Abstract:
Post-transcriptional regulation of ribonucleic acid (RNA) is mainly orchestrated by the dynamic
interplay between the RNA molecule itself and the interacting proteins. Through these interactions, RNAs can be transported to distinct cellular places, sequestered away into granules or,
among other things, actively translated into proteins. Together with the RNA they bind to,
RNA interacting proteins form ribonucleoprotein complexes (RNPs), which are reasoned to be
the true regulatory units for post-transcriptional regulation. Identifying the binding partners
of RNA thus provides valuable details to better understand their function.
Multiple techniques have been developed to characterize the interactions between RNAs and proteins. A more recent approach adopts a method initially developed to study proteinprotein interaction: proximity labeling. The elegant RNA-BioID assay enables the detection of
RNA–protein interactions and was presented by Mukherjee et al. in 2019 [1]. Starting from this publication, this thesis aims to advance, apply and follow-up on the RNA-BioID method.
The main aim of this study was to take the RNA-BioID one step further by applying it to a
lesser studied group of RNAs, the so-called long non-coding ribonucleic acid (lncRNA). By setting up a similar assay as Mukherjee and colleagues, the hope was to gain insight into the
protein interactome of two lncRNA candidates (NORAD and OIP5-AS1). As a first step, the
fusion construct, including the labeling enzyme necessary for proximity labeling, was stably
introduced into HeLa cells. This was achieved for two different labeling enzymes (microID and
TurboID) and their performance was compared. In the second step, the endogenous genes of the
two lncRNA candidates was tried to manipulate with several different CRISPR/Cas9 strategies.
As this failed repeatedly, the approach was changed to an in vitro proximity labeling method
to avoid the necessity of genetic manipulation of the cells. After expression and purification of
the proximity labeling enzyme, together with DNA probes that should direct it to the correct RNA target, it was applied to fixed and permeabilized HeLa cells. Prior, it was shown that the
probes which finally tether the enzyme to the RNA are correctly localized. The enzyme did so too, given the right probe concentration. However, having tested several different conditions,
the labeling still did not appear at the expected position. Therefore, the approach was not
further continued.
The last part of this thesis aimed to follow-up on the initial RNA-BioID assay. The β-actin
mRNA interactome has been detected in an unbiased screen in mouse embryonic fibroblast
cells. Several protein candidates which were found as potential interactors have been selected
for verification (hnRNP K, TDP-43, PTBP1, and hnRNP R) to provide insights into their
biological relevance. The interaction was analyzed by pull-down of the protein and RT-qPCR
of the co-purified RNAs. If β-actin mRNA is bound by the RNA-binding protein, it is pulled
down along with the protein. A successful enrichment of β-actin mRNA could be shown for PTBP1 and hnRNP R.
lncRNA