NOVEL APPROACH FOR EXPRESSION CLONING OF APP SECRETASES

点击翻译按钮获取中文摘要

The amyloid beta protein is an approximately 4 kD secreted protein normally found in human plasma and cerebrospinal fluid. Abeta is invariably deposited as insoluble amyloid fibrils in the brains of patients with Alzheimer's disease (AD), and there is increasing evidence that Abeta deposition plays an important role in AD pathogenesis. Abeta is released from the larger amyloid beta protein precursor (betaAPP) through cleavage on the amino and carboxyl side of Abeta by proteolytic activities referred to as beta and gamma secretase, respectively. betaAPP is also cleaved at Abeta 16 by a third secretase, a secretase, which may serve to prevent amyloid deposition by bisecting the Abeta peptide. Despite their important role in Alzheimer's disease, none of the betaAPP secretases has yet been isolated or cloned. Here, I propose to isolate betaAPP secretase cDNAs using a system in which the fission yeast Schizosaccharomyces pombe expresses a chimeric cDNA construct consisting of a secreted derivative of the human placental alkaline phosphatase (SEAP) linked to the C-terminal 105 amino acids of betaAPP (SEAP-APP). I have shown that the secretases in mammalian cells readily process SEAP-APP so that Abeta and secreted SEAP-containing derivatives (sSEAP-APP) are efficiently secreted. In contrast, S. pombe expressing SEAP-APP does not produce secreted derivatives indicating that it lacks endogenous secretase activity. In my initial studies, S. pombe cells expressing SEAP-APP were transformed with a human cDNA (HeLa cell) expression library, and 80,000 transformants, each of which could contain several cDNAs, were screened for their ability to produce secreted derivatives. Five transformants contained putative secretase cDNAs that resulted in the production of sSEAP-APP. Thus complementation in S. pombe appears to be an effective and convenient method to isolate putative human secretase cDNAs. In this application, I propose to employ this system and modifications of it to isolate as many independent candidate betaAPP-secretase cDNAs as possible. I will then determine which of these cDNAs encode authentic betaAPP secretases by determining whether coexpression with authentic betaAPP in S. pombe results in cleavage at the predicted site(s).