Exploration of Protein-Protein Interactions in Disease

Abstract

Towards examining the complex between TRIM28 and members of the MAGE proteins. TRIM28 has been labelled the master regulator of the human genome and functions in multiple pathways to influence transcription, stem-cell maintenance, differentiation and DNA repair. As a member of the TRIM protein family TRIM28 shares a conserved domain architecture consisting of a RING domain, 2 Bbox domains and a Coiled-coil region. The C-terminal domain of TRIM proteins are varied with TRIM28 having a PHD-BROMO domain. Members of the MAGE protein family are melanoma associated antigens that are restrictively expressed in germline cells but are often upregulated in cancer cells making them a marker of tumorigenesis. Recent reports have shown MAGE proteins to modulate TRIM28 E3-ubiquitin ligase activity resulting in the degradation of p53 and AMPK. To understand how MAGE proteins modulate TRIM28 activity we attempted to express and purify components of TRIM28 and MAGE proteins for biophysical and structural analysis. The TRIM28 Ring-Bbox-Coiled-coil (RBCC) construct was expressed and purified. Analysis of the multimeric state of TRIM28 RBCC using SEC-MALLS shows it forms higher order oligomers. This assembly can be inhibited by introduction of an assembly mutation located in the Bbox1 domain. Expression and purification of the MAGE proteins MAGE A3 and MAGE C2 was problematic as both MAGE proteins consistently formed high molecular weight soluble aggregates. Several strategies were undertaken in an attempt to purify soluble protein including protein truncation, altering the fusion tags, chaperone expression cells, changing the buffer conditions, and codon harmonization. Characterising the interaction between Fv1 and members of the mammalian ATG8 family. The co-evolution of retroviral pathogens and mammals has led to development of intracellular defence systems to combat retroviruses. Retroviral replication can be restricted at multiple stages of the retroviral life-cycle by anti-viral proteins termed restriction factors. Fv1 is the prototype restriction factor that was first discovered to determine susceptibility of mouse cells to MLV. Fv1 has phenotypic similarities to the unrelated restriction factor TRIM5α, both recognise the viral capsid lattice and block replication prior to integration of the provirus into the host genome. The similarity extends to the functional domain architecture of an N-terminal dimerization domain and a C-terminal recognition domain. TRIM5α is proposed to act as an autophagic receptor for recruitment of HIV-1 capsid to the autophagosome by interacting with the mammalian ATG8 protein family. The ATG8 family bind proteins containing a well characterised LIR binding motif. The LIR contains hydrophobic residues that bind two conserved hydrophobic pockets in ATG8 proteins. Recently, the Fv1 N-terminal domain was demonstrated to bind to the ATG8 protein LC3B. We investigated binding of Fv1 to the ATG8 family by sedimentation analytical centrifugation (SV-AUC) experiments, demonstrating Fv1 bound all 6 members. A saturation binding experiment was undertaken on 5 ATG8 proteins showing the binding affinity for Fv1 is very weak. To identify the Fv1 binding site we made a series of Fv1 (20-200) constructs containing single aromatic-to-alanine mutations and investigate binding to LC3B by SV-AUC. We found the binding activity of mutant Fv1 proteins was not altered. Our results suggest Fv1 binds the ATG8 family through an A-typical binding motif.