Biomolecular NMR Assignments

, Volume 5, Issue 2, pp 245–248

NMR assignments of ubiquitin fold domain (UFD) in SUMO-activating enzyme subunit 2 from rice


  • Rintaro Suzuki
    • Protein Research UnitNational Institute of Agrobiological Sciences
  • Wataru Tsuchiya
    • Protein Research UnitNational Institute of Agrobiological Sciences
  • Heisaburo Shindo
    • Protein Research UnitNational Institute of Agrobiological Sciences
    • Protein Research UnitNational Institute of Agrobiological Sciences

DOI: 10.1007/s12104-011-9310-9

Cite this article as:
Suzuki, R., Tsuchiya, W., Shindo, H. et al. Biomol NMR Assign (2011) 5: 245. doi:10.1007/s12104-011-9310-9


The small ubiquitin-related modifier (SUMO) is a ubiquitin-like post-translational modifier that alters the localization, activity, or stability of many proteins. In the sumoylation process, an activated SUMO is transferred from SUMO-activating enzyme E1 complex (SAE1/SAE2) to SUMO-conjugating enzyme E2 (Ubc9). Among the multiple domains in E1, a C-terminal ubiquitin fold domain (UFD) of SAE2 shows high affinity for Ubc9, implying that UFD will be functionally important. We report NMR chemical shift assignments of UFD in SAE2 from rice. Almost all the resonances of UFD were assigned uniquely, representing a single conformation of UFD in solution. This is a contrast to the previous report for the corresponding UFD of human SAE2 which shows two conformational states. The secondary structure prediction of UFD in rice SAE2 shows the similar overall structure to the crystal structures of UFD in other E1 proteins such as SAE2 of human and yeast, ubiquitin-activating enzyme of yeast, and NEDD8-activating enzyme E1 catalytic subunit of human. Concomitantly, differences in the length of helices, strands, and loops are observed, particularly in the binding region to E2, supposing the variation in the UFD-E2 binding mode which may play a critical role in determining E1-E2 specificity.



Biological context

Small ubiquitin-related modifier (SUMO) is a member of a protein family consisting of ubiquitin and ubiquitin-like proteins. Sumoylation is a post-translational modification and alters functions of many cellular proteins through changes in localization, activity, or stability (Geiss-Friedlander and Melchior 2007). SUMO is first activated at its C-terminus by the SUMO-activating enzyme E1. SUMO E1 is a heterodimer composed of two subunits, SUMO-activating enzyme subunit 1 and 2 (SAE1 and SAE2). After thioester bond formation between the C-terminal carboxy group of SUMO and the catalytic Cys residue in SAE2, the thioester is transferred to the SUMO-conjugating enzyme E2 (Ubc9). Finally, Ubc9 transfers SUMO to the substrate protein, usually with assistance from SUMO E3 ligases.

It has been proposed that Ubc9 is recruited to the SUMO-E1 thioester-linked conjugate through its multiple binding sites or domains including the conjugated SUMO, the Cys domain of SAE2 with catalytic Cys residue, and the C-terminal ubiquitin fold domain (UFD) of SAE2. Among them, UFD showed the highest affinity for Ubc9 and thus it was considered to play an essential role in SUMO transfer to Ubc9 (Lois and Lima 2005; Wang et al. 2007).

There have been reported many crystal structures of UFDs of SUMO E1 (SAE2), ubiquitin E1 (ubiquitin-activating enzyme UBA1), and NEDD8 E1 (NEDD8-activating enzyme E1 catalytic subunit UBA3). However, no solution structure of UFD is available to date. Here we report a full-assignment of NMR spectra of UFD from rice to open the way to elucidate the three-dimensional structure of UFD and protein–protein interactions in solution.

Methods and experiments

Uniformly 13C/15N-labeled UFD, comprising residues 436-550 of rice SAE2, was expressed as a GST-fusion protein in E. coli and purified by GSH-Sepharose column, and the GST-tag was cleaved by PreScission Protease (GE Healthcare), followed by gel filtration chromatography. The recombinant protein thus obtained contains an additional amino-acid sequence of Gly-Pro-Leu-Gly-Ser at the N-terminus. Solutions used to record NMR spectra contained the purified recombinant UFD (1.18–1.28 mM) dissolved in 10 mM potassium phosphate buffer, pH 7.0, with 100 mM NaCl and 10% 2H2O.

All NMR spectra were acquired at 25°C on a Bruker DMX 750 spectrometer equipped with pulse-field gradient. 1H, 13C, and 15N sequential resonance assignments were obtained using two-dimensional (2D) and three-dimensional (3D) heteronuclear NMR experiments: 2D 1H-15N HSQC, 2D 1H-13C HSQC, 3D HNCO, 3D CBCA(CO)NH, 3D HNCA, 3D HNHB, 3D HCACO, 3D HCCH-TOCSY, 3D CCONH, 3D HCCONH, 3D HBHACONH, 3D 15N-separated NOESY-HSQC, 3D 13C/15N-separated NOESY-HSQC, and 3D 13C-separated NOESY-HSQC. The spectra were processed using NMRPipe software (Delaglio et al. 1995) and analyzed using SPARKY 3 software (Goddard and Kneller, University of California, San Francisco).

Assignments and data deposition

High-quality NMR data for the UFD were obtained, as illustrated in the 1H-15N HSQC spectrum (Fig. 1). Backbone assignments were completed for all residues except for C′ of N-terminal Gly originating from the expression vector and C′ of Leu467. The assignments of side chain resonances were completed up to 98.3%. Five proline residues out of six assumed trans-conformation as demonstrated by the chemical shifts of Cβ and Cγ atoms characteristic to trans-form (Schubert et al. 2002). Observation of strong HN-Hδ NOEs between residues Xaa-Pro confirmed this assumption. Similar NOEs were also observed for residue Pro468, although chemical shifts of the Cβ and Cγ atoms could not specify trans-conformation.
Fig. 1

1H-15N HSQC spectrum of UFD of rice SAE2 and its assignments. The spectrum was collected in 10 mM potassium phosphate buffer, pH 7, at 25°C on a Bruker DMX750 NMR spectrometer. Backbone resonance assignments are indicated in a one-letter amino acid code. Side-chain NH2 amide resonances of Asn and Gln, connected by horizontal lines, and side-chain resonanes of Trp and Arg are labeled in italic. Note that the two Arg side-chain amide signals shown in dashed contours are folded back under the experimental condition employed

As shown in Fig. 2, the regions of regular secondary structure of the UFD were derived from both the chemical shift index calculated by the program RCI (Berjanskii and Wishart 2005) and the prediction made by the program TALOS+ (Shen et al. 2009). The overall structure of UFD of rice SAE2 was similar to the X-ray crystal structures of UFD of human or yeast SAE2 (Olsen et al. 2010; Wang et al. 2010). It was also similar to secondary structures of the UFDs in human NEDD8-activating enzyme E1 catalytic subunit (UBA3) and in yeast ubiquitin-activating enzyme E1 (UBA1), except that the UFD of SAE2 has an additional β6-strand in the C-teminus (Huang et al. 2007; Lee and Schindelin 2008). However, the length of secondary structural elements as well as the connecting loops varies throughout the domain. Particularly, the difference among E2-binding regions of UFD marked with dots in Fig. 2 suggests considerable variations in interaction mode between UFD and E2.
Fig. 2

Sequence alignments of UFDs based on the secondary structures, residues (marked with dots) associated with binding to E2, and order parameters S2 for UFD of rice SAE2. Secondary structures predicted for UFD of rice SAE2 in solution using RCI and TALOS+ are compared with those observed for the UFD of human and yeast SAE2, human UBA3, and yeast UBA1 in the crystalline state (PDB ID: 3KYC, 3ONG, 2NVU, and 3CMM, respectively). Gray-colored secondary structures of the rice SAE2 was predicted by TALOS+ alone but not by RCI. The alignment is done manually so that the topologically equivalent residues in secondary structures are aligned. Conservatively substituted residues are highlighted in black background. The residues with small letters originate from the expression vector. The residues of ScSAE2 and HsUBA3 which have close contacts with E2 in crystal structures and the E2-binding region of HsSAE2 reported by Wang et al. (2009) are marked with dots. The order parameters S2 for the rice SAE2 predicted by the program RCI are shown above the alignment

UFD of human SAE2 showed two conformational states in the region K486-E499 comprising the Ubc9-binding surface (Wang et al. 2009). It was pointed out that this region is under equilibrium between minor unfolded conformer and major folded conformer containing β3-strand, and that such a conformational flexibility is important in the molecular recognition. In case of the UFD of rice SAE2, however, all residues were assigned uniquely in the 1H-15N HSQC spectrum and the conformational multiplicity was not observed, indicating that the UFD has a unique conformation or a rapid conformational equilibrium if any. The order parameters S2 estimated by RCI indicated considerable flexibility of the short region E482-L484 where no secondary structure was suggested by either RCI or TALOS+ programs (Fig. 2). These residues are located in the insertion at the C-terminus of the β3-strand. Although the flexibility of the loop may take part in the E2 recognition in rice, the dynamic property of the recognition site of UFD seems different among species. Further structural investigation of UFD in rice SAE2 is needed to uncover the character and the interaction mode with Ubc9.

The chemical shifts for the rice UFD have been deposited in the BioMagResBank ( under accession number 11427.

Copyright information

© Springer Science+Business Media B.V. 2011