Evaluation of the potential role of the tyrosine kinase, Arg (Abl2) in Alzheimer's disease pathogenesis
MetadataShow full item record
Alzheimer's disease (AD) is the most common cause of dementia, and has become one of the leading causes of death in the United States.;The major pathologic hallmarks of AD are senile plaques and neurofibrillary tangles. Senile plaques are composed of aggregated amyloid beta (Abeta) peptide, which is secreted following sequential cleavage of the transmembrane amyloid precursor protein (APP). Neurofibrillary tangles (NFTs) are composed of aggregates of a hyperphosphorylated microtubule-associated protein tau referred to as paired helical filaments (PHFs). The focus of this work is on two other aspects of AD pathology, namely tau phosphorylation and cell cycle re-entry.;The vast majority of characterized tau phosphorylation sites are serine/threonine sites, particularly proline-directed kinase sites. Only recently has anyone investigated phosphorylation at any of the five tyrosine residues present in the full-length tau protein. Proteomic analyses revealed phosphorylation of tyrosine 394 (Y394) of tau in samples derived from PHF-tau. Additional phosphorylation was also observed at Y197, but not reproduced in other studies. Our laboratory began investigating the Abl family of non-receptor tyrosine kinases as candidates for Y394 phosphorylation just prior to the first report of c-Abl phosphorylation of this site.;The Abl family of tyrosine kinases includes two members, c-Abl (Abl or Abl1) and Arg (Abl-related gene or Abl2). The Abl family kinases share 90% identity in the N-terminal domains, including SH3, SH2, and kinase domains. However, they each possess a unique C-terminus containing actin-binding domains and proline-rich motifs, but divergent localization domains. Abl contains three nuclear localization sequences (NLS) and a nuclear export sequence (NES), while Arg lacks a functional NLS, and thus doesn't localize to the nucleus. Both kinases have been implicated in cytoskeletal organization, oxidative stress response, apoptosis, cell cycle and malignant transformation. In addition, Abl has been implicated in the physiology of APP processing and neuronal Abeta toxicity. While often overlooked, Arg is expressed in the brain at greater levels than Abl, and given its involvement in potentially AD-relevant processes, was selected as the target of our investigation into the role of Abl family kinases in the pathogenesis of AD.;Initial experiments sought to determine the importance of tyrosine phosphorylation of tau in AD and whether Arg phosphorylates tau at the tyrosine sites observed in PHFs. We generated phospho-specific antibodies recognizing tau phosphorylated at the Y394 site. Immunohistochemical staining revealed Y394-phosphorylated tau in NFTs, dystrophic neurites and neuropil threads in affected regions of the AD brain, while nearly absent in normals. In addition, Y394-phosphorylated tau was also present in small granules within hippocampal pyramidal neurons. Using a number of biochemical and genetic strategies, we also demonstrated that Arg phosphorylates tau in vitro at Y394, Y197, and Y310. Additional experiments suggested Y394 as a priming site for additional tau phosphorylation. These findings indicate that both Arg and Abl may be useful targets for studying AD pathogenesis.;In the next step, transgenic mice were constructed to more rigorously determine whether Abl and Arg activition induce neurodegeneration. To this end, activated mutant versions of the human ABL (Ab1PP) and ARG (ArgPP) genes were conditionally expressed in the mouse forebrain. Expression of activated Abl led to substantial increases in astrogliosis, inflammation, and neuronal loss in the CAI region of the hippocampus within 9 weeks of initial expression. Surprisingly, mice expressing a homologous mutant form of Arg failed to show signs of degeneration, even after 6-7 months. Given the paucity of unique domains in these related kinases, phenotypic differences are likely a consequence of substrate specificity or nuclear activity. A comparison study revealed increased JNK activation and cyclin D1 expression in AblPP mice, that was not observed in Arg mice, indicating activation of cellular stress and cell cycle pathways. There is also evidence suggesting that a novel phosphorylation site of Cdk5 may represent a critical difference between the two lines of transgenic mice. Given the findings of this work, it seems that while Arg may participate in many of the same cellular processes, Abl is a better candidate for modeling the neurodegenerative processes of AD. (Abstract shortened by UMI.).