Lenti-GFP (green fluorescent protein) image showing the expression of the siRNA (left) and a microtubule-associated protein 2 (MAP2) immunolabelling illustration (right) showing that the neuronal dendritic structure is preserved after treatment. Image courtesy of E. Masliah, University of California, Davis, USA.

The aberrant processing of amyloid precursor protein (APP), leading to a build-up of amyloid-β peptide (Aβ1–42), is a key molecular event in the pathogenesis of Alzheimer's disease. An increase in the activity of the proteolytic enzyme β-secretase (BACE1), which contributes to this abnormal processing, is associated with neurodegeneration, accumulation of APP products and Alzheimer's disease. Consequently, the inactivation of BACE1 might represent a viable therapeutic strategy for the treatment of this disease. Using the relatively new technique of RNA interference (RNAi) to reduce BACE1 activity in APP transgenic mice, the groups of Masliah and Verma recently demonstrated the efficacy of such an approach.

Lentiviral vectors, chosen for their ability to transduce non-dividing cells such as neurons, were engineered to carry selected 'short hairpin' RNA (shRNA) molecules complementary to target sequences in the open reading frame of human BACE1. Transcription of these results in the expression of small interfering RNA (siRNA) molecules capable of inhibiting the translation of complementary mRNA species, thereby silencing the target gene BACE1.

Following validation of lentisiBACE1 constructs in cell culture, where these were tested for their ability to knockdown the BACE1 protein and reduce levels of Aβ1–42, the most promising construct was selected for trial in a mouse model of Alzheimer's disease.

Mice expressing high levels of double-mutant human APP received lenti-siBACE1 constructs through intracranial injection to the susceptible neuronal populations in the hippocampus. These mice were chosen as they recapitulate many of the features of Alzheimer's disease, including neurodegeneration, memory impairment and amyloid plaques.

As in the cell culture assays, there was a significant reduction in the expression of BACE1 in the hippocampus, which contributed to a significant reduction in APP proteolysis and Aβ plaque formation in the same region. In addition, there was a marked reduction in neurodegeneration in the hippocampus of lenti-siBACE1-injected transgenic mice that was not observed in the neocortex where lenti-siBACE1 constructs had not been injected. Deficits in spatial learning and memory observed in transgenic mice in the Morris water maze were also reversed by siRNA treatment.

Although there are still questions surrounding the accuracy of delivery and long-term expression potential, RNAi is a promising viable therapeutic strategy for the treatment of Alzheimer's disease and other neurodegenerative disorders.