Promoter activity of the beta-amyloid precursor protein gene is negatively modulated by an upstream regulatory element

Alzheimer's disease (AD) is characterized by the aggregation of the amyloid beta-peptide (Abeta) which is generated from a larger beta-amyloid precursor protein (betaAPP). An overexpression of the betaAPP gene in certain areas of the AD brain has been suggested to be an important factor in the neuropathology of AD. Here we have further characterized an upstream regulatory element (URE) located between -2257 and -2234 of the human betaAPP promoter. In addition to its location in the promoter, BLAST search reveals that URE is present in several introns of the betaAPP gene and is also detected in many other genes. For functional studies, two promoter regions were cloned upstream of the reporter gene, chloramphenicol acetyl transferase (CAT): (i) phbetaE-B - the plasmid that contains the human (h) promoter region (-2832 to +101) including URE, and (ii) prhbetaE-B - the plasmid that contains the rhesus (rh) promoter region excluding URE as it lacks a 270 bp region of the hbetaAPP promoter (-2435 to -2165). Transient transfection studies indicate that phbetaE-B displayed significantly less CAT-promoter activity than prhbetaE-B in C6, PC12 and SK-N-SH cells. To determine the role of URE in a heterologous promoter, a pbetaURE construct was made by subcloning URE in an enhancerless promoter vector pCATP. The pbetaURE-CAT construct displayed threefold to fourfold less promoter activity than pCATP when different cell lines were transfected with the plasmids. URE interacts with a novel protein(s) as determined by the electrophoretic mobility shift assay (EMSA). Although the core DNA region of URE resembles with the NF-kB element, URE-binding protein is not related to the NF-kB transcription factor. When EMSA was performed with specific competitors in different cell lines, the labeled URE probe was not competed by the oligonucleotides specific for either the AP3, NF-1 or NF-kB transcription factor. The migration of the URE-protein complex was different from the NF-kB-protein complex in the EMSA gel. A distinct URE-specific nuclear factor was also detected in frontal cortex of a normal human brain. These results suggest that the URE region acts as a repressor element, that the URE-binding protein is not related to the known transcription factors tested, and that the protein is present in astrocytic, neuroblastoma, PC12 cells and in the human brain.