Gene Therapy for Huntington’s Disease - Gene Delivery and Therapy for Neurological Disorders - page 132

III promoters [ 86 ]. Most importantly, cloning mature inhibitory
sequences into miRNA scaffolds has been shown to mitigate toxic-
ity associated with shRNA.
Artificial miRNAs demonstrate improved safety profile over
shRNAs when the same inhibitory sequences are utilized without
compromising their silencing efficacy. In a CAG140 knockin model
of HD, two out of three shRNAs directed against human and
mouse Htt mRNA transcripts induced substantial neurotoxicity
following AAV-mediated expression in the mouse striatum, despite
exhibiting similar levels of HTT knockdown [ 87 ]. A mismatch
control shRNA (sh2.4mis) also produced toxicity in the absence of
HTT suppression, indicating that toxic effects are independent of
silencing activity. Furthermore, striatal toxicity correlated with
increased levels of the resultant mature RNA sequences. Cloning
two sequences which are toxic in the context of shRNA (sh2.4 and
sh2.4mis) into an artificial miRNA backbone (mi2.4 and mi2.4mis)
attenuated neurotoxicity considerably and reduced expression of
the mature RNA products in wild-type mice. Therefore, toxicity
associated with shRNA may be partly attributed to high levels of
mature inhibitory RNAs. In spite of the difference in expression,
the degree of Htt mRNA silencing was comparable between sh2.4
and mi2.4. The safety of sh2.4 and mi2.4 was further evaluated in
a subsequent study [ 88 ], where they were delivered to N171-82Q
HD mice striatum via AAV vectors. In addition to high mature
inhibitory RNA expression, there was a small build-up of unpro-
cessed shRNA precursors in some sh2.4-treated striata, suggesting
possible saturation of the endogenous miRNA biogenesis pathway.
The therapeutic efficacy of mi2.4, which targets exon 2 of human
and mouse Htt transcripts, was also investigated in the same study.
At 7 and 11 weeks posttreatment, N171-82Q mice which received
AAV-mi2.4 displayed improved rotarod performance compared to
controls. The analysis of striatal tissue revealed a 75 % decrease in
mutant human and wild-type mouse Htt mRNA levels at 13 weeks
after treatment. Nonallele-specific silencing led to changes in tran-
scriptional profile associated with loss of normal Htt function simi-
lar to that observed by Drouet et al. in their lentiviral shRNA study
[ 75 ]. Despite these alterations, treatment with mi2.4 reduced the
mortality of mice with more than 75 % survival at 20 weeks of age,
in contrast to only 45 % survival in control HD mice.
Compared to rodents, non-human primate models offer the
opportunity for comprehensive evaluation of safety and efficacy
because they are closer to humans in size, physiology, and behav-
ioral capabilities [ 89 ]. An inhibitory sequence designed to silence
mouse, rhesus, and human HTT was cloned into a miR-30-based
scaffold and then into an AAV vector for delivery (AAV-miHDS1).
Successful knockdown of striatal Htt mRNA in mice by 60 % with-
out toxic effects was confirmed before the vector treatment was
applied to monkeys. AAV-miHDS1 significantly reduced rhesus