Alzheimer's disease (AD) is a common dementia with characteristic neuropathology. Although a number of animal models have been proposed which exhibit some of the neurochemical and pathological abnormalities seen in AD, there is no model which can be utilized to definitively study the pathogenesis and treatment of this disease. We propose to develop a transgenic mouse model of AD in which selected chromosomal regions, thought to contain the disease-causing gene(s), are dissected from patients with familial Alzheimer's disease (FAD) and inserted into the germ line of recipient mice (""""""""transomic mice""""""""). We have developed the methodology to precisely microdissect chromosome bands and microinject the dissected genomic DNA, in its natural context, into fertilized mouse ova. The regions to be microdissected from chromosomes from tissue culture cells are 21q11-q22 in patients with early onset FAD (with and without an amyloid precursor protein (APP) 717 mutation) and 19q13 from patients with late onset FAD. Experiments will be performed in which dissected fragments from chromosome 21 will contain both the putative FAD locus and the APP gene, as well as experiments in which the FAD locus and APP gene are dissected separately and microinjected into mouse embryos. The identical regions will be dissected from normal individuals as a control for gene dosage. 21q11-q22 will also be dissected from Dutch patients with hereditary cerebral hemorrhage with amyloidosis (HCHA-D) to determine if different APP mutations induce varying phenotypes in transgenic mice. Live births from infected embryos will be screened for human sequences using a PCR method to detect human specific Alu sequences, and heterozygous and homozygous transomic mouse strains developed by selective matings. Brains of transomic mice will be examined in detail for neuropathologic lesions, particularly senile plaques and neurofibrillary tangles, and by immunohistochemistry for astrocytosis and beta amyloid deposition. DNA from transomic mouse strains exhibiting neuropathology will be studied with respect to the organization and expression of specific human sequences (particularly the APP gene) using filter hybridization techniques as well as determining the location of human sequences in the transomic mouse genome using in situ hybridization. In summary, this methodology has the potential of establishing a legitimate mouse model for AD and to unambiguously identify regions of the human chromosome map which cause specific neuropathology in FAD.
