Scalable methods for the discovery of autophagy and disease genes

Macroautophagy (henceforth referred to as autophagy) is a major and conserved cellular process in which cells deliver cytoplasmic contents to lysosomes for degradation. As autophagy has been linked to various human diseases, the discovery and characterisation of autophagy genes is of interest for both basic science and medical research. This thesis develops scalable experimental and computational methods for the discovery of autophagy genes and elucidation of their relationships to human disease. To enable scalable genetic screens for autophagy, we develop two high-throughput autophagy assays: (1) an autophagic flux assay, SRAI-LC3B; and (2) an assay for levels of the ATG12-ATG5 conjugate, an important early autophagy marker. SRAI-LC3B can be used to assess autophagic flux via flow cytometry or microscopy and responds robustly to established pharmacological and genetic controls. Additionally, we exploit the optical properties of SRAI-LC3B to develop a novel high-throughput autophagy assay in a human neuronal model. By conducting a targeted screen with the ATG12-ATG5 assay, we identify two chaperone proteins as novel autophagy regulators. These chaperones act together to regulate autophagy at multiple stages, by mechanisms likely to involve stabilising the VPS34 complex and enabling DNM2-mediated autophagosome scission. Computational techniques such as network propagation and machine learning can help to condense large, complex data into testable predictions. We exploit these tools to develop a model that predicts new autophagy genes by analysing systematic datasets. Top predictions were screened using the SRAI-LC3B assay, resulting in eighteen new candidate autophagy genes. Predicted genes were enriched >5-fold in significant screen hits compared to a randomly selected control set. Subsequently, we develop a systematic procedure for process-to-phenotype prediction, which analyses interaction networks to predict diseases that may be associated with autophagy. A targeted screen of top predictions identifies twenty-eight diseases in which disruption to one or more associated genes caused significant changes in autophagy, raising the prospect that dysfunctional autophagy may play a role in some of these diseases.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2022

Author dc:creator

• Djajadikerta, Alvin

Advisor dc:contributor.advisor

• Rubinsztein, David

• autophagy

Licence

• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.91092

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/343665