Structural and Functional Analysis of Human MID1 B-box1 E3 Ligase Domain: Implications to X-linked Opitz Syndrome and Overall MID1 Function

Proper function of the human MID1 protein is required for midline development during embryogenesis and is the focus of this dissertation. Mutations in MID1 are associated with X-linked Opitz Syndrome (XLOS), a congenital condition characterized by midline abnormalities that include cleft lip/palate, wide spaced eyes and LTE cleft. In cells derived from XLOS patients, it was revealed that loss of function of MID1 is associated with an increased cellular concentration of the alpha4 and PP2A proteins. The exact mechanism and details of how XLOS-observed mutations affect the function and structure of MID1 have not been well characterized. This dissertation is divided into six chapters that describes in detail the various insights obtained on the structure and function of MID1. Specifically, both structural and functional studies were performed on the B-box1 domain to gain understanding of how mutations within this domain affects the structure and overall function of MID1. Furthermore, structural and functional studies were performed with another region of MID1 called the COS domain leading to the confirmation that MID1 directly binds microtubules and the identification of a novel microtubule binding fold.;In chapter 1, a general introduction of the human MID1 protein, current understanding of its function, XLOS, the ubiquitination pathway and techniques used are presented to provide the necessary background to understand and appreciate the experiments and results described in each subsequent chapter.;In chapter 2, structural and functional data of the COS region of MID1 is presented. This region is responsible for microtubule association but the mechanism is unclear. Using three-dimensional (3D) NMR spectroscopy, the first solution structure of the MID1 COS domain was solved and was found to adopt a helix-loop-helix structure. Functional studies confirmed the COS domain in conjugation with the coiled-coil domain (CC-COS) binds to microtubules, representing the first demonstration that MID1 directly binds with microtubules rather than a microtubule-associated protein. Modeling of the CC-COS protein revealed a spectrin-like fold or three helical bundle, a novel microtubule binding fold.;In chapter 3, structural studies were performed on three XLOS-observed mutations (A130T, C142S and C145T) located within the MID1 B-box1 domain. Using two-dimensional (2D) NMR, these mutations were shown to cause the complete unfolding of the B-box1 domain. Interestingly, an unfolded B-box1 did not have any effect on the structure and stability of the neighboring domains. In parallel studies, it was shown that the B-box1 domain with the structural destabilizing mutations could no longer ubiquitinate alpha4, a key substrate of MID1.;In chapter 4, structural and functional studies were performed on another XLOS-observed mutation, P151L, within the B-box1 domain. Using 2D NMR, it was shown that the P151L mutation did not disrupt the B-box1 structure, unlike the previous three mutations described in chapter 3. Interestingly, the P151L mutant B-box1 exhibited greater E3 ligase activity compared with wild-type, by an unknown mechanism that was investigated in work described in chapter 5. The P151L B-box1 domain failed to bind and catalyze the ubiquitination of alpha4. An increase in alpha4 directly associates with an increase in PP2A.;In chapter 5, the B-box1 domain is interrogated for its E3 ligase activity given that it adopts a similar RING-like structure common to all E3 ligases. These are proteins that facilitate the ubiquitination and subsequent proteasomal-mediated degradation of proteins required for cellular homeostasis. Substrate ubiquitination is determined by the mechanism of interaction between the RING-type E3 ligases and the E2 enzymes. The B-box1 domain exhibits weak E3 ligase activity. NMR and fluorescence binding studies revealed the B-box1 domain binds on a completely different site on the E2 compared with the prototypical RING binding interface. The B-box1 domain also binds much tighter than RING domains. These observations might explain why the B-box1 domain exhibits weak E3 ligase activity compared with some RING E3 ligases.;In chapter 6, we built on observations from the P151L mutation that the E3 ligase activity of the B-box1 domain can be altered. We introduced mutations/amino acids that are conserved in RING domains within the B-box1 domain. All the mutants exhibited greater E3 ligase activity compared with the wild-type. The increased activity correlated with the increased binding affinity observed with fluorescence binding experiments. Together, these data indicate that the E3 ligase activity of the B-box1 domain can be altered while maintaining its binding affinity to the E2 enzyme. However, further studies need to be performed to understand how the mutations contribute to the enhanced E3 ligase activity of the B-box1 domain.