Professor Ron Hay FRS FRSE - Honorary Programme Leader
Research
Role of SUMO Modification in Gene Expression
My laboratory is interested in mechanisms that control the expression of our genes. In particular how the ubiquitin-like proteins are linked to important regulators of gene expression and how this influences which genes are switched on or off. We are particularly interested in how SUMO modification alters transcription and have identified and determined the mechanism of action of a number of the gene products required for SUMO modification and for deconjugation (Figure 1a). This work utilizes a diverse array of approaches ranging from X-ray crystallography to biochemistry and cell biology. We have proposed that transient SUMO modification may leave proteins with a “history of modification” even after the modification has been removed. The importance of these pathways is manifest in the many disease states that result when these processes fail.
SUMO Conjugation
Our laboratory has established conjugation with the Small Ubiquitin-like Modifier (SUMO) as an important regulatory mechanism in eukaryotes. By analysing the site of modification in a number of proteins we proposed a SUMO consensus modification site consisting of the sequence yKxE, where "y" represents a large hydrophobic amino acid and " x " represents any amino acid (Figure 1b). We further demonstrated that this site constitutes a transferable signal that confers the ability to be modified with SUMO on proteins to which it is linked. In chordates there are 3 members of the SUMO family. Although SUMO-2 and SUMO-3 are 97% identical they share only 50% sequence identity with SUMO-1 and appear to be functionally distinct. We demonstrated that in contrast to SUMO-1, SUMO-2 and SUMO-3 could form poly-SUMO-2 chains.
Figure 1 SUMO modification (a) The SUMO cycle of conjugation and deconjugation. SUMO (Su, yellow circle) is processed by a SUMO specific protease (SENP, red circle) prior to being activated and covalently linked to the SUMO E1 activating enzyme (SAE1/SAE2, green boxes). SUMO is then transferred to the SUMO E2 conjugating enzyme (Ubc9, grey ellipse) that carries out target (blue trapezoids) modification with the aid of a SUMO E3 ligase (E3, purple ellipse). Deconjugation to release free SUMO and target is mediated by a SENP (red circle). (b) Lysine acceptor residues subject to SUMO modification are usually found in the SUMO modification motif yKxE (where y is I, V or L). This motif is found in the N-terminal region of SUMO-2 and SUMO-3, but not in SUMO-1. Thus SUMO-2 (and SUMO-3) can form polymeric chains via conjugation at the yKxE motif.
SUMO specific ubiquitin E3 ligases
Although we reported the existence of these chains in 2001, it is only recently that their function has been revealed. We recognised that the RING domain containing protein Rnf4 also contained multiple SUMO interaction motifs (SIMs) and demonstrated that it could function as a ubiquitin E3 ligase with a unique specificity for polySUMO chains. We further showed that Rnf4 is the ubiquitin ligase responsible for arsenic inducible, proteasomal degradation of the Promyelocytic Leukaemia (PML) protein. In Acute Promyelocytic Leukaemia (APL) the PML protein is fused to the Retinoic Acid Receptor and the disease can be effectively treated by arsenic administration. Arsenic induces modification of PML with SUMO and subsequent proteasomal degradation of PML. Our identification of Rnf4 as the E3 ligase responsible for the SUMO-dependent degradation of PML provides the molecular basis for the therapeutic action of a drug currently used to treat leukaemia (Tatham et al., 2008). The objective of present work is to determine the signal transduction pathway, activated by arsenic, which leads to increased SUMO modification of PML. X-ray crystallography and NMR spectroscopy are being employed to determine the structure of the ubiquitin E3 ligase Rnf4, bound to its poly SUMO substrate and its cognate E2 conjugating enzyme.
Figure 2. A model for RNF4 mediated ubiquitylation of polySUMO-2 modified PML. Schematic showing PML protein with a linked poly-SUMO-2 chain being targeted for ubiquitylation by RNF4. RNF4 contains multiple SUMO Interaction Motifs (SIMs) and a RING domain.
SUMO targets
In research that is currently underway we are using stable cell lines containing TAP-tagged versions of SUMO (expressed at close to physiological levels) and Stable Isotope Labelling with Amino Acids in Cell culture (SILAC) coupled to high resolution mass spectrometry to carry out quantitative temporal analysis of the SUMO proteome as cells respond to various challenges. This is a productive area of research that provides a system wide view of SUMO modification, amenable to mathematical analysis. Many further analyses are planned: as cells progress through the cell cycle; exposure of cells to arsenic, DNA damaging agents and cytokines.
Work from our own and other laboratories indicated that proteins involved in transcriptional regulation were important targets for SUMO modification. The transcriptional regulators p300 and CPB contain domains responsible for transcriptional repression and we demonstrated that these regions contained two copies of the SUMO consensus modification sequence. Mutations that reduced SUMO modification, increased p300 mediated transcriptional activity and expression of a SUMO specific protease or catalytically inactive Ubc9 relieved repression, demonstrating that p300 repression was mediated by SUMO conjugation (Girdwood et al., 2003). We further demonstrated that SUMO modified p300 recruited a histone deacetylase and that it was this histone deacetylase that was responsible for transcriptional repression. This was the first insight into the mechanism by which SUMO modification could mediate transcriptional repression. Subsequent papers from other laboratories confirmed that this was a general mechanism of SUMO-dependent transcriptional repression. We are currently using whole genome siRNA screening to identify additional components of the SUMO transcriptional repression pathway.
SUMO E3 ligases
Figure 3. Identification of the RanBP2 binding site on Ubc9. (A) A representative region of 1H-15N TROSY spectra of Ubc9 showing the superposition of the spectra of Ubc9 when free (black peaks) and in complex with RanBP2 (red peaks). Resonances of unbound Ubc9 are labelled with the corresponding residue number. (B) Ribbon diagram of the three-dimensional structure of Ubc9 showing (in red) the regions of most significant chemical shift perturbation upon complex formation with RanBP2. The active site Cys93 in Ubc9 is shown in orange. Mutants of Ubc9 with residues at the RanBP2 binding interface fail to interact with RanBP2. Gel shift assay where RanBP2 in complex with Ubc9 has lower mobility than free RanBP2 allowing comparison of unbound and bound forms for each Ubc9 type (quantitation in right panel).
While the SUMO E1 and Ubc9 can conjugate SUMO-1, -2 and -3 onto target proteins SUMO E3 ligases are required for efficient modification. To establish the mechanism by which these ligases enhance modification, NMR was used to identify the surface of Ubc9 engaged by the SUMO ligase RanBP2 (Figure 3). In contrast to known ubiquitin E2-E3 interactions, RanBP2 bound to the “back” of Ubc9 on the opposite face to that used for substrate recognition. (Tatham et al., 2005). We demonstrated that this SUMO E3 ligase enhanced substrate modification, not by direct interaction with substrate, as is the case with most ubiquitin E3 ligases, but by engaging the Ubc9 thioester and aligning it in a configuration optimal for conjugation to substrate. Additional SUMO E3 ligases have been identified and their activation and mechanism of action are being investigated.
SUMO specific proteases
Figure 4. Structure of catalytically inactive SENP1 with SUMO substrates. Superposition of the C603A SENP1, SUMO-1-RanGAP1 complex with the C603A SENP1 SUMO-1 FL complex. In the C603A SENP1, SUMO-1-RanGAP1 complex RanGAP1 is shown in red ribbon, C603A SENP1 in blue ribbon and SUMO-1 in cyan ribbon. The isopeptide bond between K524 of RanGAP1 and G97 of SUMO1 is shown in sticks with carbon yellow, nitrogen blue and oxygen red. The C603A SENP1 SUMO-1 FL complex has SENP1 coloured in light blue and SUMO-1 in deep purple.
SUMO specific protease SENP1 processes SUMO-1, SUMO-2 and SUMO-3 to mature forms and deconjugates them from modified proteins. To establish the proteolytic mechanism we determined structures of catalytically inactive SENP1 bound to SUMO-1 modified RanGAP1 and to unprocessed SUMO-1 (Figure 4). In each case the scissile peptide bond is kinked at a right angle to the C-terminal tail of SUMO-1 and has the cis configuration of the amide nitrogens (Shen et al., 2006). This demonstrated how the scissile bond was manipulated into the correct orientation for the cleavage reaction. Biochemical analysis from this and other laboratories has indicated that the SUMO specific proteases SENP6 and SENP7 both display strong specificity for polySUMO-2 chains. Our present efforts are directed towards understanding the roles of these proteins in vivo and determining the structure of diSUMO-2 bound to catalytically inactive forms of SENP6 and SENP7.
NEDD8 specific proteases
Figure 5 Structure of NEDD8 specific protease NEDP1 (blue space fill) in a transition state complex with NEDD8 (pink ribbon diagram). It shows the C-terminus of NEDD8 disappearing into the NEDP1 cleavage tunnel.
The ubiquitin-like protein NEDD8 is essential for activity of SCF-like ubiquitin ligase complexes and affects the transcriptional activity of p53. While characterising what we initially thought were SUMO specific proteases we identified NEDP1, the first NEDD8 processing enzyme to be reported. NEP1 is highly conserved throughout evolution and equivalent proteins are present in yeast, plants, insects and mammals. NEP1 can process the NEDD8 precursor and deconjugates NEDD8 from a wide variety of substrates including the cullin component of SCF-like complexes, p53 and ribosomal proteins (Xirodimas et al., 2008). Although NEDD8 and ubiquitin are highly related in sequence and structure, their attachment to a protein leads to different biological effects. It is therefore critical that NEDP1 discriminates between NEDD8 and ubiquitin. To determine the basis of this specificity we determined the crystal structure of NEDP1 in isolation and in a transition state complex with NEDD8. Binding of NEDD8 induced a dramatic conformational change in a flexible loop of NEDP1 that swings over the C-terminus of NEDD8 locking it a structure optimal for catalysis (Figure 5). Structural, mutational and biochemical studies identified key residues involved in molecular recognition. This demonstrated how a single residue difference in the C-terminus of NEDD8 and ubiquitin contributed to the ability of NEDP1 to discriminate between these two highly related ubiquitin-like proteins (Shen et al., 2005).
Selected publications:
- Tatham, M.H., Geoffroy, M.C., Shen, L., Plechanovova, A., Hattersley, N., Jaffray, E.G., Palvimo, J.J. and Hay, R.T. (2008). RNF4 is a poly-SUMO-specific E3 ubiquitin ligase required for arsenic-induced PML degradation. Nat Cell Biol. 10, pp. 538-546.
- Shen, L., Tatham, M.H., Dong, C., Zagorska, A., Naismith, J.H., Hay, R.T. (2006). SUMO protease SENP1 induces isomerization of the scissile peptide bond. Nat Struct Mol Biol. 13, pp. 1069-1077.
- Shen, L.N., Liu, H., Dong, C., Xirodimas, D., Naismith, J.H. and Hay, R.T. (2005). Structural basis of NEDD8 ubiquitin discrimination by the deNEDDylating enzyme. NEDP1. EMBO J. 24, pp. 1341-1351.
- Tatham, M.H., Kim, S., Jaffray, E., Song, J., Chen, Y., Hay, R.T. (2005). Unique binding interactions among Ubc9, SUMO and RanBP2 reveal a mechanism for SUMO paralog selection. Nat Struct Mol Biol. 12, pp. 67-74.
- Mendoza, H.M., Shen, L.N., Botting, C., Lewis, A., Chen, J., Ink, B. and Hay, R.T. (2003). NEDP1, a highly conserved cysteine protease that deNEDDylates Cullins. J Biol Chem. 278, pp. 25637-25643.
- Girdwood, D., Bumpass, D., Vaughan, O.A., Thain, A., Anderson, L.A., Snowden, A.W., Garcia-Wilson, E., Perkins, N.D. and Hay, R.T. (2003). P300 transcriptional repression is mediated by SUMO modification. Mol Cell. 11, pp. 1043-1054.
University of Dundee