A Structure-Function Analysis of the Meiosis-Specific E3 SUMO Ligase Zip3
During meiotic prophase I, pairing of homologous chromosomes and interhomolog recombination (crossovers) are essential for proper chromosome segregation. During these landmark events, the assembly of the Synaptonemal Complex (SC), a conserved tripartite proteinaceous structure, forms between homologous chromosomes. The processes of SC assembly and crossover recombination are intimately linked, but the molecular and functional relationship between the two is obscure. In budding yeast, the E3 SUMO ligase Zip3 has been shown to have roles in both SC assembly and recombination, however, the molecular interactions that enable its functions in both processes are unknown. Here, we report the results of a complete structure-function study that identifies key regions of Zip3 important for SC assembly and Ecm11 SUMOylation (important for SC assembly) and Msh4 phosphorylation and interaction with the SC transverse filament protein Zip1 (important for SC assembly and recombination). In contrast to expectations, we found that multiple domains across the length of Zip3 protein are required for each of the Msh4 phosphorylation, SC assembly, and Ecm11 SUMOylation functions, suggesting that Zip3 binds multiple partners and that all these partner interactions are important for the completion of each Zip3 meiotic function. Of these interactions, we hypothesize the identity of three potential interactors and their binding sites on the Zip3 protein: First, the SC transverse filament protein Zip1 might indirectly interact with SP-RING domain of Zip3 using the mediator proteins the E2 SUMO ligase Ubc9 and SUMO. This interaction is necessary for Zip3 binding to chromosomes and for Zip3/Zip1 co-localization at polycomplex. Second, the Zip3 alpha-helical region (residues 101-150) is also likely involved in recruiting the Zip2 protein, since the Msh4 protein, which is recruited by Zip2, fails to co-localize robustly in mutants missing this Zip3 region. This interaction is important for Msh4 phosphorylation since Msh4 is phosphorylated after recruitment to recombination sites. Third, Zip3 residues 100-482, excluding F231 and residues 388-400, likely bind an unknown SUMO ligation factor and enable its regulatory SUMOylation via Zip3, resulting in the attenuation of Ecm11 SUMOylation. During our in-depth analysis of Msh4 phosphorylation, we also believe we identified a further interactor of Zip3: Some small mutations of Zip3 seem to modulate Msh4 modification, producing only partially phosphorylated Msh4 in comparison to that seen in wildtype. We theorize that Zip3 helps recruit phosphorylation machinery to recombination sites and brings the machinery and Msh4 together, subsequently allowing Msh4 to be properly modified. Mutation of Zip3 in some domains seem to only hamper, rather than eliminate, this process. Besides identifying protein partners of Zip3, we also identified mutants of Zip3 that uncharacteristically altered its meiotic functions. Mutation of the coiled-coil domain created a mutant that failed to assemble SC at recombination sites while aggressively blocked SC assembly at centromere sites, producing less total SC than the zip3 null strain. Mutants in the alpha-helical region, especially deletion of residues 122-136, showed full SC assembly while lacking almost all Msh4 phosphorylation, showing an uncoupling of the SC assembly and crossover recombination processes since phosphorylated Msh4 is usually necessary for both. Taken all this data together, Zip3 clearly shows different regions of the protein promote SC assembly and crossover recombination, further solidifying Zip3 as a bridging factor between these processes. We propose a molecular model in which each region of the Zip3 protein is necessary for the nucleation and maturation of the recombinosome, a complex of pro-SC and pro-crossover factors localized to chromosomes. Within this model, the absence of any region of Zip3 impacts either SC assembly, crossover recombination, or both processes.