Spatial Defenition Of The HELA Transcriptome : Optimization and improvement of Halo-sequencing

Abstract

Background and aims: After their transcription many RNAs are specifically localized to distinct areas of the cell. Localization can alter RNA’s fate by controlling its availability for translation, degradation, and interactions with proteins. It can also create local translatomes, where the translated proteins are more abundant compared to the rest of the cell. Even though RNA localization is linked to several important cellular events and even diseases, studying RNA localization remains difficult. Traditional methods are often limited in either spatial resolution or throughput. RNA localization has been studied using imaging-based methods which provide high resolution but low throughput, and biochemical or mechanical fractionation can provide high throughput but is limited in spatial accuracy. In recent years, RNA proximity labelling techniques have been developed to combat the limitations of existing methods. The methods use specifically localised guide proteins that can biochemically tag RNAs in their immediate vicinity. Halo-seq, originally developed in 2020, is a proximity labelling method that uses HaloTagged guide proteins and a dibromofluorescein (DBF) Halo-ligand to alkynylate RNAs near the guide protein. The method has not seen widespread adaptation due to the difficulty of manufacturing the DBF Halo-ligand. In collaboration, we synthetized the DBF Halo-ligand using an improved synthesis approach and the goal of this thesis was to establish and optimize a Halo-seq system including comparison with a traditional fractionation technique.

Methods: Clonal cell lines expressing Halo-fusion proteins destined to nucleus, nucleolus or whole cell were established. HaloTag-proximal RNA was biotinylated by adapting and optimizing the Halo-seq protocol originally published by Engels, et al. (2022). Biotinylated RNA was enriched using a streptavidin pulldown followed by RNA sequencing. RNAs enriched or depleted in pulldown samples compared to their inputs were determined using command line and R-based bioinformatics tools. In parallel, nuclear-cytosolic fractionation was performed and RNA isolated from the subcellular he fractions were sequenced.RNA abundance in the fractions were determined and compared with the Halo-seq enriched RNAs.

Results: Analysis of differential RNA abundances between Halo-seq pulldown and input samples showed that Halo-seq can specifically identify localized RNAs both in membrane bound or membraneless compartments. Compared to nuclear-cytosolic fractionation, Halo-seq showed superior spatial resolution but was limited in sensitivity.In conclusion, the established and optimized Halo-seq enables the detection of Halo-proximal RNAs in different subcellular compartments.