WO1999058711A1

WO1999058711A1 - Method for diagnosing or assessing a predisposition to alzheimer's disease

Info

Publication number: WO1999058711A1
Application number: PCT/AU1999/000362
Authority: WO
Inventors: John Kwok; Peter Robert Schofield
Original assignee: Garvan Institute of Medical Research
Current assignee: Garvan Institute of Medical Research
Priority date: 1998-05-13
Filing date: 1999-05-13
Publication date: 1999-11-18
Anticipated expiration: 2000-11-13
Also published as: AUPP350198A0

Abstract

Methods for diagnosing or assessing predisposition to Alzheimer's disease (AD), especially early onset AD, are disclosed. The methods involve the determination of the presence of an AD-linked marker(s) on chromosome 5 or analysis of allelic variation in relation to an AD susceptibility gene on chromosome 5.

Description

METHOD FOR DIAGNOSING OR ASSESSING A PREDISPOSITION TO

ALZHEIMER'S DISEASE

Field of the Invention: This invention relates to methods for diagnosing or assessing a predisposition to Alzheimer's disease (AD).

Background to the Invention:

Alzheimer's disease (AD) is the most common cause of dementia, estimated to affect 1 to 6% of people over the age of 65 years and between 10 to 20% of those over 80 [1]. AD is distinguished from other forms of dementia by key pathological features in the brain. Firstly, there are a large number of senile plaques in the extracellular spaces between neurones. The plaques have central cores of beta-amyloid peptide (β-A4) fibrils, surrounded by degenerated neurites and glial cells [2]. β-A4 consists of a sequence of hydrophobic amino acids of 39 to 43 amino acids in length [3] derived from proteolytic cleavage of a larger glycoprotein, the amyloid precursor protein (APP) [4]. Secondly, neurofibrillary tangles (NFTs) are found within neurones. The major components of NFTs are paired helical filaments, which in turn are composed of hyperphosphorylated form of TAU, a microtubule associated protein [2]. Finally, there is extensive neuronal loss in the cerebral cortex and hippocampus [2].

The exact contribution of each neuropathological feature to the clinical symptoms of AD is unclear [reviewed in 5]. However, several studies have suggested that the β-A4 deposits can be directly neurotoxic, in part through the generation of free radicals [5], or that β-A4 can disrupt ionic homeostasis and lead to severe effects on cellular processes and induction of neuronal cell death [5]. The amyloid cascade hypothesis postulates that the deposition of β-A4 is the central causative event in AD and that the NFTs, cell death and dementia follow as a direct result of this deposition [6]. The amyloid cascade hypothesis predicts that mutations in genes, which lead to the overproduction of APP or subsequent mismetabolism, would underlie the genetic basis of AD.

Familial Alzheimer's Disease is a dominantly inherited disorder that presents prior to 65 years of age and accounts for 10% of all AD cases. However, since the neuropathology of familial AD patients is indistinguishable from that of sporadic AD, the study of these familial forms of AD has led to a greater understanding of AD. Linkage analysis has been an important tool in the elucidation of the genetic events involved in familial AD. Chromosome 21 - APP sene

Genetic analysis detected linkage between an AD locus and several markers on the long arm of chromosome 21 containing the APP gene [7]. Comparisons of the APP gene nucleotide sequence between normal and affected individuals revealed several missense mutations which are found only in AD patients. All the mutations appear to cluster within or adjacent to the sequence which encodes β-A4 [reviewed in 8]. Each mutation has a subtly different biochemical effect, either altering the nature of the β-A4 sequence or the metabolism of APP which increases production of the insoluble β-A442 isoforni. Ultimately, all APP mutations result in an increase in the rate of β-A4 deposition.

Chromosome 14 and 1 - Presenilin genes

A major locus responsible for AD was shown to map to the long arm of chromosome 14 [9]. In 1995, Sherrington et al. [10] reported the positional cloning of the chromosome 14 AD gene, known as presenilin-1 (PS-1). The homologue of PS-1, known as presenilin-2 (PS-2) on chromosome lq31-42, was shown to be the causative locus in a series of Volga-German pedigrees [11]. Over thirty-seven distinct point mutations, and a splice-site mutation which results in the deletion of a small portion of the protein, have been identified in the presenilin genes [12]. Considerable evidence suggests that all known AD genetic loci may act in the same amyloid cascade pathway as mentioned above. For example, clonal cell lines and transgenic mice which express mutant forms of the presenilin genes also have elevated production of the amyloidogenic form of the amyloid peptide, β-A442 [13]. Chromosome 19 - Apolipoprotein E sene

Linkage analyses of late onset AD has defined a region on chromosome 19ql3.2 [14] which contains the candidate gene apolipoprotein E (Apo E) [15]. The Apo E gene contains as three co-dominant alleles, e2 (allele frequency of .08), e3 {.77) and e4 (.15) [16]. Association studies have shown that the frequency of the e4 allele (estimates ranging from .34 to .50) was significantly increased in AD patients (both familial late-onset and sporadic) [16]. Apo E may play an important role as a modifier gene, interacting with the APP gene or other AD genetic loci to produce the AD phenotype. The e2 allele confers protection against development of late-onset AD [17].

In work leading to the present invention, a panel of sixteen familial AD families were screened for missense mutations in the three AD genes and their ApoE genotype status determined by PCR-restriction enzyme digest assay [20]. Two families, Aus-1 and Aus-2 had mutations in the APP gene (Table 1), one of which is a novel mutation (Leu723Pro). Causative loci in three pedigrees were attributable to known mutations in the PS-1 gene. In a fourth pedigree, a novel PS-1 mutation was identified in exon 7 (M233T), which is homologous to a pathogenic PS-2 mutation (M239V). In one early- onset case, another novel PS-1 mutation was identified in exon 8 (R278T). Two families have late-onset AD associated with an ApoE e4/e4 genotype. Of the six remaining families and the other early-onset case, none have missense mutations in the coding region of APP or presenilin genes, or are associated with ApoE alleles. Thus, the results of the screening suggested that other novel loci, which have now been mapped to chromosome 5, are involved in the etiology of familial AD cases and possibly sporadic AD cases as well.

Summary of the Invention:

In a first aspect, the present invention provides a method of diagnosing or assessing an individual's predisposition to Alzheimer's disease (AD), especially early onset AD, comprising a step of determining the presence of a AD-linked marker(s) on chromosome 5.

Preferably, the method comprises determining the presence of a AD- linked marker(s) on the short arm of chromosome 5. More preferably, the method comprises determining the presence of an AD-linked marker(s) between the chromosomal loci D5S1987 and D5S648. The method may involve the determination of a single AD-linked marker but, more preferably, involves the determination of the presence of at least two AD-linked markers. Preferably, the AD-linked marker(s) are microsatellite marker(s). More preferably, the AD-linked marker(s) are selected from D5S416, D5S1987 and D5S648 and combinations thereof. Preferably, the step of determining the presence of a AD-linked marker(s) comprises PCR amplification and gel electrophoresis to determine the presence of a specific microsatellite allele for a given marker.

In a second aspect, the present invention provides a method of diagnosing or assessing an individual's predisposition to AD, especially early onset AD, comprising a step of analysing allelic variation in relation to an AD susceptibility gene on chromosome 5.

Preferably, the method of the second aspect comprises analysing allelic variation in relation to an AD susceptibility gene on the short arm of chromosome 5. More preferably, the method of the second aspect comprises analysing allelic variation between the chromosomal loci D5S1987 and D5S648.

Detailed Disclosure of the Invention: The method of the first and second aspects of the invention may be used with similar methods for other AD-linked markers and, therefore, may provide a valuable clinical test. As such, the method of the first and second aspects may be used with methods involving determining the presence of the novel APP mutation, Leu 723 Pro and/or novel PS-1 mutations (i.e. M233T and R278T) and PS-2 mutation (i.e. M239V) described above.

The present invention may also be used to guide the cloning of the AD susceptibility gene on chromosome 5 and the identification of the allelic variation in the susceptibility gene that results in the allelic variant of the gene providing a higher relative risk to carriers. Testing directly for this allelic variation will be the preferred test. Cloning of the susceptibility gene on chromosome 5 may also allow identification of agents that block or enhance the action of this gene. Such agents are likely to be of value in the clinical treatment of AD.

Thus, in a further aspect, the present invention provides an isolated polynucleotide molecule, comprising a nucleotide sequence substantially corresponding to the nucleotide sequence between the human chromosomal loci D5S1987 and D5S648 and which includes an AD susceptibility gene.

Preferably, the AD susceptibility gene is an early onset AD susceptibility gene. By the term "early onset AD" we refer to AD wherein the mean age of onset is within the range of about 30 to about 50 years of age. The terms "comprise", "comprises" and "comprising" as used throughout the specification are intended to refer to the inclusion of a stated step, component or feature or group of steps, components or features with or without the inclusion of a further step, component or feature or group of steps, components or features.

The invention will hereinafter be described with reference to the following non-limiting example and accompanying figures.

Brief description of the accompanying figures: Figure 1 provides the results of haplotype analysis of TAS-1 pedigree.

Disease haplotype spanning 5 markers was determined using the least number of meiotic recombinations (X) to explain allele presentation. Present ages are shown in brackets.

Figure 2(a) and (b) provides the results of a Three-Point linkage analysis of candidate chromosomal markers. Markers flanking the minimal haplotype region are indicated by vertical arrows.

Figure 3 diagrammatically shows the position of the chromosomal 5ρ markers used in linkage analyses for the TAS-1 pedigree.

Figure 4 provides the results of haplotype analysis of KK pedigree. The analysis involved three individuals (III.4, III.5 and III.6), and showed that a recombination occurred between markers D5S630 and D5S1987.

Example 1: Identification of TAS-1 AD Pedigree

Unlinked Familial AD Pedigrees Seven pedigrees in the panel of sixteen familial AD families (Table 1) are not linked to known AD genes and are available for linkage analysis to detect novel AD genetic loci. The largest pedigree (TAS-1) consists of a multi-generational family where AD is inherited as a Mendelian autosomal dominant trait (see Fig 1). There are a total of thirty- two individuals of which nine members are affected. AD was confirmed by autopsy in two cases (III.13 and III.17 in Fig 1). Both cases had cortical atrophy, numerous senile plaques in the frontal and temporal cortex and neurofibrillary tangles in the hippocampus as well as congophilic angiopathy, a common feature of familial AD. Table 1 : AD Pedigrees

Family Mean Age Pathology Mutation/ of Onset Confirmed Risk Factor

ΛUS-1 49 Yes APP V717I

AUS-2 56 No APP L723P

EOAD-1 45 Yes (a) PS-1 Δex 10

EOAD-3 50 No(a) PS-1 ΔexlO

EOAD-6 39 Yes PS-1 P264L

COL 47 No PS-1 G280A

PERTH-2 43 No PS-1 M233T

P-2 37 Yes (a) PS-1 A278T

BP-4 66 No Apo E e4/e4

LOAD-3 67 No Apo E e4/e4

TAS-1 46 Yes -

LYN 50 No -

KK 63 No -

PERTH-1 35 Yes -

PERTH-3 57 No -

P-l 36 Yes -

(a) Spastic paraparesis also found in this pedigree

Linkage Analysis of TAS-1 Pedigree Simulation linkage analysis was performed using the SLINK program on TAS-1 pedigree to determine the power of this pedigree to detect linkage.

The linkage model assumed that markers had 6 alleles of equal frequency.

Using 10 members of the pedigree (3 affected and 7 apparently unaffected), simulation with 1000 replicates produced a maximum expected LOD score of 2.56 and an average expected LOD score of 0.89 at zero recombination fraction.

In order to detect the AD locus segregating within TAS-1 pedigree, the ten members were used for a genome-wide screen using highly polymorphic microsatellite markers. To screen the entire genome, 358 fluorescent microsatellite markers placed at an approximate spacing of 10 cM, were used

(ABI Linkage mapping Set 1, Perkin Elmer). Genotyping data was analysed using the LINKAGE program [23] assuming dominant inheritance of the disease locus, a disease allele frequency of 0.001 and a phenocopy rate of 0.0001. Age-dependent liability classes were assigned from the distribution of the known age of onset in the family. A penetrance of 95 percent was assumed at the mean age of onset plus two standard deviations (46.3 years + 9 years).

Exclusion of Known Familial AD Genes

The coding region of candidate AD genes (APP, PS-1, PS-2, ApoE) of affected members of TAS-1 pedigree had been previously screened by direct PCR-cycle sequencing, thus enabling the determination of the Apo E status of each member. However, since linkage analysis would allow further full exclusion of the four known AD genes, the candidate chromosomal regions of Iq31-lq32 (PS-2), 14q24.3 (PS-1), 19ql3.1 (Apo E) and 21q21 (APP) were screened. Four markers which span each chromosomal region were analysed. Two-point analysis (MLINK) gave no support of linkage with markers flanking known genes: D1S413, D1S249, D1S425, D1S213, D3S1566, D3S1603, D3S1271, D3S1278, D14S258, D14S74, D14S68, D14S280, D17S798, D17S791, D17S787, D17S808, D19S221, D19S226, D19S220, D19S414, D21S1256, D21S1253, D21S263, D21S1252. All LOD scores were negative for each chromosomal region, with LOD scores ranging from -0.22 to -4.8 at zero recombination fraction. This analysis indicated that the pathogenic mutation within the TAS-1 pedigree does not involve the known AD genes. A Novel Candidate AD Chromosomal Region Two-point linkage analysis of the genotype data for the entire panel of microsatellite markers revealed eight LOD scores greater than 1.00. Two-point analysis (MLINK):

- fully penetrant dominant disease model with 7 liability classes and disease allele freq of 0.001. - liability classes: 1) 0-5 yrs - 0.01

2) 6-llyrs - 0.08

3) 12-19yrs - 0.22

4) 20-35yrs - 0.46

5) 36-45yrs - 0.71 6) 46-54yrs - 0.91

7) >55yrs - 0.95. 8

- 8 loci gave lod scores over 1.00: D2S125, D5S416, D5S419, D5S424, D14S283, D15S165, D18S64, D20S115.

- Chr5p: 3 markers gave lods > 1.0, D5S416 giving a max lod = 1.12, 3 other markers gave lods of 0.77-0.8 ; D5S406, D5S630, D5S647. - repeat of MLINK with genotype for sample 01 gave lod = 1.42 for

D5S416.

Multi-point analysis (LINKMAP) of above markers with their flanking markers gave some positive lods but none were significant and none of the flanking markers had positive lod scores in MLINK: - Chr2: D2S338 - D2S125 (most distal Chr2 marker) gave max. lod = 0.

- Chrl4: D14S261 - 7.1- D14S283 - 12.8 - D14S80 gave max. lod = 1.0 between D14S283 and D14S80, but D14S80 gave lod = -1.16 in MLINK.

- Chrlδ: D15S128 - 14.7 - D15S165 - 30.8 - D15S117 gave max. lod = 1.2 between D15S165 and D15S117, but D15S117 gave lod = -2.32 in

MLINK.

- Chr 18: D18S474 - 13.8 - D18S64 - 12.0 - D18S68 gave max. lod = 0.5.

- Chr 20: D20S95 - 4.9 - D20S115 - 11.9 - D20S189 gave max. lod = 0.7. However, further examination of these regions revealed that most of them are probably weak false positives as LOD scores from flanking markers were zero or negative. Only one region consistently gave LOD scores of greater than 1.00. Two-point analysis of the candidate chromosomal arm (i.e. the short arm of chromosome 5) revealed three markers with a LOD scores of greater than 1.00. At one end, two adjacent markers (D5S416 and D5S419) gave LOD scores of 1.42 and 1.24 at zero recombination fraction (Table 2).

The more centromeric marker D5S424, gave a LOD score of 1.39. Three point analysis (using LINKMAP program of the LINKAGE package [23]) of the entire chromosomal arm gave a peak LOD score of 1.33 around the first two markers D5S416 and D5S419 (Fig 2a). In order to clarify the genotype data, a haplotype analysis was performed using markers on the candidate chromosomal arm. A disease haplotype spanning five markers was found in all three affected individuals (Fig 1). However, the putative disease haplotype is also found in one apparent unaffected (11.12). This explains the comparatively low LOD score obtained for this pedigree since the affection status of 11.12 places an obligate meiotic recombination between the disease locus and all markers within the candidate chromosomal region. MLINK analysis of Chr5p markers without sample 08 gave max lod score = 1.86 at D5S416 (Table 3). Multi-point gave max lod = 2.14 between D5S630 and D5S416.

Reducing penetrance to 60% and including sample 08 did not alter the two-point analysis.

If II.12's status is changed to affected, the maximum LOD score increases to 2.33 at D5S416 (Table 4), a value which is 92% of the maximum simulated LOD score for the pedigree. Similarly, as shown in Figure 2b, the multipoint LOD score increases to 2.49 between D5S630 and D5S416. Thus, individual 11.12 may be a confounding 'escapee' who has inherited a disease haplotype or the disease gene which may not be fully penetrant [24, 25].

Table 2 : Chromosome 5p Markers

Cm θ=0.0 θ=0.1 θ=0.2 θ=0.3 θ=0.4 θ=0.5

D5S406 0 0.78 0.68 0.58 0.44 0.25 0.0

D5S630 8 0.8 0.88 0.69 0.43 0.15 0.0

D5S416 17 1.42 1.29 1.05 0.75 0.39 0.0

D5S419 30 1.24 1.08 0.88 0.64 0.33 0.0

D5S426 43 -0.55 -0.35 -0.19 -0.08 -0.02 0.0

D5S418 50 -1.26 0.85 0.79 0.59 0.3 0.0

D5S407 57 -0.97 0.94 0.85 0.62 0.31 0.0

D5S647 63 0.77 0.69 0.57 0.42 0.23 0.0

D5S424 73 1.39 1.08 0.79 0.51 0.25 0.0 Multi-point analysis - max lod = 1.33 between D5S416 and D5S419

10

Table 3: < Chromosome 5 Markers with sample 08 removed θ = 0.0 θ = 0.1 θ = 0.2 θ = 0.3 θ = 0.4 θ = 0.5

D5S406 0.69 0.67 0.58 0.42 0.22 0.0

D5S630 1.61 1.22 0.81 0.4 0.06 0.0

D5S416 1.86 1.48 1.09 0.69 0.3 0.0

D5S419 1.54 1.24 0.93 0.61 0.28 0.0

D5S426 -0.34 -0.22 -0.12 -0.05 -0.01 0.0

D5S418 -0.79 0.47 0.46 0.29 0.09 0.0

D5S407 -0.95 0.64 0.62 0.45 0.21 0.0

D5S647 0.65 0.62 0.53 0.39 0.21 0.0

D5S424 1.36 1.05 0.75 0.46 0.19 0.0

Multi-point max lod = 2.14 between D5S630 and D5S416

Table 4 : Chromosome 5p Markers with sample 08 as affed ted θ = 0.0 θ = 0.1 θ = 0.2 θ = 0.3 θ = 0.4 θ = 0.5

D5S406 0.65 0.73 0.65 0.49 0.26 0.0

D5S630 2.12 1.68 1.2 0.68 0.21 0.0

D5S416 2.33 1.9 1.43 0.94 0.44 0.0

D5S419 1.68 1.4 1.08 0.73 0.36 0.0

D5S426 -0.03 -0.01 0.0 0.01 0.0 0.0

D5S418 -0.93 0.65 0.7 0.55 0.3 0.0

D5S407 -1.65 0.5 0.65 0.53 0.29 0.0

D5S647 0.68 0.66 0.58 0.44 0.24 0.0

D5S424 1.08 0.95 0.75 0.52 0.25 0.0

Multi-point max lod = 2.49 between D5S630 and D5S416

Example 2: Identification of KK Pedigree

Haplotype analysis of KK Pedigree The KK pedigree represents a family with heritable clinically- diagnosed early onset AD. This family does not, however, possess any detectable mutations in the known familial AD genes (APP, PS-1, PS-2, Apo E). Haplotype analysis was performed in the manner described for the TAS-1 pedigree, on three individuals of the KK pedigree. 11

Haplotype analysis (Fig 4) showed that a recombination occured between the markers D5S630 and D5S1987 in one unaffected individual (III6). When combining these results with the results obtained from the TAS- 1 family, it can be determined that the minimal disease haplotype corresponds to a 17.6 cM region defined by the markers D5S1987 and

D5S648.

Example 3: Positional cloning and functional studies of the novel AD gene

Positional cloning of the novel AD gene At present, the candidate disease haplotype spans a region of approximately 17.6 cM. However, analysis of other pedigrees should significantly refine this region. A physical contig of YAC, BAC and Pi clones will be constructed across the minimal disease haplotype region. This will facilitate the mapping of various novel and previously identified coding sequences. Techniques for detecting coding sequences of genes in defined subchromosomal regions, such as exon trapping and direct cDNA selection [26], will be used to detect novel coding sequences. Moreover, several promising candidates genes have already been mapped to novel AD region, including genes which code for ligands involved in neuronal survival and cell death. Candidate genes will be screened for pathogenic mutations which are found only in the affected individuals of each pedigree [27]. Functional studies on the novel AD gene

Once the gene has been isolated, biochemical and molecular studies can be employed to determine whether the effects of the pathogenic mutation follow the amyloid cascade pathway. This may be done by using biochemical analyses previously established by the present applicant For example, to determine whether the APP Leu723Pro mutation affects the production of β-A4, various forms of APP cDNAs were expressed in CHO cells. Stable transfectants carrying the normal and mutant versions of APP cDNAs were selected and APP cleavage products immunoprecipated with antibodies which specifically recognise various isoforms of β-A4 [28]. This demonstrated that the expression of the Leu723Pro mutation results in a twofold increase in production of β-A4₄₂. This two-fold increase is also seen for the London or Swedish APP mutations and is comparable with other studies [8, 29], 12

The London APP mutation, as well as several presenilin mutations have been shown to induce apoptosis in transfected cells via G-protein pathways [30, 31]. We have demonstrated that the Leu723Pro mutation is also capable of inducing apoptosis in transiently transfected CHO cells as determined by the distinctive cleavage of genomic DNA into nucleosomal length (Fig 3b). When levels of apoptosis were quantified using a sandwich ELISA assay, the Leu723Pro mutation was able to induce levels of apoptosis comparable to the London mutation. Such techniques will be used to analyse the biochemical pathways of the pathogenic mutation in the novel AD gene.

Coexpression of the novel AD gene with the known wild type and mutant AD genes in cell lines, or (in future studies) in transgenic mice, will allow for the elucidation of the normal biological role of the gene. Immunocytochemical studies using antipeptide antibodies should allow for the localisation of the gene product to tissue and cellular specific areas as well as determined its relative topology if located in a cell membrane. The presence of conserved sequence motifs within the gene product and/or homologous genes in other species should also give an indication as to the function of the gene product. Hypotheses of AD gene function can then be tested using a series of analyses to test for possible involvement in a diverse range of cellular events including ligand binding, channel activity, phosphorylation, proteolytic cleavage and secondary messenger activation.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive. 13

References:

1 Clark RF and Goate AM (1993) Molecular genetics of Alzheimer's disease. Arch. Neurol 50: 1164-1172.

2 Iqbal K et al. (Eds) (1995) Research advances in Alzheimer's Disease and related disorders. John Wiley and Sons UK.

3 Masters C et al. (1985) Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc NatlAcad Sci USA 82: 4245-4259.

4 Kang J et al. (1987) The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor. Nature (London) 325: 733-736. 5 Iverson LL et al. (1995) The toxicity in vitro of β-amyloid protein.

Biochem J 311: 1-16.

6 Hardy JA and Higgins GA (1992) Alzheimer's disease: The amyloid cascade hypothesis. Science 256: 184-185.

7 St George-Hyslop PH et al. (1987) The genetic defect causing familial Alzheimer's disease maps on chromosome 21. Science 235: 885-890.

8 Hardy J (1997. Amyloid, the presenilins and Alzheimer's disease. Trends Neurosci 20: 154- 159.

9 Schellenberg GD et al. (1992) Genetic linkage evidence for a familial Alzheimer's disease locus on chromosome 14. Science 258: 668-671. 10 Sherrington R et al. (1995) Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease. Nature (London) 375:

754-760.

11 Levy-Lehad E et al. (1995) Candidate gene for the chromosome 1 familial Alzheimer's disease locus. Science 269: 973-977. 12 Dasilva HAR et al. (1997) Presenilins and early-onset familial

Alzheimer's disease. NeuroReport 8:1-12.

13 Citron M et al. (1997) Mutant presenilins of Alzheimer's disease increase production of 42-residue amyloid beta-protein in both transfected cells and transgenic mice. Nature Med 3: 67-72. 14 Pericak- Vance MA et al. (1991) Linkage studies in familial Alzheimer's disease: Evidence for chromosome 19 linkage. Am J Hum Genet 48: 1034- 1050.

15 Das HK et al. (1985) Isolation, characterisation, and mapping to chromosome 19 of the human apolipoprotein E gene. J Biol Chem 260: 6240- 6247. 14

16 Tsai M et al. (1994) Apolipoprotein E: Risk factor for Alzheimer's disease. Am J Hum Genet 54: 643-649.

17 Corder EH et al. (1994) Protective effect of apolipoprotein E type 2 allele for late-onset Alzheimer's disease. Nature Genet. 7: 180-184. 18 Hutton M et al. (1996) Complete analysis of presenilin 1 gene in early onset Alzheimer's disease. NeuroReport 7: 801-805 .

19 Campion D et al. (1996) Mutations of the presenilin I gene in families with early-onset Alzheimer's disease. NeuroReport 7: 1582-1584

20 Hixson JE et al. (1990) Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with Hhal. J Lipid Res 31:545-548.

21 Hazan Jei al. (1993) Autosomal dominant familial spastic paraplegia is genetically heterogenous and one locus maps to chromosome 14q. Nature Genet 5: 163-167

22 Lizcano-Gil L.A. et al. (1997) Association of late-onset spastic paraparesis and dementia. Am J Hum Genet 68: 1-6.

23 Lathrop GM et al. (1984) Strategies for multilocus linkage analysis in humans. Proc NatlAcadAci USA 81: 3443-3446.

24 Rossor MN et al. (1996) Incomplete penetrance of familial Alzheimer's disease in a pedigree with a novel presenilin-1 gene mutation. Lancet 347: 1560.

25 St George-Hyslop P et al. (1994) Alzheimer's disease and possible gene interaction. Science 263: 537.

26 Gardiner K et al. (1995) Getting the message: identifying transcribed sequences. Trends Genet 11: 77-79. 27 Grompe M. (1993) The rapid detection of unknown mutations in nucleic acids. Nature Genet 5: 111-117.

28 Fuller S. et al. (1995) Intracellular production of βA4 amyloid of Alzheimer's disease. Biochemistry 34: 8091-8098.

29 Suzuki N.ef. al. (1994) An increased percentage of long amyloid b- protein secreted by familial amyloid β protein precursor (βAPP717) mutant. Science 264: 1336-1340.

30 Yamatsuji T et al. (1996) Expression of V642 APP mutant causes cellular apoptosis as Alzheimer trait-linked phenotype. EMBOJ 15: 498-509.

31 Wolozin B et al. (1996) Participation of presenilin 2 in apoptosis. Science 274: 1710-1713

Claims

15Claims:-

1. A method of diagnosing or assessing an individual's predisposition to Alzheimer's disease (AD), comprising a step of determining the presence of an AD-linked marker(s) on chromosome 5.

2. A method according to claim 1, which method comprises determining the presence of an AD-linked marker(s) on the short arm of chromosome 5.

3. A method according to claim 1, which method comprises determining the presence of an AD-linked marker(s) between the chromosomal loci D5S1987 and D5S648.

4. A method according to any one of the preceding claims, wherein the

AD-linked marker(s) is/are microsatellite marker(s).

5. A method according to claim 4, wherein the AD-linked marker(s) is/are selected from D5S416, D5Sl987and D5S648 and combinations thereof.

6. A method according to any one of the preceding claims, wherein the step of determining the presence of an AD-linked marker(s) comprises PCR amplification and gel electrophoresis.

7. A method of diagnosing or assessing an individual's predisposition to

Alzheimer's disease (AD), comprising a step of analysing allelic variation in relation to an AD susceptibility gene on chromosome 5.

8. A method according to claim 7, which method comprises analysing allelic variation in relation to an AD susceptibility gene on the short arm of chromosome 5.

9. A method according to claim 7, which method comprises analysing allelic variation in relation to an AD susceptibility gene between the chromosomal loci D5S1987 and D5S648. 16

10. A method according to any one of claims 7 to 9, wherein the step of analysing allelic variation comprises determining microsatellite alleles by PCR amplification and gel electrophoresis.

11. A method according to any one of the preceding claims, wherein the Alzheimer's disease is an early onset Alzheimer's disease.