SUMO-SIM interactions and their role in substrate recognition and autoregulation of SUMO-dependent ubiquitin ligases

我们以及其他实验室对酵母和哺乳动物蛋白质的研究已经确定了SUMO相互作用基序（SIMs），该基序介导蛋白质与小泛素相关修饰子SUMO的非共价相互作用。这种SIMs的存在被发现是一类新的泛素连接酶（ULS）的特征，它特异性地识别sumoylated蛋白。我们和其他人将人类RNF4蛋白定性为第一个哺乳动物ULS，并确定了早幼粒细胞白血病蛋白（PML）作为其底物。量热分析显示，RNF4通过其多个SIMs优先结合到SUMO链上。生物信息学分析确定了其他几种将泛素连接酶结构域与SIMs结合的哺乳动物蛋白。这些连接酶中有几种会影响酵母中表达的SUMO偶联物水平或人类细胞中定位于PML小体的SUMO偶联物水平，这表明SUMO- sim相互作用与其功能有关。我们观察到酵母和人的ULS蛋白本身都被聚合、泛素化和降解，这表明它们受到自调节反馈控制。我们将结合突变生化分析来描述SIMs个体或组合在识别底物识别中不同类型的SUMO修饰和这些泛素连接酶的自SUMO化中的作用。我们将进一步研究生理条件（如各种应激类型）如何影响酵母和人类细胞中ULS蛋白的蛋白水解控制。我们已经建立了一个数据库筛选方案，可以识别高概率的SIMs，并应用该方法预测酵母基因组中编码的SUMO相互作用蛋白。我们将验证选定的候选物，并通过将其应用于人类和其他模式生物的蛋白质组学分析来扩展这种方法。此外，我们将修改方法来识别非共识的SIMs和功能相关的基序。

Studies on yeast and mammalian proteins in our as well as other laboratories have identified SUMO interaction motifs (SIMs) that mediate non-covalent interaction of proteins with the small ubiquitin-related modifier SUMO. Presence of such SIMs was found to be a characteristic feature of a novel class of ubiquitin ligases (ULS) that specifically recognize sumoylated proteins. We and others characterized the human RNF4 protein as the first mammalian ULS and identified the Promyelocytic Leukemia Protein (PML) as its substrate. Calorimetric assays revealed that RNF4 via its multiple SIMs preferentially binds to SUMO chains. Bioinformatical analyses identified several other mammalian proteins that combine ubiquitin ligase domains with SIMs. Several of these ligases affect SUMO conjugate levels upon expression in yeast or localized to PML bodies in human cells suggesting that SUMO-SIM interaction is relevant to their function. We observed both for yeast and human ULS proteins that they are themselves sumoylated, ubiquitylated and degraded suggesting that they subject to an auto-regulatory feedback control. We will use a combination of mutational biochemical analyses to characterize the role of individual or combinations of SIMs in the recognition of distinct types of SUMO modification in substrate recognition and auto-sumoylation of these ubiquitin ligases. We will further investigate how physiological conditions such as various stress types affects proteolytic control of ULS proteins in yeast and human cells. We have established a database screening protocol that allows to identify high probability SIMs and applied this method to predict SUMO interacting proteins encoded in the yeast genome. We will validate selected candidates and extend this approach by applying it to the analysis of the proteomes of humans and other model organisms. Furthermore, we will modify the approach to identify non-consensus SIMs und functionally related motifs.