Analysis of multi-omics datasets to dissect the mechanisms of mRNA post-transcriptional regulation by RNA modification

3' untranslated regions (3' UTRs) play critical roles in controlling mRNA translation and
stability. They contain a complex assortment of sequence and secondary structural
elements that mediate interactions with RNA-binding proteins (RBPs) and miRNAs. Of all
human cell types, brain-specific mRNA isoforms have the longest 3' UTRs and therefore
their regulation is likely to be most complex, which allows them to control the precise
timing and location of protein production within neurons. Even though sequence variation
in 3' UTRs is a likely contributor to human disease, the regulatory elements in 3' UTRs are
poorly understood.
This project will focus on the regulatory elements relevant for motor neuron disease
(MND; also known as amyotrophic lateral sclerosis, ALS). MND is a common but incurable
disorder in which mRNA metabolism and protein homeostasis are disrupted. To study how
the relevant protein-RNA complexes assemble on the 3' UTRs, human induced pluripotent
cells (hiPSCs) with specific mutations will be differentiated into motor neurons. The
project will ask how the sequence and secondary structural elements in the 3' UTRs
control mRNA translation in human motor neurons, and how changes in the composition of
protein-RNA complex can disrupt this regulation in MND.
The project will involve collaborative work between the groups of Jernej Ule and Nick
Luscombe, and will thus combine experimental and computational techniques. Spinal cord
motor neurons will be generated by differentiation from hiPSCs. While the details of the
project will be decided in consultation with both supervisors, one potential project will
employ hybrid crosslinking and immuneprecipitation (hiCLIP) to study protein-RNA and
RNA-RNA interactions in the 3' UTRs in motor neurons. Genome editing will be used to
create hiPSCs with mutations in specific RBPs or in the sequence or structural elements
within 3' UTRs. This will define the disease phenotypes of 3' UTR variation, based on which
we will build computational models predicting the molecular impact of such mutations in
3' UTRs.