Abstract
Post-translational modifications serve to regulate the function, activity, abundance or cellular localisation of proteins in a dynamic and timely manner. One form of such modifications involves the covalent attachment of a conserved 76-residue protein called ubiquitin (Ub) to the ε-amino group of Lys side chains. In many cases, Ub attachment (or ‘ubiquitylation’) tags the modified proteins for degradation by the 26S proteasome. Ubiquitylation plays a key role in cell signalling because, by controlling the abundance of specific proteins, it can influence the outcome of certain pathways.
Ub attachment requires the activity of three groups of enzymes: E1 Ub-activating enzymes, E2 Ub-conjugating enzymes and E3 Ub-protein ligases. They are often simply referred to as E1 enzymes, E2 enzymes and E3 ligases, respectively. The human genome possesses two genes encoding for E1 enzymes, around 40 genes encoding for E2 enzymes, and more than 600 genes encoding for E3 ligases. The great majority of these E3 ligases possess Really Interesting New Gene (RING) domains, which directly recruit a Ub-loaded E2 enzyme (E2~Ub conjugate) to promote Ub transfer.
Several members of the Inhibitor of Apoptosis (IAP) protein family, such as the human cellular IAP 1 and 2 (cIAP1 and cIAP2, respectively), are RING-type E3 ligases. Their E3 ligase activity is essential for regulating pathways that mediate apoptosis, cell division and the immune response. cIAP1 and cIAP2 mainly ubiquitylates components of the Tumour Necrosis Factor α (TNFα) Receptor 1 (TNF-R1)-associated protein complex. This process leads to stabilisation of the complex and promotes the expression of anti-apoptotic genes. The realisation that many cancer cell lines over-express cIAP1 and cIAP2 to promote their survival has sparked the development of a new class of drugs named SMAC-mimetic (SM) compounds. Binding of SM compounds to the Baculoviral IAP Repeat 3 (BIR3) domain of cIAP1 and cIAP2 has been shown to promote their self-ubiquitylation (or ‘auto-ubiquitylation’) and proteasomal degradation, as well as sensitise the cells to other killing agents.
Studies by previous members of our group have shown that dimerisation of the C-terminal RING domain of cIAP2 is essential for ubiquitylation. Using purified recombinant cIAP1 and cIAP2 proteins, this study revealed that cIAP1 adopted a predominantly monomeric conformation in solution. This was in contrast to the more dimeric cIAP2. Measurements using a number of biophysical methods also showed that the addition of SM compounds promoted RING domain-dependent dimerisation of cIAP1. Dimerisation was not accompanied by changes in the secondary structural contents, suggesting that SM compounds only promoted domain re-arrangements. Furthermore, subsequent activity assays pointed to a role of the BIR3 domain in inhibiting RING domain dimerisation. This first part of the project highlights the importance of a non-catalytic (BIR3) domain in regulating the activity and oligomeric state of the catalytic RING domain. In addition, the results provide insights into the mechanism of SM compound-induced activation of cIAP1.
In addition to the BIR3 and RING domains, cIAP1 and cIAP2 also possesses a Ub-binding domain called the Ubiquitin-Associated (UBA) domain. Others have suggested that the UBA domain had important roles in the induction of anti-apoptotic genes, and/or the proteasomal degradation of cIAP1 and cIAP2 following auto-ubiquitylation. However, these studies did not find any evidence for the role of the UBA domain in influencing E3 ligase activity. It should be noted that, in these studies, Ala substitution was introduced to the conserved motif of the UBA domain. Herein, it is demonstrated that this commonly-introduced mutation destabilises the UBA domain fold and causes artificial alterations in the E3 ligase activity pattern of the resulting mutant cIAP1 protein. With this in mind, a new fully-folded UBA domain mutant cIAP1 protein displaying impaired Ub binding was generated. E3 ligase activity comparison between wild-type cIAP1 and the new mutant protein revealed that the UBA domain functioned to accelerate the initial rate of auto- and substrate ubiquitylation. This rate enhancement was likely caused by direct E2~Ub conjugate recruitment by the UBA domain. Interestingly, similar observations were also made using cIAP2, suggesting that the mechanism is conserved. Overall, this part of the project uncovered an additional layer of E3 ligase activity regulation in cIAP1 and cIAP2, which has been previously overlooked.