Many neurological diseases are in need of effective treatments. CRISPR/Cas9-mediated genome editing holds enormous potential as a therapeutic strategy for the inherited forms of these neurological diseases. It is considered “once-for-all,” as its effects can last a lifetime. One bottleneck in the translational application of genome editing for treating brain disorders is the absence of brain-wide delivery approaches to transfer genome-editing components into the whole brain. Current delivery approaches for brain genome editing utilize intraparenchymal injection [1], which only enables a limited brain region to receive genome-editing components. As many neurological disorders affect multiple brain regions (e.g., Alzheimer’s disease [AD] affects the cortex and hippocampus), the localized delivery of such genome-editing components may restrict their beneficial effects. Therefore, an efficient brain-wide delivery of genome-editing components is urgently needed for the development of therapies for inherited neurological diseases.

Modified adeno-associated virus (AAV) variants that can cross the blood–brain barrier (BBB) have recently been developed, enabling widespread transduction in the brain after intravenous delivery [2] [3]. Therefore, there is a great potential to develop a therapeutic strategy for neurological disorders that enables brain-wide delivery of genome-editing components by administration of a BBB-crossing virus. However, the ability of such a virus to deliver genome-editing components to ameliorate the disease phenotypes has not yet been demonstrated successfully in vivo.  

Therefore, we aimed to develop a brain-wide genome-editing strategy by combining CRISPR/Cas9-mediated genome editing and the BBB-crossing virus delivery method. As a proof-of-concept study, we investigated whether this system could efficiently edit a familial Alzheimer’s disease (FAD) mutation (APP Swedish mutation) and alleviate AD-associated pathologies.

Firstly, we adopted the PX601-based AAV-Cas9 construct [4] with modification of the promoter by replacing the cytomegalovirus (CMV) promoter with a smaller elongation factor 1-alpha short (EFS) promoter. We packaged this construct into AAV9, which can infect the brain cells with high efficiency via intrahippocampal injection. We then found that intrahippocampal delivery of this AAV9-Cas9 system into the brains efficiently edited the APP Swedish mutation in the virus-infected brain regions (~30% of genome-editing efficiency), suggesting this Cas9 system works well in vivo.

Next, we packaged the Cas9 construct into a BBB-crossing AAV variant, AAV-PHP.eB. We could not detect any editing events after systemic delivery of this AAV-PHP.eB-Cas9 system, and only detected the Cas9 protein in a few brain cells. However, when we co-injected this AAV-PHP.eB packaged Cas9 virus and a GFP virus, we observed strong GFP signals. These findings suggest that the viral construct of Cas9 could not express well at low viral copy numbers, resulting in a barely detectable expression of Cas9.

Given that Cas9 and GFP driven by 2 viral constructs showed a drastic difference in their expression signals, we compared their viral expression constructs and examined whether any critical element was missed in the viral construct that expresses Cas9. We found that the Cas9 construct did not contain a sequence of Woodchuck Hepatitis Virus (WHP) Posttranscriptional Regulatory Element (WPRE), which is for the enhancement of transgene expression in viral vectors. However, it is technically not feasible for us to insert the full-length form of WPRE into the Cas9 construct as the size of the Cas9-WPRE construct exceeds the package limit of the AAV virus (around 5 kb). We then came across a paper on the different truncated forms of WPRE [5]. After integrating a truncated form of WPRE into the Cas9 construct without exceeding the package limit of the AAV virus, this system significantly enhanced the expression of the Cas9 protein. Moreover, compared to the editing efficiency of local intrahippocampal injection, this modified system achieved a similar editing rate in multiple brain regions. We further modified this construct to specifically target neurons by using a neuron-specific human synapsin 1 (Syn) promoter. With these optimizations, this new BBB-crossing, neuron-specific, genome-editing system can achieve effective, efficient brain-wide genome editing. In our proof-of-concept study, we demonstrated that this brain-wide genome-editing strategy can successfully disrupt the APP Swedish mutation and lead to beneficial outcomes including the alleviation of a brain-wide amyloid pathology and improved cognitive functions.

To our knowledge, this is the first demonstration of brain-wide genome editing at the adult stage via modified AAV variants. This study demonstrates the feasibility of using this genome-editing approach for efficient brain-wide gene editing and provides experimental evidence that this approach reduces disease pathology and improves cognitive functions. Our brain-wide genome-editing strategy can be applied to other monogenetic brain disorders in which multiple regions are affected, such as amyotrophic lateral sclerosis (which is caused by mutations in the C9orf72 gene and affects motor neurons in the primary motor cortex, brainstem, and spinal cord) and Parkinson’s disease (which is caused by mutations in the LRRK2, PARK7, PINK1, PRKNs, or SNCA genes and affects hippocampus, thalamus, and anterior cingulate functions). We believe this study serves as a crucial step in promoting the use of genome editing in brain disease treatment, specifically in developing precision medicine for inherited forms of neurodegenerative disorders.

Our paper: Duan Y., Ye T., Qu Z., Chen Y., Miranda A., Zhou X., Lok K.C., Chen Y., Fu A.K.Y., Gradinaru V. and Ip N.Y. Brain-wide Cas9-mediated cleavage of a gene causing familial Alzheimer’s disease alleviates amyloid-related pathologies in mice. Nat Biomed Eng (2021). doi: 10.1038/s41551-021-00759-0.PMID: 34312508  

https://doi.org/10.1038/s41551-021-00759-0

References:

[1] Kimura, S. and Harashima, H. Current Status and Challenges Associated with CNS-Targeted Gene Delivery across the BBB. Pharmaceutics 12,12 (2020).

[2] Chan, K.Y., et al. Engineered AAVs for efficient noninvasive gene delivery to the central and peripheral nervous systems. Nature neuroscience 20, 1172–1179 (2017).

[3] Deverman, B.E., et al. Cre-dependent selection yields AAV variants for widespread gene transfer to the adult brain. Nature biotechnology 34, 204–209 (2016).

[4] Ran, F.A., et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature 520, 186–191 (2015).

[5] Choi, J.H., et al. Optimization of AAV expression cassettes to improve packaging capacity and transgene expression in neurons. Mol Brain 7, 17 (2014).