9orf82 downstream

9 10 112534645 SNP_A-2001430 rs7077757

10 11 8830651 SNP_A-2060836 rs4910068

11 11 9021842 SNP_A-4192560 rs7125458 SCUBE2 intron 12 14 95019742 SNP_A-2114676 rs1187738 C14orf49 upstream 13 15 35957646 SNP_A-2073562 rs11853877 MEIS2 upstream 14 15 35957844 SNP_A-4194683 rs16965830 MEIS2 upstream 15 18 33609512 SNP_A-2118637 rs2114517 BRUNOL4 upstream 16 20 11550966 SNP A-2045721 rs6078266

Table 7C (part 2)

NEUROPATH REPLICATION COHORT

No. Minor Major MAF Affected MAF Unaffected CHISQ P OR L95 U95

1 C T 0.4938 0.3876 3.854 0.04962 1.541 0.9998 2.374

2 C G 0.525 0.4157 4.041 0.0444 1.553 1.01 2.389

3 G A 0.4625 0.3523 4.227 0.0398 1.582 1.021 2.453

4 C T 0 4625 0.3539 4.12 0.04237 1.571 1.015 2.431

5 C T 0.4813 0.3539 5.629 0.01766 1.693 1.094 2.62

6 A G 0.3618 0.4773 4.45 0.03489 0.621 0.3984 0.9679

7 C T 0.1118 0.2159 6.329 0.01188 0.4573 0.2462 0.8494

8 G A 0.01899 0.07303 5.391 0.02024 0.2457 0.06869 0.8786

9 T C 0.1812 0.2865 5.168 0.02301 0.5513 0.3287 0.9245

10 C T 0.3375 0.2191 5.922 0.01495 1.816 1.12 2.944

11 A G 0.3312 0.1761 10.57 0.001149 2.316 1.387 3.868

12 T C 0.5 0.3876 4.315 0.03777 1.58 1.025 2.434

13 A C 0.2125 0.1205 4.991 0.02547 1.97 1.079 3.595

14 G A 0.2188 0.125 5.229 0.02221 1.96 1.094 3.512

15 C T 0.2434 0.1529 4.175 0.04103 1.782 1.02 3.114

16 T A 0.5625 0.4326 5.689 0.01707 1.686 1.096 2.594

Table 8A (part i)

SNP INFORMATION

No. CHR Physical Position Aflymetrix ID dbsnp RS ID Gene Gene Relationship 1 1 21752655 SNP_A-4211511 rs950922 ALPL intron

2 1 196366613 SNP_A-1924594 rs16842422 LHX9 downstream 3 3 52468315 SNP_A-2293723 rs9855470 NISCH intron 4 9 136484947 SNP_A-2111568 rs4565533 COL5A1 upstream 5 10 112534645 SNP_A-2001430 rs7077757 6 12 54287594 SNP_A-1920262 rs10747758 OR10P1 upstream 7 12 60812599 SNP_A-1803270 rs11174339 USP15 upstream 8 12 90561550 SNP_A-2288842 rs11106228 BTG1 downstream

9 12 93848520 SNP_A-2208078 rs249153 NDUFA12 downstream 10 16 26556972 SNP_A-2190350 rs12162084 HS3ST4 downstream 11 16 50838191 SNP_A-1801171 rs1345316 12 18 71899460 SNP_A-2264132 rs10871528 LOC284274 upstream 13 19 50114786 SNP_A-2236481 rs4420638 APOC1 downstream 14 20 44997127 SNP_A-2213592 rs6094514 EYA2 intron

Table 8A (part 2)

NETHERLANDS BRAIN BANK POSTMORTEM COHORT

Table 8B

SNP INFORMATION

No. CHR Physical Position Affymefrix ID dbsnp RS ID Gene Gene Relationship

1 1 95331855 SIMP_A-1893969 rs6593670 TMEM56 upstream 2 1 108878546 SNP_A-1904964 rs7533905 FAM102B upstream

3 1 108885823 SNP_A-1865439 rs6666724 FAM102B upstream 4 1 108996418 SNP_A-1834828 rs10157512 C1 orf59 intron 5 2 202181323 SNP_A-2129507 rs12464623 ALS2CR11 iπtroπ 6 3 19530275 SNP_A-2301054 rs6549944 KCNH8 intron 7 12 12146023 SNP_A-2002538 rs4763782 BCL2L14 intron 8 13 74601462 SNP_A-4247955 rs7985837 TBC1 D4 downstream 9 18 7365434 SNP A-1782916 rs9807727 PTPRM upstream

NETHERLANDS BRAIN BANK POSTMORTEM COHORT

No. Mr Mj MAF Affected MAF Unaffected CHISQ OR L95 U95

1 C T 0.2991 0.45 4.866 0.02739 0.5216 0.291 0.9348

2 A C 0.0625 0.15 4.868 0.02736 0.3778 0.1549 0.9214

3 A G 0.06696 0.1667 5.86 0.01549 0.3589 0.1522 0.846

4 C T 0.05804 0.1333 3.918 0.04776 0.4005 0.1578 1.017 5 T C 0.02232 0.1167 10.41 0.001254 0.1729 0.0528 0.5661

6 C T 0.2207 0.35 4.23 0.03971 0.526 0.2835 0.9759

7 C A 0.4509 0.3 4.426 0.03539 1.916 1.039 3.532

8 G T 0.3211 0.4833 5.401 0.02013 0.5056 0.2829 0.9036

9 A G 0.3036 0.1552 5.111 0.02377 2.373 1.104 5.104

2008/061068

Table 8C

.. _ _u SNP INFORMATION

No. CHR Physical Position Affymetπx ID dbsnp RS ID Gene

Gene Relationship

1 2 195582541 SNP_A-4236859 rs7581004

2 4 158770749 SNP_>2123398 rs6833943 GRIA2 downstream

3 10 14166119 SNP_A-2000983 rs898717 FRMD4A intron

4 20 4093021 SNP. A-1885462 rs927091 SMOX introπ

h _l

No