Since SUMO was simultaneously identified by different research groups in 1996 the number of proteins that have been found to be SUMO-modified has continued to grow and grow. This list now contains several thousand proteins covering a wide variety of functions. While most of SUMO’s functions are associated with nuclear organisation and transcript processing Tsur et al (link is external). show in a paper published in Developmental Cell that SUMO also functions away from the nucleus, at adherens junctions.
Adherens junctions are protein complexes that occur at the cell-cell junctions in epithelial cells. They interact with neighbouring cells and the cell’s underlying actin cytoskeleton through a core protein complex of E-cadherin, ?-catenin and ?-catenin. These junctions play a crucial role in the development and maintenance of epithelial integrity, but to allow tissue remodelling and cell migration to take place these cell-cell contacts have to be disassembled.
Broday and her lab previously used a proteomics approach to profile SUMO targets in C.elegans. This study was intended to follow up on their initial findings by examining the effect of mutating the SUMO protease, ulp-2, to mimic constitutive sumoylation. When the team mutated this SUMO protease, they found that it resulted in severe embryonic morphogenic defects, which were due at least in part to the maintenance of E-cadherin sumoylation.
The endocytosis and recycling of E-cadherin has previously been shown to be essential to the turnover of adherens junctions in cell culture and model systems. Tsue et al. identified three lysine residues in the tail of HMR-1, the C.elegans homologue to E-cadherin, that can be sumoylated. They demonstrated that sumoylation of the HMR-1 tail reduces its interaction with the ?-catenin homologue, HMP-2. In addition to sumoylation of the three lysine residues in the tail of HMR-1, the neighbouring serine residue must also be phosphorylated to allow a stable interaction with HMP-2, demonstrating the importance of cross-talk between these two signalling systems.
To test the potential role of sumoylation and desumoylation in vivo they examined the role of E-cadherin sumoylation at two distinct stages- embryo morphogenesis and epithelial polarisation. Mutating the sumo protease, ULP-2, or expressing a HMR-1:Sumo chimera, to mirror constitutive E-cadherin sumoylation, slows down AJ formation. Embryos that were able to compact their adherens junctions then failed to reorganise these junctions to the apical membrane of the cell for polarisation. ULP-2 overexpression, which results in no sumoylation of the E-cadherin homologue, also gives rise to similar phenotypes in C.elegans. Taken together these results suggest that HMR-1 may need to be continuously sumolyated and desumolyated to allow epithelial morphogenesis to proceed.
This study places sumoylation at centre stage in the maintenance and dynamics of adherens junctions. It suggests that sumoylation might carry out its role in adherens junctions by affecting how E-cadherin and ?-catenin interact, and so sumoylation might impinge on how ?-catenin/ ?-catenin link E-cadherin to the cytoskeleton. As the strength of adherens junctions depends in part on the formation of E-cadherin cis-complexes, continuous sumoylation might weaken these junctions by affecting their formulation.
Stemming from this work are questions remaining to be explored, such as: whether sumoylation affects the mechnosensitivity of adherens junctions?; what controls the cycle of sumoylation and desumoylation? and what role the cycles of sumoylation and desumoylation play in the epithelial-mesenchymal transitions in cancer development?
This is the only image I can find of an adherens junction that is available for reuse, but it shows all the main components with cadherins linking the cells and the attachment to the cytoskeleton.