The role of MUL1 in the regulation of rig-i-like receptor (RLR) mediated inflammatory and antiviral response

Mitochondria ubiquitin ligase 1 (Mul1) is a mitochondrial protein located on the outer mitochondrial membrane of the organelle. Mul1 consists of 352 amino acids, has two putative transmembrane domains (at the N- (called here TM1) and C-terminus (TM2) respectively), as well as a C-terminal RING domain. Mul1 has been previously reported to have E3 ligase activity, while a role has been suggested in modifying mitochondrial dynamics. Furthermore, Mul1 was originally identified from a genomic screen as a putative activator of the prototypic inflammatory transcription factor, NFκB, which indicates a possible role in the innate immunity. Attention in this project is focused on the role of Mul1 in regulating RIG-I-like Receptors (RLRs) mediated antiviral responses. RLRs consist of RIG-I and MDA5, which are cytoplasmic pathogen recognition receptors that recognise viral RNA species during viral infection and transmit downstream signalling via a mitochondrial adaptor protein known as MAVS, resulting in the expression of proinflammatory and antiviral genes such as NFκB and Type I interferon. Excessive inflammation and activation of antiviral response can be detrimental to the host and clinically manifests as chronic inflammation or death. Therefore, the activation of RLRs and its downstream signalling require tight regulations to avoid excessive inflammation. In my research, I have demonstrated that Mul1 colocalises with MAVS on the mitochondria by immunocytochemical staining and confocal imaging. Furthermore, Mul1 was found to interact with the constitutively active forms of both RIG-I and MDA5 in coimmunoprecipitation studies. Previous reports have shown that RIG-I and MDA5 require polyubiquitination for activation and signal transduction. In the presence of Mul1, I found that there was enhanced SUMOylation of the RLRs, and reduced polyubiquitination of these receptors. These findings suggest that Mul1 may antagonise the polyubiquitination, and thus the activation status, of RLRs by regulating the post-translational modification of these receptors. To understand the role of Mul1 in RIG-I and MDA5 mediated antiviral responses, Mul1 siRNA depletion was performed in primary human lung fibroblasts, resulting in potentiated antiviral response following stimulation with Sendai virus, influenza A virus and transfected poly(I:C), which activate the RLRs. Together, these studies suggest a role for Mul1 as a negative regulator of RLR mediated antiviral signalling. This thesis also presents the first ever characterisation of a Mul1 gene-deficient mouse model. The Mul1 expressing gene in this mouse model is replaced with the gene expressing a LacZ reporter under the control of the natural Mul1 promoter, allowing for the assessment of endogenous Mul1 expression levels in a range of tissues using LacZ reporter assays. Mul1-/- mice exhibit significantly smaller body weight (22%) compared to wild type mice. Quantitative RT-PCR and LacZ reporter assay revealed that Mul1 is highly expressed in the brain, heart and testes. Intriguingly, preliminary studies using transmission electron micrographs of liver tissues from a Mul1-/- mouse demonstrated that although the mitochondrial morphology is normal, the endoplasmic reticulum (ER) displayed abnormal morphology. This finding, although inconclusive at present, provides clues suggesting that although Mul1 is a mitochondrial protein, it may have a more crucial function in maintaining ER morphology and the interaction between ER and mitochondria. Critically, in vivo influenza A infection of Mul1-/- mice further corroborated the results from the in vitro infection experiments, in which Mul1-/- mice demonstrated potentiated antiviral response after infection. Since Mul1 is a mitochondrial protein, I have also investigated the structural determinants required for the targeting of Mul1 to the mitochondria. Using immunocytochemical staining and confocal imaging, with the ectopic expression of Mul1 (and mutants with deletions with respect to TM1 and TM2), I have shown that TM1 appears to be more important for Mul1 targeting to the mitochondria compared to TM2. Critically, Mul1 expression at high levels was associated with fragmented mitochondrial morphology, suggesting that Mul1 may alter mitochondrial morphology only when present at high levels. Significantly, mitochondrial morphology did not differ between wild type and Mul1-/- MEFs. Hence, Mul1 may not be essential in maintaining the processes governing mitochondrial morphology under normal circumstances, but possibly under conditions of increased expression in yet to be identified conditions. Overall, the findings from my research support the hypothesis that Mul1 is a negative regulator of RLR antiviral signalling. Mul1 appears to alter RLR activation by means of regulating the post-translational modification of these receptors. In vivo experiments provided further clues on the biological role of Mul1 beyond the regulation of RLR signalling. Furthermore, the possible role of Mul1 in maintaining ER morphology and ER-mitochondrial interaction may also contribute to the regulation of essential metabolic and physiological functions where Mul1 is highly expressed, such as the brain, heart and testes. In conclusion, this thesis has provided the first crucial insight of the biological role of Mul1, a novel mitochondrial protein, and provides the basis for further studies in elucidating the role of Mul1 in cellular function. Ultimately, understanding the regulation of inflammation pathways will provide greater insights into disease progression and development, leading to opportunities of therapeutic intervention.