CRISPR Screening for Neurological Diseases

Neurological diseases, such as neurodegenerative, neurodevelopmental, and neuropsychiatric disorders, remain a significant health concern as many lack effective treatment options. While there have been significant advances in the identification of genetic variants contributing to neurological disease using next-generation sequencing (NGS), the understanding of the cellular and molecular basis of how these variants contribute to disease is still lacking.

CRISPR-based screens are now increasingly used to perform large-scale, genome-wide functional studies to elucidate disease mechanisms and identify therapeutic targets—but the inherent limitations of using patient tissue samples and animal models means that using them for the study of neurological disease remains difficult. This article discusses the challenges of performing CRISPR screens in neurological disease models and how the development of human induced pluripotent stem cell (iPSC)-derived disease models could potentially provide a new approach to performing functional studies for neurological disorders.

CRISPR-based screening for functional studies

The CRISPR-Cas9 system was originally identified as a bacterial adaptive immune system, protecting prokaryotes against viral infection. It was then repurposed into a powerful gene-editing tool, enabling researchers to make targeted and specific changes to gene sequence—the Type II CRISPR system from S .pyogenes is the most widely used for gene editing, requiring only a short guide RNA (sgRNA) to target the Cas9 endonuclease to the 3–NGG-5’ protospacer adjacent motif (PAM) site. If the sgRNA contains significant homology with the sequence immediately adjacent to the PAM site, then a double-strand break (DSB) is introduced at the target loci, triggering the activation of endogenous DNA repair pathways.

Gene knockout relies on the activation of the Non-Homologous End Joining (NHEJ) pathway—as this is error prone, the resulting repair will often cause an indel at the cut site and subsequent frameshift mutation. In the presence of an exogenous DNA template, the DSB will initiate Homologous Repair (HR) facilitating alterations to gene expression of gene knock in. Gene expression can also be modulated with CRISPRi and CRISPRa, which utilize a catalytically inactive version of Cas9 (dCas9) tethered to regulatory regions.

The programmable nature of CRISPR-Cas9 and simplicity of targeting have allowed scientists to design guide RNAs against every gene in the genome. The ability to perform gene editing in multiplex also means that scientists can use CRISPR for genome-wide functional genomic screening. Pooled CRISPR screens are now increasingly popular, where lentiviral delivery of the sgRNA results in the integration into the genome, providing a barcode that can be identified by NGS. However, in some cases, the use of an arrayed CRISPR screen may allow for other phenotypes to be screened, such as cell morphology.

The challenge of studying neurological disease

An important consideration when embarking on a CRISPR screen is selecting an appropriate model system; it needs to be both biologically relevant and technically feasible for use in a genome-wide screen. When performing a CRISPR screen, it can be a good idea to use a scalable model for the initial screen, and then follow up any hits in a more relevant surrogate system. However, when studying neurological disorders, the selection of such models comes with significant challenges.

The most biologically relevant model to use in CRISPR screening is primary cells—but primary cells of the human nervous system are difficult to obtain and cannot be expanded, and so are therefore not technically feasible. However, animal models such as the mouse are not biologically relevant, as disease-relevant cell types, such as glia and neurons, differ between species and there are significant genetic differences that make modeling disease-causing variants difficult. Using non-human primate models does circumvent these limitations, but ethical and cost considerations mean that they are also not a feasible alternative for use in a high-throughput screen.

The majority of CRISPR screens published on neurological disorders so far have used cancer cell lines, but these too have limitations. Many neurological diseases are caused by changes in the function of neurological cells—which can therefore only be successfully characterized in the relevant cell as there is a danger that disease-relevant genes may not be expressed in a model cell line and so would not be detected. There is also a crucial limitation of using cancer cell lines for the study of neurodegeneration in which cell survival is a key phenotype—essential genes in cancer cell lines differ from those in stem cells and neurons, and so the control of cancer cell survival is very different to those neurologically relevant cell types.

iPSC technology: a new neurological model?

An exciting development that could provide a potential workaround for these limitations is the use of induced pluripotent stem cells (iPSCs) as an alternative surrogate model. iPSCs are artificial stem cells created by reprogramming somatic cells, such as skin fibroblasts and peripheral blood cells, by over-expressing so-called Yamanaka factors, which are four specific transcription factors that induce pluripotency.

Once created, iPSCs can then be further differentiated into neuronal subtypes, glia, and neurons and so could be a good alternative for neurological disease modeling—iPSCs can be derived from healthy donors, as well as from patients with neurological disease. Those iPSCs created from donors with familial or sporadic disease have been shown to exhibit a disease phenotype,¹ for example, iPSC-neurons derived from patients with Alzheimer’s disease (AD) show several of the key phenotypes associated with the disease, such as production of AB peptides.² There is inherent variability between donors, but because iPSCs are less prone to genetic drift during extended cell culture, they can more easily be used to engineer an isogenic cell-line pair. Researchers can use CRISPR gene editing to either correct disease-associated mutations or alternatively introduce a disease-associated variant in a healthy donor.

CRISPR screening in iPSC-derived cells

Most importantly for CRISPR screening, iPSCs can be expanded in culture prior to differentiation, thereby achieving the required cell numbers for performing a genome-wide, high-throughput screen. The first such CRISPR screen was performed in neurons derived from human iPSCs, where a sgRNA library was introduced via lentiviral vector into iPSCs which were then differentiated into neurons.³ This CRISPR screen really demonstrated the importance of using a biologically relevant model, as resulting hits associated with both the survival of neurons and essential housekeeping genes was significantly different to those generated in screens using cancer cell lines, or the undifferentiated iPSCs themselves. In addition, by utilizing cell imaging, the researchers were able to look at how neuron-specific morphological features, such as neurite length and branching patterns, are altered with specific gene knockdown that would not have been possible if a different, irrelevant cell type was used.

The CRISPR revolution

CRISPR-based screening has revolutionized the study of many diseases, allowing the elucidation of cellular and molecular mechanisms that drive the development of many diseases. The integration of two technologies, CRISPR-based screening and iPSC-disease modeling is perhaps the key to unpicking the mechanisms and identifying potential therapeutic targets for neurological disorders.⁴

There are still some challenges to overcome, such as the current limit of disease-relevant cell types possible from iPSCs, and the difficulty in recapitulating older neurons, as neurodegeneration is associated with many neurological disorders. Many neurological disorder phenotypes also rely on interactions between different cell types—however, iPSC-derived brain organoids and assembloids also offer an exciting possibility for future study using CRISPR screening methodologies. New CRISPR innovations, such as prime editing, could also help with the creation of disease-associated variants in iPSCs, providing the means to study and evaluate the phenotypic consequences of gene expression, thereby increasing our understanding of neurological disorders.

Key Takeways

• Many neurological disorders do not have effective therapeutic options due to a lack of understanding as to the cellular and molecular pathways that drive disease.
• CRISPR screening is now increasingly used to study the phenotypic consequences of gene expression on a genome-wide scale to elucidate biological pathways
• Due to the inherent limitations in using primary cells, such as neurons and glia, and significant genetic differences in animal models and immortalized cell lines, the study of neurological disorders is hampered by a lack of appropriate biological models
• Induced pluripotent stem cells (iPSCs) are artificial stem cells derived from human somatic cells that can then be further differentiated into biologically relevant models for use in high-throughput CRISPR screens for the study of neurological disease

References

1. Han, Steve S W et al. “Constructing and deconstructing stem cell models of neurological disease.” Neuron vol. 70,4 (2011): 626-44.

2. Israel, Mason A et al. “Probing sporadic and familial Alzheimer's disease using induced pluripotent stem cells.” Nature vol. 482,7384 216-20. 25 Jan. 2012,

3. Tian, Ruilin et al. “CRISPR Interference-Based Platform for Multimodal Genetic Screens in Human iPSC-Derived Neurons.” Neuron vol. 104,2 (2019): 239-255.e12.

4. Kampmann, Martin. “CRISPR-based functional genomics for neurological disease.” Nature reviews. Neurology vol. 16,9 (2020): 465-480.