Structure and Analysis of a Complex between SUMO and Ubc9 Illustrates Features of a Conserved E2-Ubl Interaction

Abstract

SummaryThe SUMO E2 Ubc9 serves as a lynchpin in the SUMO conjugation pathway, interacting with the SUMO E1 during activation, with thioester linked SUMO after E1 transfer and with the substrate and SUMO E3 ligases during conjugation. In this manuscript, we describe the structure determination of a non-covalent complex between human Ubc9 and SUMO-1 at 2.4 Å resolution. Non-covalent interactions between Ubc9 and SUMO are conserved in human and yeast insomuch as human Ubc9 interacts with each of the human SUMO isoforms, and yeast Ubc9 interacts with Smt3, the yeast SUMO ortholog. Structural comparisons reveal similarities to several other non-covalent complexes in the ubiquitin pathway, suggesting that the non-covalent Ubc9-SUMO interface may be important for poly-SUMO chain formation, for E2 recruitment to SUMO conjugated substrates, or for mediating E2 interactions with either E1 or E3 ligases. Biochemical analysis suggests this surface is less important for E1 activation or di-SUMO-2 formation, but more important for E3 interactions and for poly-SUMO chain formation when the chain exceeds more than two SUMO proteins.

Identification of a molecular recognition feature in the E1A oncoprotein that binds the SUMO conjugase UBC9 and likely interferes with polySUMOylation

“…This MoRF is necessary and sufficient to interact with the SUMO conjugase UBC9. E1A binds UBC9 by a surface previously shown to bind SUMO noncovalently and has a role in polySUMOylation (Capili and Lima, 2007;Knipscheer et al, 2007). Our results suggest that this viral MoRF mimics an interaction normally present in uninfected cells and functions to inhibit polySUMOylation.…”

“…The N-terminal helix of UBC9 shows a weak noncovalent interaction with the C-terminus of SUMO (Capili and Lima, 2007;Knipscheer et al, 2007). As this helix is also important for binding E1A, we compared the sequence of E1A with the C-terminus of SUMO.…”

“…We tested UBC9 point mutants that affect the surface that binds SUMO noncovalently for their ability to bind E1A. The R13E, R17E and H20D UBC9 mutants that do not bind SUMO (Capili and Lima, 2007;Knipscheer et al, 2007) also failed to interact with E1A (Figure 7a). Furthermore, the G23R and V25R mutants that retained noncovalent SUMO binding (Knipscheer et al, 2007) still bound E1A (Figure 7a).…”

“…In the crystal structure of the noncovalent association of UBC9 with SUMO-1 (Capili and Lima, 2007) (accession number 2pe6), R17 and H20 in UBC9 make ionic contacts with D86 and V87 in SUMO-1, respectively. D86 and V87 correspond to E118 and V119 in the E1A peptide, which explains why mutation of R17 and H20 in UBC9 also affected E1A binding.…”

Twists and turns in ubiquitin‐like protein conjugation cascades

“…These include a variety of types of interactions between E2s and UBLs. [99][100][101][102] Also, pathwayspecific E2-E3 interactions outside of the catalytic domains may tether an E2 to a nonconserved E3 domain during E1-mediated UBL loading, and allowing the UBL-loaded E2's rapid return to a conserved E3 catalytic domain for UBL transfer.…”