Silencing of ApoE with Divalent siRNAs Drives Activation of Immune Clearance Pathways and Improves Amyloid Pathology in Mouse Models of Alzheimer’s Disease

Abstract

The most common genetic risk factor for late-onset Alzheimer's disease (AD) is the APOE4 allele, with evidence for gain- and loss-of-function mechanisms. ApoE knockout in mice abrogates AD phenotypes but causes severe atherosclerosis due to the role of liver ApoE in cholesterol homeostasis. Previous attempts to inhibit brain-specific ApoE with anti-sense oligonucleotides only modestly reduced ApoE expression and had no effect on amyloid burden in adult AD mice. Here, we optimized a divalent small interfering RNA (di-siRNA) to selectively and potently silence ApoE in the brain. Silencing brain ApoE in AD mice significantly reduced amyloid plaque formation without affecting systemic cholesterol levels, confirming that brain and liver APOE pools are spatially and functionally distinct. Mechanistically, APOE appears to be a scaffold for beta-amyloid aggregation that limits clearance by microglia. Di-siRNAs from this study can be taken to pre-clinical and clinical trials to accelerate development of AD-modifying therapies and establish siRNA-based modulation of ApoE as a viable path towards therapeutic development.

A combinatorial approach for achieving CNS-selective RNAi

“…Although in the short-term, partial reduction of liver Apoe by di-siRNA does not translate into detectable serum cholesterol changes in wild-type or AD mouse models ( 36 ) and can be controlled by dose reduction, long-term partial reduction may have a more significant impact. Thus, we sought to devise a method to block any silencing of APOE in the liver.…”

A programmable dual-targeting di-valent siRNA scaffold supports potent multi-gene modulation in the central nervous system

“…Branched di-valent siRNA has been shown to achieve sustained (up to 6 months), efficacious silencing of gene targets in the CNS (1, 18, 22, 23). To test the effect of di-valent siRNA configuration on silencing efficacy, we synthesized single-targeting branched and linear configurations against a non-targeting control (NTC) sequence an Msh3 sequence, or an Htt sequence (Figure 1).…”

“…To test whether the dual-targeting divalent siRNA scaffold could be reprogrammed to silence other disease-relevant genes, dual-targeting divalent siRNAs and controls were synthesized using two previously validated siRNA sequences, Apoe and Jak1 (Figure 6A) (22, 26). Apolipoprotein E (APOE) is a well-characterized, leading risk factor for AD (27), and Janus kinase 1 (JAK1) is a major signal transduction mediator of many inflammatory pathways, including the interferon pathway (26, 28).…”

Di-valent siRNA Mediated Silencing of MSH3 Blocks Somatic Repeat Expansion in Mouse Models of Huntington’s Disease

“…Di-valent siRNAs, when administered by CSF infusion, achieve 30-40% retention of the injected dose in the CNS ( 31 ) ( 57 ), where it broadly distributes ( 30 ) –to brain structures highly affected in HD –and can silence a gene target for up to six months. Importantly, di-valent siRNAs show limited accumulation in liver and kidney and no detectable presence in the colon, a tissue in which MSH3 silencing is associated with cancer ( 58, 59 ).…”

“…The inherent sequence specificity of siRNAs and their duration of effect provide a powerful therapeutic paradigm for treating neurodegenerative disorders. The discovery and development of di-valent ( 30 ) and lipophilic siRNA ( 62, 63 ) has opened siRNA drugs to CNS indications; several compounds are now in clinical or late preclinical development ( 58 ).…”