Periodic Reporting for period 1 - TRIM MAGE complexes (Deciphering the mechanism of the dual functional E3 ligase TRIM28 in cancer cells)
Reporting period: 2021-04-01 to 2023-03-31
Ubiquitination is the post-translational modification of substrate proteins and involved in the regulation of almost all cellular signalling pathways. It is catalysed by a three protein enzymatic cascade, comprised of E1 activating, E2 conjugating and E3 ligating enzymes. TRIM family proteins are RING-type E3 ubiquitin ligases that are characterised by the presence of an N-terminal tripartite motif that harbours the catalytic RING domain and a C-terminal variable substrate-binding domain. TRIM28 is a part of class VI TRIM proteins that has unique properties: Instead of substrate binding domains, it harbours C-terminal PHD-Bromo domains which mediate its role in transcriptional repression. Additionally, while most TRIM proteins characterised so far require RING dimerisation for ubiquitination activity, all three members of class VI of TRIM ligases, including TRIM28, have recently been shown to contain monomeric RINGs and indeed show no apparent E3 ligase activity. Nevertheless, ubiquitination activity has been described for TRIM28 in a cellular context, especially in cancer cells, in conjunction with MAGE proteins. MAGE proteins are aberrantly expressed in many cancer types and have been proposed to regulate and activate TRIM28 E3 ligase activity to mediate proteasomal degradation of substrates, such as p53 and AMPK. However, the regulation and molecular details of MAGE dependent activation of TRIM28 E3 ligase activity remain unknown. Furthermore, it is unclear, how TRIM28 suppresses E3 ligases activity while being part of transcriptional complexes, as there is currently no evidence in the literature that suggests that both functions of TRIM28 are functionally connected.
The goal of the action was to use a multi-disciplinary approach to study MAGE-dependent regulation of TRIM28 ligase activity. Unbiased proteomic studies together with in vitro biochemical, biophysical, and structural experiments tried to identify cellular signals such as posttranslational modifications and interaction partners that can modulate MAGE-TRIM28 complexes and regulate the interplay between its transcriptional activity and E3 ligase activity. To uncover how MAGE proteins modulate substrate recognition, E2 recruitment and ubiquitin ligase activity of TRIM28, and to establish a molecular model of the MAGE-TRIM28 complex, structural and mechanistic studies were performed.
The goal of the action was to use a multi-disciplinary approach to study MAGE-dependent regulation of TRIM28 ligase activity. Unbiased proteomic studies together with in vitro biochemical, biophysical, and structural experiments tried to identify cellular signals such as posttranslational modifications and interaction partners that can modulate MAGE-TRIM28 complexes and regulate the interplay between its transcriptional activity and E3 ligase activity. To uncover how MAGE proteins modulate substrate recognition, E2 recruitment and ubiquitin ligase activity of TRIM28, and to establish a molecular model of the MAGE-TRIM28 complex, structural and mechanistic studies were performed.
To understand the molecular mechanism of the regulation of TRIM28 activity by MAGE protein, the work performed during the action was divided into three work packages (WPs):
WP1: Investigate if additional interaction partners and PTMs contribute to the regulation of MAGE-TRIM28 activity
A significant part of this WP was to generate a MAGE-dependent interactome map of TRIM28 to identify novel interactors and potential regulators. After establishing the cellular tools required for this WP, protocols were generated that enabled proximity-based labelling of TRIM28 interactors and their identification by proteomics. Although the workflow established generated reproducible data, TRIM28 was found to have a very high number of interactors, potentially due to unspecific binding or DNA mediated interactions, which made it difficult to rationally discriminate between real binders and unspecific background. However, results from these experiments opened a new experimental strategy to study a different and unusual function of TRIM28: its activity of a SUMO E3 ligase.
WP2: Characterise the effect of novel binders and PTMs of TRIM28 on its function
This WP aimed to validate novel binders of TRIM28 in the context of MAGE proteins. As this WP depended on results from WP1, their validation and effect on TRIM28 could not be studied in further detail for reason explained in the previous paragraph. However, several cellular and in vitro assays provided further evidence that TRIM28 is a TRIM family member that is devoid of ubiquitin E3 ligase activity and may be regarded as a “pseudo” E3 ligase.
WP3: Elucidate the molecular mechanism of MAGE-dependent modulation of TRIM28 E3 ligase activity
The goal of WP3 was to generate a molecular model of MAGE dependent regulation of TRIM28 activity using biochemical, biophysical, and structural approaches. However, despite extensive work aiming to reproduce results from the literature, no MAGE-dependent regulation TRIM28 activity could be shown. Furthermore, several biophysical techniques could not show reliable binding of recombinant MAGE proteins to TRIM28, and structural approaches did not yield a structural description of the complex.
Several alternative approaches to study the activity of TRIM28 alone were undertaken, but no activity of TRIM28 could be shown in diverse assays and structural studies on either isolated domains or the full-length protein did not yield structural insights due to the highly flexible nature of TRIM28.
As several of the experiments performed in WP1-3 did not yield the expected results, an alternative function of TRIM28 mediate by its unique C-terminal domains was analysed. Using a broad range of biochemical and structural experiments allowed me to show that TRIM28 does not work as a novel bona fide SUMO E3 ligase despite its apparent SUMOylation activity, but instead might serve as a multi-protein interaction platform to mediate substrate sumoylation at different genomic locations.
WP1: Investigate if additional interaction partners and PTMs contribute to the regulation of MAGE-TRIM28 activity
A significant part of this WP was to generate a MAGE-dependent interactome map of TRIM28 to identify novel interactors and potential regulators. After establishing the cellular tools required for this WP, protocols were generated that enabled proximity-based labelling of TRIM28 interactors and their identification by proteomics. Although the workflow established generated reproducible data, TRIM28 was found to have a very high number of interactors, potentially due to unspecific binding or DNA mediated interactions, which made it difficult to rationally discriminate between real binders and unspecific background. However, results from these experiments opened a new experimental strategy to study a different and unusual function of TRIM28: its activity of a SUMO E3 ligase.
WP2: Characterise the effect of novel binders and PTMs of TRIM28 on its function
This WP aimed to validate novel binders of TRIM28 in the context of MAGE proteins. As this WP depended on results from WP1, their validation and effect on TRIM28 could not be studied in further detail for reason explained in the previous paragraph. However, several cellular and in vitro assays provided further evidence that TRIM28 is a TRIM family member that is devoid of ubiquitin E3 ligase activity and may be regarded as a “pseudo” E3 ligase.
WP3: Elucidate the molecular mechanism of MAGE-dependent modulation of TRIM28 E3 ligase activity
The goal of WP3 was to generate a molecular model of MAGE dependent regulation of TRIM28 activity using biochemical, biophysical, and structural approaches. However, despite extensive work aiming to reproduce results from the literature, no MAGE-dependent regulation TRIM28 activity could be shown. Furthermore, several biophysical techniques could not show reliable binding of recombinant MAGE proteins to TRIM28, and structural approaches did not yield a structural description of the complex.
Several alternative approaches to study the activity of TRIM28 alone were undertaken, but no activity of TRIM28 could be shown in diverse assays and structural studies on either isolated domains or the full-length protein did not yield structural insights due to the highly flexible nature of TRIM28.
As several of the experiments performed in WP1-3 did not yield the expected results, an alternative function of TRIM28 mediate by its unique C-terminal domains was analysed. Using a broad range of biochemical and structural experiments allowed me to show that TRIM28 does not work as a novel bona fide SUMO E3 ligase despite its apparent SUMOylation activity, but instead might serve as a multi-protein interaction platform to mediate substrate sumoylation at different genomic locations.
Although the project failed to successfully show MAGE-dependent regulation of TRIM28 activity, it identified some new key aspects of TRIM28. Together with other family members of class VI TRIMs as well as members of other TRIM sub-families, TRIM28 is in a unique position within TRIM ligases: as one of several inactive TRIMs, this raises more general interesting questions about the function of the inactive RING domain, which normally mediates ubiquitination activity. Together with other domains of these large multi-domain proteins, the inactive RING domain could adopt alternative functions, for example in protein-protein interactions linked to non-ubiquitin ligase functions of the proteins. Alternatively, the RING domain might have lost its function during the evolution of the TRIM family proteins.