Jessica Cale

The type-1 fibrillinopathies are a family of connective tissue disorders of which Marfan syndrome is the most common, affecting between 2-3 in 10,000 individuals. Marfan syndrome is a multisystem disorder characterised by ocular, skeletal and cardiovascular abnormalities and can be caused by any one of over 2800 unique mutations reported across the fibrillin-1 (FBN1) gene. FBN1 encodes the large extracellular glycoprotein, fibrillin-1; the fibrillin-1 monomers aggregate to form the backbone of microfibrils. Fibrillin-1 has both structural and regulatory roles, including the regulation of transforming growth factor-beta. This regulation is critical in maintaining extracellular matrix stability and dysregulation of this function is one of the keystones of the Marfan syndrome pathogenesis. Mutations in FBN1 can result in reduced fibrillin-1 expression, loss-of-function or the production of two different fibrillin-1 proteins that are unable to interact to form functional microfibrils. The result in all three cases is a lack of functional microfibrils and destabilisation of the extracellular matrix. The current standard of care relies heavily on surgical intervention and lifelong use of medications to slow disease progression, thus the need for new therapeutic options that target the cause of disease. This thesis focused on developing a suite of short synthetic nucleic acid sequences, known as antisense oligonucleotides, to selectively manipulate FBN1 pre-mRNA splicing. We hypothesised that the removal of an amenable mutation-associated exon would result in one of the following scenarios. For missense mutations, removing the mutation-associated exon from affected and unaffected transcripts would eliminate the aberrant sequence and restore homogeneity between fibrillin-1 monomers. For splice-site and in-frame deletion mutations, excluding the mutation-associated exon from the remaining healthy transcripts would restore the domain periodicity and monomer homogeneity. Lastly, for mutations resulting in a premature termination codon, excluding the mutation-associated exon from the affected transcripts would restore the reading frame, rescuing transcript functionality. The mutation-associated exon would also need to be removed from the unaffected transcripts to maintain monomer homogeneity. For each of these scenarios, we hypothesised that the internally truncated proteins produced would be capable of forming functional microfibrils, thereby reducing the severity or slowing the progression of the Marfan syndrome phenotype. As an initial proof-of-concept for this project, antisense oligonucleotide sequences targeting FBN1 exon 52 were assessed. A promising sequence induced dose-dependent exon skipping in healthy control cells allowing us to observe the formation of healthy fibrillin-1 fibres with 0% exon skipping, loss of extruded fibrillin-1 fibres with 50% skipping; mimicking the disease-like state, and subsequent re-appearance of extracellular fibrillin-1 fibres with greater than 80% skipping indicating that the internally truncated fibrillin-1 monomers are capable of forming aggregates. Similarly, we demonstrate that FBN1 exons 47 and 59 can be efficiently excluded, and sufficient skipping can result in fibrillin-1 fibre formation. However, many of the FBN1 exons targeted were not as readily excised from the mature mRNA. Comparison of three antisense oligonucleotide chemistries revealed the promising efficacy of the newer thiophosphoramidate morpholino oligomer chemistry. Similar to the commonly used phosphorodiamidate morpholino oligomer, the thiophosphoramidate morpholino oligomer sequences resulted in efficient and consistent FBN1 exon 52 skipping. Both chemistries also had little effect on paraspeckle protein distribution, an indicator of toxicity, unlike the third, 2′OMe-PS, chemistry that caused gross paraspeckle protein disruption. Therefore, thiophosphoramidate morpholino oligomer should be included in the repertoire of chemistries routinely used in studies developing antisense therapeutics. Lastly, while we demonstrate that >80% exon skipping can lead to fibrillin-1 microfibril-like formations in vitro, we could not confirm the functionality of these fibres nor the effect of exon skipping on the Marfan syndrome phenotype. Nevertheless, this study demonstrates proof-of-concept and lays a solid foundation for further development of antisense oligonucleotides to treat the type-1 fibrillinopathies.

Background: Rare diseases are chronic and debilitating and while individually they affect less than 1 in 2,000 people, collectively they have a huge impact affect around 1.8 million Australians. Approximately 80% of these have a genetic origin, however only 30% of patients receive a formal molecular diagnosis. RASopathies are a group of rare disease that are caused by a mutations in the genes involved in the RAS-MAPK pathway, the most common of which is Noonan syndrome, which is characterised by heart defects, short stature, chest deformities, and specific craniofacial features. This study focuses on the effects of a novel c.173C>T (p.Thr58Ile) mutation, found in the NRAS gene of a patient diagnosed with Noonan-like syndrome, on the localisation and function of the NRAS protein. In doing so this study aimed to validate the role of mutations in the patient’s disease and provide information toward the creation of an experimental pipeline for the validation of other rare variants. Methods: U87-MG cells were transiently transfected with the NRAS constructs tagged with GFP2, and stained with specific antibody markers to determine localisation of the proteins using confocal microscopy. Functional studies included an Annexin V apoptosis assay, using flow cytometry to detect and quantify levels of apoptosis in transfected and untransfected populations, and the prediction of conserved domains using online bioinformatic tools. Results and Conclusions: Preliminary results from this study suggest that there is a difference between the localisation and function of the mutant and wild type proteins. The mutant protein was seen to co-localise with the Golgi apparatus as expected if the mutant protein was constitutively active. However the mutant protein was also observed to co-localise with the markers for the plasma membrane, nucleus and endoplasmic reticulum indicating that the mutation affects more than just the activation of the protein. An increase in apoptosis observed in NRASMUT transfected population, in comparison to both the wild type and untransfected populations, indicates that the mutation is engaging the pro-apoptotic function of NRAS associated with an increased activation of the RAS–RAF–MAPKK–MAPK pathway. Bioinformatic analyses identified that the mutation is location within a number of motifs involved in the binding of GTP and thus the activation and inactivation of NRAS, directing further studies toward the proliferation, activation and interaction of NRAS with effectors.