Defining cell state regulators in cancers using single-cell analysis and CRISPR-Cas9 screening

As high-throughput transcriptional profiling becomes more sophisticated and accessible, our understanding of cancer heterogeneity and its impact on clinical outcomes are being realised. Whilst genomic variation has been extremely useful for cancer stratification and development of targeted therapies, they do not always underpin variable therapeutic responses, especially those that display higher levels of plasticity. Development of single-cell profiling has had a particular impact for elucidating the composition of cell states occurring within individual tumours, and offers high-dimensional data which can be utilised for unsupervised signature extraction. The aim of this project was to identify clinically relevant signatures of transcriptional heterogeneity in cancers by mining a pan-cancer single-cell RNA sequencing dataset spanning in 198 cell line models across 22 cancer types [1]. A dimensionality reduction method which resolves continuous expression signatures at multiple resolutions was used to resolve a range of behaviours, from consistent intra-sample cell states to cancer subtypes. In the analysis of melanoma models, three main signatures were defined that reflected distinct subtypes characterised by their differential invasive and proliferative properties. Genes highlighted as putative regulators of these sub- type signatures were screened using a single-cell RNA-seq coupled CRISPR- knock-out approach, with regulator potential uncovered for multiple targets including SOX10, MITF, EIF3G, PRPF19, RPS27A, and CDC20. Cell line annotations achieved through previous high-throughput screens were also leveraged to uncover associations between the defined heterogeneous expression signatures with features reflecting genetic variants, gene essentiality, and drug response. CRISPR perturbation screening was again used to validate the potential for putative melanoma subtype regulators to modulate the response to Rac inhibition, with results suggesting an overlap in gene regulatory networks between melanoma subtype and context specific responses to this inhibitor.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2024

Author dc:creator

• Edwards, Olivia

Advisor dc:contributor.advisor

• Adams, David

• Biology
• Cancer
• CRISPR
• Genetics
• RNA-sequencing
• Single-cell

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/1e5901a0-4e30-4eb4-a020-c7a57819ab7f/download
• https://creativecommons.org/licenses/by-nd/4.0/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.110713

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/371568