The HTT1a protein initiates HTT aggregation in a knock-in mouse model of Huntington’s disease

The mutation that causes Huntington's disease is a CAG repeat expansion in exon 1 of the huntingtin gene (HTT) that leads to an abnormally long polyglutamine tract in the huntingtin protein (HTT). Mutant CAG repeats are unstable and increase in size in specific neurons and brain regions with age, a phenomenon that constitutes the first step in the pathogenesis of the disease. In the presence of an expanded CAG repeat, cryptic polyA sites in intron 1 of the HTT pre-mRNA can become activated leading to the polyadenylation of a prematurely terminated transcript, HTT1a. This encodes the HTT1a protein, which is known to be very aggregation-prone and highly pathogenic. Given that the longer the CAG repeat the more HTT1a is generated, could the production of HTT1a be the mechanism through which somatic CAG repeat expansion exerts its pathogenic consequences? Resolving this issue is very important for the design of therapeutic approaches to lower huntingtin levels. We have used a CRISPR-Cas9 approach to prevent the production of HTT1a in a knock-in mouse model of Huntington's disease. All potential cryptic polyA sites were deleted from Htt intron 1 in HdhQ150 mice and colonies were established that were heterozygous for the intron 1 deletion on a mutant allele (HdhQ150ΔI) and heterozygous for the deletion on a wild-type allele (WTΔI). The CAG repeat sizes in the HdhQ150 and HdhQ150ΔI colonies were well-matched at approximately 195 CAGs. As predicted, the deletion of the cryptic polyA sites from Htt intron 1 prevented the generation of the Htt1a transcript in the HdhQ150ΔI mice. However, very low levels of the HTT1a protein were detected, which resulted from a Htt readthrough product of exon 1 and exon 2, that had retained the deleted intron and terminated at a cryptic polyA site in intron 2. HdhQ150, HdhQ150ΔI, wild-type and WTΔI mice were studied until 17 months of age. Immunohistochemical and homogeneous time resolved fluorescence analysis showed that HTT aggregation in both HdhQ150 and HdhQ150ΔI brains contained HTT1a, but the dramatic decrease in soluble HTT1a levels in HdhQ150ΔI brains delayed the appearance of aggregated HTT1a by several months. Although this delay in aggregate pathology only partially reversed transcriptional dysregulation, the biomarkers NEFL and BRP39 (YKL40) remained at wild-type levels in HdhQ150ΔI mice at 17 months of age. These data demonstrate that the production of HTT1a initiates HTT aggregation and that it is important to target HTT1a in huntingtin-lowering therapeutic strategies.