Solution structure of SUMO from Trypanosoma brucei and its interaction with Ubc9.
Qiang Shang,Chao Xu,Jiahai Zhang,Xuecheng Zhang,Xiaoming Tu
DOI: https://doi.org/10.1002/prot.22409
2009-01-01
Abstract:Post-translational modification by SUMO (small ubiquitin-related modifier) is an important mechanism that regulates a wide variety of cellular functions, such as nuclear transport, gene expression, stress response, cell cycle control, oncogenesis, and response to virus infection.1-3 SUMO proteins are conjugated to target proteins by an enzymatic cascade involving SUMO activating enzyme E1 (Aos1/Uba2, also named SAE1/SAE2),4-6 SUMO conjugating enzyme E2 (Ubc9),7, 8 and E3 ligase. Trypanosoma brucei, the causative agent of African sleeping sickness, is a unicellular protozoan. The sequences of SUMO in Trypanosoma brucei (Tb-SUMO) are 37% and 33% identical with that of human SUMO-1 and yeast Smt3, respectively (Supporting Information Figure S1A). Here, the solution structure of truncated Tb-SUMO (residues 7–108) from T. brucei was determined by NMR spectroscopy. Its interaction surface with human Ubc9 was also determined by chemical shift perturbation using NMR spectroscopy. This is the first structure of SUMO determined in protist. The 3D structure of Tb-SUMO is highly conserved with that of SUMO-1 and Smt3. Tb-SUMO interacts with human Ubc9 through residues located on the β-sheet of Tb-SUMO, which is also similar to that of SUMO-1 and Smt3. All these results suggest the evolutionary conservation of SUMO and sumoylation in eukaryotes. The DNA encoding residues 7–108 of SUMO in T. brucei was cloned into the expression vector pET22b(+) (Novgen). The recombinant vector was transformed into E.coli BL21 (DE3) for expression. Recombinant Tb-SUMO has 110 amino acid residues, including a C-terminal His tag (LEHHHHHH). The Tb-SUMO proteins were purified with a Ni-chelating column (Qiagen). NMR samples contained 0.6 mM Tb-SUMO, 50 mM phosphate buffer (pH 6.25), 100 mM sodium chloride, 2 mM EDTA in either 90% H2O/10%D2O or 100% D2O.9 Human GST-Ubc9 fusion protein was produced in E. coli DH5α cells.10 The fusion protein was purified on a GSH-Sepharose 4B column (Pharmacia) according to the manufacturer's specification. The fusion protein was digested with 50 units/mL thrombin (Pharmacia) for 16 h at 20°C. All NMR data were collected at 298 K on a Bruker DMX500 spectrometer with self-shielded z-axis gradients. NMR data processing was carried out using NMRPipe and NMRDraw. [1H-15N] HSQC, HNCACB, CACB (CO)NH, HNHA, HCC(CO)NH, H(CC)CONH, 3D 15N-edited NOESY, and 13C-edited NOESY spectra were required for structure analysis. The data were analyzed with Sparky 3 software. All software was run on a Linux system. Structures were calculated using CYANA.11 Initially, only typical medium- and long-range NOEs defining secondary structures from 13C-edited NOESY spectrum were used together with NOEs obtained from 15N-edited NOESY for structure calculation. Dihedral angle restraints, hydrogen bonds, and all other NOEs were introduced in consecutive steps. The statistics on experimental constraints, coordinate precision, and stereochemical quality of 20 structures with lowest energy were analyzed with MOLMOL and PROCHECK. To investigate the human Ubc9 binding site on Tb-SUMO, 0.4 mM 15N-labeled Tb-SUMO was prepared. Unlabeled human Ubc9 was titrated into the sample containing Tb-SUMO. Six different molar ratios of Ubc9 and Tb-SUMO (1:6, 1:3, 1:2, 2:3, 5:6, and 1:1) were designed. The concentration of both proteins at the end of the titration was ∼0.25 mM. The final molar ratio of Ubc9 and Tb-SUMO is about 1:1. The HSQC spectra for the binding surface on Tb-SUMO were recorded at 298 K. The 3D structure was calculated using CYANA 2.1.11 About 97% chemical shifts in backbone and 95% chemical shifts in total were assigned. The best-fit superposition of the ensemble of the 20 lowest-energy calculated structures is shown in Figure 1A. The atomic coordinates have been deposited in the Protein Data Bank as entry 2K8H. The statistical parameters in Table I showed a high-quality NMR structure of Tb-SUMO. The NMR solution structure of Tb-SUMO consists of two α-helices (α1, 56–67; α2, 88–91) and one mixed β-sheet (β1, 33–39; β2, 43–49; β3, 73–76; β4, 80–82; β5, 96–102) [Fig. 1(B)]. The β-sheet consists of five β-strands composed of three antiparallel β-strands and one parallel β-strand. The structure represents the typical ubiquitin-like fold. Pairwise backbone (residues L32–G108) Cα RMSD between the average NMR structure of Tb-SUMO and the average NMR structure of human SUMO-1 is 2.5 Å. For the same residue range, the RMSD value between Tb-SUMO and Smt3 is 3.1 Å. The structural comparison indicates that structure of Tb-SUMO is highly similar with that of human SUMO-1 and yeast Smt313 (Supporting Information Figure S1B and 1C), suggesting the conservation of SUMO protein among eukaryotes. NMR structure of Tb-SUMO: A: Backbone superposition of 20 selected conformers (residue range, 31–108) from the final CYANA 2.111 calculation; B: ribbon representation of the minimized averaged structure with the secondary structure elements highlighted. This figure was produced with MOLMOL.12 Previous studies indicate that Ubc9 can interact with SUMO proteins through noncovalent binding.14-16 No Ubc9 homologue has been identified in T. brucei yet. Therefore, the human Ubc9 was chosen to investigate the interface between Tb-SUMO and Ubc9 in this study. The chemical shift perturbation was used to map the Ubc9-binding interface on Tb-SUMO. Residues 31, 38–43, 75–80, 92–102, and 104 have the most significant changes in chemical shifts on complex formation with Ubc9 (Figure S2). Similar to other SUMO, these residues are located in β strands 1, 3, 5, the loop between β1 and β2 and the region between β3 and β5 [Fig. 2(A); Supporting Information Figure S3A]. Among the surface residues affected by interaction between Ubc9 and Tb-SUMO, six negatively charged residues, D41, D78, E94, D95, I99, and D100, are highly conserved among SUMO family members [Fig. 2(B)]. These residues, together with E89, D96, D97, and E104, constitute a large negative potential surface (Supporting Information Figure S3B and S3C). The result suggests an important role for electrostatic attraction in the interaction between Tb-SUMO and human Ubc9. The binding surface on Ubc9 has been reported previously.15 These residues are mainly on the helix α1, β-strands 1–3, and loop between α1 and β1. These highly conserved and positive residues on this region constitute a large positive potential surface. This surface is apparently compatible with the negative potential surface of Tb-SUMO. Besides, there are some hydrophobic residues affected significantly in this region. The complementary electrostatic potential plus hydrophobic interaction contribute to the specific binding of Tb-SUMO to human Ubc9. Ubc9 binding region and conserved residues on the surface of Tb-SUMO: A: residues whose combined chemical shift changes were more than the mean value are colored in red; B: residues which are conserved on the Ubc9 binding surface are colored in dark pink. Another interesting observation in this study comes from the N-terminal region of Tb-SUMO. Residue T30, which is located on the N-terminal region near to the β1-sheet, exhibits an obvious change on Ubc9 binding. Residue T30 is a polar residue and might contribute to the hydrophobic interaction between SUMO and Ubc9. This is unique for Tb-SUMO since N-terminal region of SUMO from other organisms has never been found to interact with Ubc9. One obvious discrepancy between SUMO and ubiquitin lies in the N-terminal extension in SUMO proteins. Whether the extension has special function is not clear. Tb-SUMO has the longer N-terminal extension than human and yeast SUMO proteins. The unique feature of N-terminal extension of Tb-SUMO regarding its interaction with Ubc9 might reflect the divergence during its evolution in eukaryotes. Gene expression in eukaryotes is regulated at several different levels including transcriptional level and post-translational level. However, in T. brucei, the lack of transcriptional factors implies the deficiency of regulation at transcriptional level which makes the post-translational regulation such as sumoylation become extremely more important in regulating the gene expression. A large body of evidence has indicated that SUMO is essential for many important nuclear events, including DNA replication and repair, kinetochore-tubulin attachment, genome integrity maintenance.1, 13, 17 Independent studies in our laboratory indicated that Tb-SUMO plays an essential role in the mitosis of cell cycle in T. brucei (Liao et al., unpublished dada). Our study, based on the sequence alignment, the similarity of Tb-SUMO structure compared with other SUMO structure and the conservation of molecular mechanism of interaction between SUMO and E2 conjugating enzyme (Ubc9), confirms the existence of SUMO and imply a SUMO system in T. brucei. In conclusion, the first solution structure of Tb-SUMO from protist has been determined. The 3D structure of Tb-SUMO and its interaction with human Ubc9 exhibit highly similarity among all SUMO from eukaryotes. All these accumulated data confirm the conservation of SUMO and sumoylation from protist to vertebrate. The authors thank F. Delaglio and A. Bax for providing NMRPipe and NMRDraw software. They also thank Goddard and Kneller for providing SPARKY software. Additional Supporting Information may be found in the online version of this article. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.