Vincent Dion

The best studied Cas9 enzyme, from Streptococcus pyogenes , is thought to be too large to be efficiently packaged into a single adeno-associated virus (AAV) together with a single-guide (sg) RNA cassette, requiring a dual-vector system. Here we have circumvented this limitation and packaged the full CRISPR system into a single AAV by using a single promoter driving the expression of both the Cas9 enzyme and a sgRNA. We compared the ability of the dual and single vector systems to induce contraction of CAG/CTG repeats in the striatum of a Huntington’s disease model. We show that the all-in-one system led to larger contractions of the CAG/CTG repeats. This approach provides a more efficient and less costly approach for in vivo gene editing and improves the translational potential of AAV-mediated delivery of Cas9. One sentence summary: Delivery of CRISPR-Cas9 nickase in a single AAV improves contraction of CAG repeats in HD.

Expanded CAG/CTG repeats cause more than 15 different diseases that all remain without a disease-modifying treatment. Because repeat length accounts for most of the variation in disease severity, contracting CAG/CTG repeats presents an attractive therapeutic avenue. Here, we show that the CRISPR-Cas9 nickase targeted to CAG/CTG repeats leads to efficient repeat contractions in Huntington’s disease (HD) patient–derived neurons and astrocytes and in myotonic dystrophy type 1 patient–derived neurons. The approach was allele selective and free of detectable off-target mutations. Injection of the Cas9 nickase into the striatum of the R6/1 mouse model of HD using adeno-associated viral vectors led to repeat contractions in more than half of striatal neurons. Upon injection, we observed a reduction in the number of inclusion bodies, an improved gene expression profile, and amelioration of locomotion deficits in the open field test. The effects were greater than expected from the contractions induced, suggesting that non–cell autonomous mechanisms may be involved. Our results provide the proof of concept that correction of CAG/CTG repeats can improve HD phenotypes in a mouse disease model.

Expanded repeat disorders remain without a disease modifying treatment. Some of the most important modifiers of disease severity include inherited repeat size, their somatic mosaicism, and whether they contain interruptions. Targeted sequencing approaches are becoming the gold standard on how these parameters are measured despite repeats being notoriously challenging to sequence. Here we developed CHARLIE (Comprehensive High-throughput Analysis of Repeat Length, Interruptions and Expansions) to identify and position interruptions within expanded repeats. We used samples from Huntington's disease and myotonic dystrophy type 1 sequenced with Illumina MiSeq and PacBio's Single-Molecule Real-Time Sequencing. We validated the pipeline against previous methods for identifying germline-inherited interruptions. One challenge that can potentially mask the identification of true interruptions and sequence variation is the accuracy of sequencing platforms, which, when applied to expanded repeats, is largely unknown. Our results suggest that each sequencing platform produces a distinct error profile and we show that PCR-free library preparation for SMRT sequencing improves sequencing accuracy. CHARLIE, therefore, provides a method that can help differentiate between trivial sequencing errors and those interruptions that modify disease presentation, which can be inherited or somatic in origin.

The expansion of short tandem repeats is a feature of over 60 different human diseases. Ongoing somatic instability throughout a patient’s lifetime can influence disease progression and has emerged as a therapeutic target. Understanding its mechanism is essential for the identification of both drug targets and therapeutic interventions. A major obstacle towards this translational goal has been to measure changes in repeat size distribution given that these are complex datasets. To address this, here we provide a new analysis method, and accompanying software, that generates delta plots, extracts the instability frequency from targeted long-read sequencing data, the bias towards expansion or contraction, and the average size of the changes. It further provides statistical analysis for comparison between treatments. We show its applicability to non-dividing cells, and in vivo datasets. Moreover, we have developed a streamlined experimental design for dividing cells, Single Clone-based Instability Assay (SCIA), that saves weeks in assessing the effect of a gene knockout on repeat instability and is ideal for an initial screen. We have validated the approach using FAN1, PMS1, and MLH1 knockouts. Using SCIA, we find that although FAN1 knockout clones showed increased frequency of expansions, the size of the expansions were smaller. This highlights the wealth of information that can be extracted and the potential for novel insights into the mechanism of repeat instability.

Expansion mutations of triplet repeat sequences cause numerous inherited neurological diseases. In some diseases, affected individuals display somatic expansions in affected tissues which have been linked to accelerated disease onset and progression. There is currently considerable interest in developing therapies to slow somatic repeat expansions to delay or block disease onset. In vitro assays are particularly important to evaluate potential therapeutic interventions. Current assays typically use physical methods to monitor triplet repeat lengths within a population of cells. While useful, most of these assays are relatively slow (∼six weeks) and are somewhat limited in sensitivity to rare events. Here, a new assay, called TRX, is described to monitor CAG•CTG triplet repeat expansions more rapidly and with better sensitivity. TRX uses human tissue culture cells expressing two fluorescent proteins. Red fluorescent protein TagRFP658 is constitutively expressed and serves as an internal control. GFP is expressed in a CAG•CTG repeat length-dependent manner, with longer repeat lengths predicted to give higher green fluorescence intensity. Standard flow cytometry allows quantification of changes in fluorescent signal as a simple readout with <2% sensitivity. Two independently derived cell lines with 63 or 59 CAG repeats yielded similar rates of TRX activity. Cells with increased green fluorescence were observed within one to two weeks of culture, with longer times leading to additional signal. The appearance of green fluorescence was partly dependent on MutSβ, the DNA MSH2-MSH3 complex, based on siRNA knockdown of MSH3. However, physical analysis of the CAG•CTG repeat tracts by MiSeq deep sequencing or capillary electrophoresis showed limited changes in the length of the repeat tracts. We conclude that the TRX assay is a promising new tool for monitoring CAG•CTG repeat expansions but that further development of the assay is needed to make it fully useful.

There is currently no disease-modifying treatment for expanded CAG/CTG repeat disorders. Given that longer repeat tracts lead to an earlier age of disease onset and faster progression, contracting them is expected to improve symptoms and/or delay onset. We have previously demonstrated that the Cas9 D10A nickase can effectively contract CAG/CTG repeats when targeted to the repeat tract itself. However, the mechanism remains unclear. Here, we tested whether nickase-mediated contractions depend on transcription or on replication using human cell models. We find that transcription promotes contractions and that they occur independently of the rate of cell division. These results support the therapeutic potential of this approach in non-dividing cells.

The contraction of CAG/CTG repeats is an attractive approach to correct the mutation that causes at least 15 neuromuscular and neurodegenerative diseases, including Huntington’s disease and Myotonic Dystrophy type 1. Contractions can be achieved in vivo using the Cas9 D10A nickase from Streptococcus pyogenes (SpCas9) using a single guide RNA (sgRNA) against the repeat tract. One hurdle on the path to the clinic is that SpCas9 is too large to be packaged together with its sgRNA into a single adeno-associated virus. Here we aimed to circumvent this problem using the smaller Cas9 orthologue, SlugCas9, and the Cas9 ancestor OgeuIscB. We found them to be ineffective in inducing contractions, despite their advertised PAM sequences being compatible with CAG/CTG repeats. Thus, we further developed smaller Cas9 hybrids, made of the PAM interacting domain of S. pyogenes and the catalytic domains of the smaller Cas9 orthologues. We also designed the cognate sgRNA hybrids using molecular dynamic simulations and binding energy calculations. We found that the four Cas9/sgRNA hybrid pairs tested in human cells failed to edit their target sequences. We conclude that in silico approaches can identify functional changes caused by point mutations but are not sufficient for designing larger scale complexes of Cas9/sgRNA hybrids.