Search

Influenza A virus (IAV) causes a contagious respiratory disease which can be fatal in at risk populations. Seasonal epidemics of IAV infection are responsible for as many as 650,000 human deaths annually. Antiviral drugs currently available for the treatment of IAV infection target viral polypeptides. However, as an RNA virus, IAV is able to mutate and rapidly develop resistance to these drugs. For this reason, new approaches which focus on targeting the host are being explored. IAV interacts with many host proteins which are necessary for efficient completion of the viral lifecycle; these host factors potentially make attractive drug targets for the treatment of IAV, as development of resistance to host-targeted treatments is predicted to be slower. Host factors required for IAV replication have been identified through a number of routes including targeted research based on previous knowledge of the virus and the host, and large scale screens. The work presented in this thesis aimed to investigate three cellular proteins that had been previously proposed as pro-viral host factors for IAV. Using RNA interference (RNAi) and chemical inhibitors, I sought to dissect the role of these proteins in the viral lifecycle and to assess their suitability as drug targets in the treatment of IAV infection. The arginine demethylase JMJD6 had been suggested to have a proviral role in IAV replication. In the process of refining previously developed RNAi procedures for depleting JMJD6, I found that a published siRNA used to study JMJD6 induced an interferon (IFN) response in cells. Here, I showed that IAV was able to replicate normally in cells in which JMJD6 was depleted by alternative RNAi methods that did not induce an IFN response and in JMJD6⁻/⁻ cells, strongly suggesting that the previous finding was likely an artefact of the approach used to deplete JMJD6. Thus JMJD6 was determined not to be a host factor of IAV or a suitable target for the treatment of IAV infection. Next, a host factor that had been identified in a genome-wide CRISPR screen was investigated. The cap 2'-O-methyltransferase, CMTR1, had been proposed to be required for preventing detection of viral mRNA via the IFN signalling pathway. In my hands, knockdown of CMTR1 using RNAi did not lead to enhanced IFN response to infection, nor did it inhibit the replication of IAV in cell culture. This did not support CMTR1 alone as an effective target for IAV treatment. However, preliminary results investigating CMTR1 knockdown in combination with the viral cap-dependent endonuclease inhibitor Baloxavir, suggested that future work should examine CMTR1 as a target to enhance other antiviral treatment. Finally, the N6-methyladenosine (m⁶A) reader protein IGF2BP1 was examined. IAV RNAs, both genomic and messenger, are known to contain the m⁶A modification, which is required for optimal replication efficiency. IGF2BP1, which binds to m⁶A modified mRNA to enhance stability, has been linked to IAV replication in host interaction screens. However, using RNAi and chemical inhibition, I did not find that IGF2BP1 was required for IAV replication or spread. Furthermore, overexpression of IGF2BP1 showed no effect on IAV replication. Together these findings rule out IGF2BP1 as an IAV host factor and drug target. Overall, the work presented in this study did not support the three host proteins investigated as being suitable drug targets for the treatment of IAV, although CMTR1 could still be investigated as a target for combination therapy. The ruling out of JMJD6 and IGF2BP1 can help to focus future work investigating host targets for IAV treatment.

Post-transcriptional regulation involving RNA modifications is known as epitranscriptomic regulation. The ability of nucleotide modifications to modulate RNA-protein interaction is the basis of epitranscriptomic regulation. There are over 170 nucleotide modifications in RNA, among them m6A modification is the most abundant nucleotide modification in eukaryotic mRNA. The adenine is methylated by a set of proteins known as writers (METTL3- METTL14 complex) and demethylated by another set of proteins known as erasers (FTO and Alkbh5). Changes in m6A levels due to abnormal expression levels of writers and erasers are associated with various diseases and conditions including cancer. Oncogenicity in cancers with decreased m6A levels due to over-expression of FTO can be reduced by inhibiting FTO. In Chapter 1 of my dissertation work, the role played by m6A modification on cancer cell viability was investigated. In this study, phage display was used to discover a peptide inhibitor that binds with m6A-modified RNA leading to the inhibition of FTO binding to its target RNA. The peptide binds to methylated RNA with high affinity and inhibits FTO binding to methylated target RNA. In addition, m6A levels increased, and cell viability decreased in cancer cells upon treatment with the peptide.The m6A modification resides within a consensus sequence known as DRACH (D=A, G, U; R=A, G; H=A, C, U). The selection pressure exerted by m6A modification and its surrounding nucleotides is not well understood. In chapter 2 of my dissertation work, the evolutionary pressure exerted by m6A modification and its surrounding nucleotides was studied. Although phage display is extensively used as a method to identify novel drugs against protein or RNA targets, in this study intrinsic strength of phage display as a directed evolution method was used to understand the selection pressure exerted by m6A modification and its surrounding nucleotides on its binding proteins. Phage display experiments were conducted using RNA constructs with the most and least abundant DRACH sequences as well as with structured RNA with m6A modification in various secondary structures. These phage display experiments illustrated that m6A-modified RNA exerts a unique selection pressure compared with an unmodified RNA regardless of its DRACH sequence and that m6A modifications can exert different selection pressures depending on the region of the secondary structure of the RNA with m6A modification. There can be many m6A modifications in mRNA and they are dynamic. This dynamic nature is due to the addition or removal of methylations in mRNA by writers and erasers as a response to various intracellular or extracellular signals. This dynamicity in m6A modifications elicits the presence of other m6A readers apart from those currently found in the literature. In the third chapter of my dissertation work, the presence of novel m6A readers that are not reported in the current literature was investigated. To identify novel m6A readers, RNA pull-down assays were conducted using RNA targets that were used in phage display with and without m6A modification. Bioinformatics methods such as multiple sequence alignment and structural analysis were used to identify potential m6A binding proteins. This study identified hnRNP A1 as a potential m6A reader. Dot blot assay and calorimetric titrations illustrated that HNRNPA1 has a higher binding affinity towards the methylated RNA than the unmethylated RNA further proving the ability of hnRNP A1 to act as an m6A reader.

Glioma is the most common form of primary brain cancer and suffers from a paucity of efficacious therapies. Decades of research have successfully characterized glioma subtypes based on genetic, epigenetic, and molecular signatures that act as important prognostic indicators. Here, we present two novel approaches to designing therapeutics that convert malignant glioma and glioblastoma subtypes into more treatment-amenable subtypes. We first describe how the IDH1mut → D-2-HG ⊣ FTO axis establishes a unique epitranscriptomic profile that we term G-RAMP, which results in reduced tumor cell proliferation in both patient tumor samples and patient-derived gliomaspheres via inhibition of the anti-apoptotic regulator ATF5. G-RAMP was characterized using MeRIP-Seq unbiased screening of N6-methyladenosine (m6A) enrichment sites in IDH1mut gliomas, and small molecule inhibitors of FTO were employed to recapitulate IDH1mut growth phenotypes in more malignant IDH1wt lines. Our second approach utilized dCas9-DNMT3a epigenetic editing platforms to induce high-density methylation of the MGMT promoter and exon 1 region. MGMT methylation is a positive prognostic indicator in gliomas that predicts patient tumor responses to the standard-of-care antineoplastic agent temozolomide. We demonstrate that unmethylated MGMT gliomas can be converted to exhibit MGMT methylated profiles, which results in decreased expression of MGMT and enhanced sensitivity to temozolomide. The utilization of epitranscriptomic and epigenetic engineering approaches thus represent two novel means of effecting subtype conversions in glioma that demonstrate promising potential as neoadjuvant therapies.

The methylation of the sixth position of adenine (m6A) is an epitranscriptomic modification that has been found to play a vital role in early cell differentiation and embryonic development. The removal of the m6A modification is facilitated by the "eraser" protein FTO. The protein FTO (fat mass and obesity-associated protein) is a non-heme Fe2+ and α-ketoglutarate-dependent dioxygenase demethylase enzyme whose activity has been recently linked to spermatogenesis and cancer progression. However, little is known about the biophysical nature of its binding to its RNA targets. First, FTO was produced in BL21 (DE3) Escherichia coli transformed with the PET45-C+1 plasmid and purified using affinity chromatography. Thin Layer Chromatography with32 P-labeled, m6A-containing target RNA which was incubated with FTO resulted in loss of the m6A modification, as indicated by changes in the spotting pattern compared to a control reaction. Additionally, FRET experiments were used to determine the effect of FTO's cofactors (α-ketoglutarate and Fe2+) on its binding thermodynamics, revealing slightly increased dissociation constant (KD) values in the absence of each cofactor (α-ketoglutarate and Ni2+, an Fe2+ analog). This suggests that FTO binding to m6A RNA is slightly strengthened when both cofactors are present in solution. FRET experiments also indicated the potential of a phage-display isolated peptide to serve as an inhibitor of FTO's demethylase activity.

RNA-binding proteins (RBPs) are critical regulators of post-transcriptional gene expression and aberrant RBP-RNA interactions can cause cancer. However, RBPs are often overlooked as therapeutically relevant targets because unlike transcription factors, altered RBP activity is more frequently caused by changes in the expression levels of RBPs and their underlying targets, as opposed to somatic mutations. It has therefore been challenging to systematically evaluate the function of RBPs in disease. We addressed the lack of characterization of RBPs in cancer by developing an approach to interrogate the function of RBPs using pooled CRISPR-Cas9 screening. We identified 57 putative RBPs with distinct roles in supporting MYC-driven oncogenic pathways and we found that disrupting YTHDF2-dependent degradation of its target transcripts triggers apoptosis of MYC-dependent cancer cells and tumors. Next, we profiled YTHDF2 function using enhanced CLIP (eCLIP) and m6A-seq analysis, which revealed extensive interactions with mRNAs encoding MAPK pathway genes. We found that mRNA stabilization of upstream MAPK pathway genes drives epithelial-to-mesenchymal transition in MYC-dependent cells and is accompanied by profound oxidative cellular stress due to increased protein synthesis. We next explored the cellular stress response elicited by YTHDF2 depletion by surveying the activation of the intrinsic and extrinsic apoptotic pathways. Loss of YTHDF2 caused activation of the unfolded protein response and subsequent mitochondrial dysfunction in MYC-dependent cells, which are especially sensitive to increases in oxidative stress. Finally, we determined that PRSS23 mRNA stabilization is necessary for driving apoptosis by promoting TCF12-mediated transcription of cancer lineage-specific transcripts and translation factors. Thus, this dissertation highlights the therapeutic relevance of RBPs by uncovering the critical role of YTHDF2 in counteracting the global increase of mRNA synthesis in MYC-driven cancers.