AIM: To investigate the pathogenesis of Aβ42 yielding and new drug targets as well as the possibility of RNA interference (RNAi) technique for treatment of Alzheimer disease (AD). METHODS: Human AD presenilin 1 (PS1)cDNA sequence was obtained from NCBI website. The three sites of RNAi action and one missense control site were selected in PS 1 cDNA through online design of Ambion company. To confirm specificity of these sites, we conducted a BLAST search of the IMAGE EST library. The corresponding double-stranded DNA was used to construct pSilencer 3.l-H1 plasmid, which could transcribe small interference RNA (siRNA). Then, the pSliencer 3.l-H1 plasmids were transfected into CHO/PS1/APP cells with SuperFect transfection reagent. The cells have been transfected with the mutant PS 1 and APP gene of AD. All the CHO/PS1/APP cells with pSliencer 3.1-H1plasmids were screened out using G418. Transcripts of PS 1 gene in CHO/PS1/APP were measured by RT-PCR, the contents of PS 1 peptide and Aβ42 production inside CHO/PS 1/APP cells were examined through Weste blot and the Aβ42 change of secretion by CHO/PS 1/APP was determined with ELISA. RESULTS: The site 3 of PS1 mRNA was inhibited by RNAi after 2 d. The effect was more obvious with the time. The peptide corresponding to PS1 gene and Aβ42 production in CHO/PS 1/APP cells were both reduced after siRNA interfere for 3 d. Aβ42 secretion by CHO/PS 1/APP cells began to reduce on d 3, and reached the most significance on d 5. There was a time-dependent relationship between the transcript of PS 1 gene and the production of Aβ42 with RNAi action. CONCLUSION: PS1is essential for γ-secretase activity. Inhibition of the PS1 can decrease the levels of Aβ42. Some sites of PS1 mRNA,for example, the site 3, may serve as a new drug target and RNAi probably can be used for treatment of AD.