Structural analysis of HERC2/UBE3A and HERC2/DOCK10 complexes provides new insights into the molecular basis of Angelman, Angelman-like and Dup15q Syndromes

UBE3A and HERC2 are two mutually interacting HECT E3 Ubiquitin ligases whose genes are altered in 15q11.2-13.1 Duplication (Dup15q) Syndrome, Angelman Syndrome (AS) and other phenotypically-related mental retardation syndromes. Using quantitative binding assays, X-ray crystallography and sequence conservation analysis, we show that the HERC2/UBE3A complex occurs in probably most animals with a central nervous system, via a conserved interface involving the RLD2 domain of HERC2 and a "DxDKDxD" motif of UBE3A. We found that HERC2 also recognizes and binds to similar DxDKDxD motifs within a handful of other proteins relevant to brain development (DOCK10, PCM1, USP35, BAZ2B, ARID4A, ARIP4, RERE and MYT1). We further investigated the interaction of HERC2 with DOCK10, a RAC1- and CDC42-GEF protein that regulates dendritic spine morphogenesis in hippocampal neurons. Both disruption of the HERC2-binding motif in DOCK10 and knockdown of HERC2 affected the GEF activity of DOCK10. We also show that the DOCK10-induced dendritic spine formation is dependent on its ability to bind HERC2. Structural modeling of full-length DOCK10, free or bound to either RLD2, CDC42 or RAC1 indicates that the GEF activation of full-length DOCK10 requires a conformational change that is stimulated by binding to HERC2. Based on our data, we propose that under pathological conditions, in developing brains with an abnormal dosage of either HERC2 or its dominant partner UBE3A, increased or decreased amounts of HERC2/DOCK10 complexes could lead to altered GTPase activation. This in turn could affect dendritic spine formation and neurodevelopment.