Portal de Periódicos da CAPES

Structural and Functional Analysis of the Human Mitotic-specific Ubiquitin-conjugating Enzyme, UbcH10

2002; Elsevier BV; Volume: 277; Issue: 24 Linguagem: Inglês

10.1074/jbc.m109398200

1083-351X

Yaqiong Lin, W.C. Hwang, Ravi Basavappa,

Endoplasmic Reticulum Stress and Disease

Cell cycle progression is controlled at several different junctures by the targeted destruction of cell cycle regulatory proteins. These carefully orchestrated events include the destruction of the securin protein to permit entry into anaphase, and the destruction of cyclin B to permit exit from mitosis. These destruction events are mediated by the ubiquitin/proteasome system. The human ubiquitin-conjugating enzyme, UbcH10, is an essential mediator of the mitotic destruction events. We report here the 1.95-Å crystal structure of a mutant UbcH10, in which the active site cysteine has been replaced with a serine. Functional analysis indicates that the mutant is active in accepting ubiquitin, although not as efficiently as wild-type. Examination of the crystal structure reveals that the NH2-terminal extension in UbcH10 is disordered and that a conserved 310-helix places a lysine residue near the active site. Analysis of relevant mutants demonstrates that for ubiquitin-adduct formation the presence or absence of the NH2-terminal extension has little effect, whereas the lysine residue near the active site has significant effect. The structure provides additional insight into UbcH10 function including possible sites of interaction with the anaphase promoting complex/cyclosome and the disposition of a putative destruction box motif in the structure. Cell cycle progression is controlled at several different junctures by the targeted destruction of cell cycle regulatory proteins. These carefully orchestrated events include the destruction of the securin protein to permit entry into anaphase, and the destruction of cyclin B to permit exit from mitosis. These destruction events are mediated by the ubiquitin/proteasome system. The human ubiquitin-conjugating enzyme, UbcH10, is an essential mediator of the mitotic destruction events. We report here the 1.95-Å crystal structure of a mutant UbcH10, in which the active site cysteine has been replaced with a serine. Functional analysis indicates that the mutant is active in accepting ubiquitin, although not as efficiently as wild-type. Examination of the crystal structure reveals that the NH2-terminal extension in UbcH10 is disordered and that a conserved 310-helix places a lysine residue near the active site. Analysis of relevant mutants demonstrates that for ubiquitin-adduct formation the presence or absence of the NH2-terminal extension has little effect, whereas the lysine residue near the active site has significant effect. The structure provides additional insight into UbcH10 function including possible sites of interaction with the anaphase promoting complex/cyclosome and the disposition of a putative destruction box motif in the structure. ubiquitin-activating enzyme ubiquitin carrier protein ubiquitin-protein isopeptide ligase ubiquitin-conjugating enzyme anaphase promoting complex/cyclosome 4-morpholineethanesulfonic acid Ubiquitin-mediated proteolysis regulates cell cycle progression at several key control points. At least two such control points occur in mitosis. One is at the transition from metaphase to anaphase and the other is at the exit from mitosis (for reviews, see Refs. 1King R.W. Deshaies R.J. Peters J.M. Kirschner M.W. Science. 1996; 274: 1652-1659Crossref PubMed Scopus (1117) Google Scholar, 2Townsley F.M. Ruderman J.V. Trends Cell Biol. 1998; 8: 238-244Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar, 3Hershko A. Philos. Trans. R. Soc. Lond. B Biol. Sci. 1999; 354: 1571-1575Crossref PubMed Scopus (80) Google Scholar, 4Nasmyth K. Peters J.M. Uhlmann F. Science. 2000; 288: 1379-1385Crossref PubMed Scopus (377) Google Scholar, 5Amon A. Nat. Cell. Biol. 2001; 3: E12-E14Crossref PubMed Scopus (22) Google Scholar). At the transition from metaphase to anaphase, the securin protein in the securin-separase protein complex is destroyed to release separase. The freed separase cleaves the protein complexes binding the sister chromatids together. Cleavage of these protein complexes is thought to facilitate sister chromatid segregation, and hence entry into anaphase. To exit from mitosis, cyclin B in the cyclin B-cdc2 complex must be destroyed. Destruction of cyclin B results in the inactivation of the cdc2 kinase. The inactivation of cdc2 is an essential event for resetting the cell cycle machinery (6Noton E. Diffley J.F. Mol. Cell. 2000; 5: 85-95Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar).To accomplish the ubiquitination of securin and cyclin B (and other proteins targeted for ubiquitin-mediated destruction), three enzyme activities, designated E11(ubiquitin activating enzyme), E2 (ubiquitin conjugating enzyme, Ubc), and E3 (ubiquitin ligase), must work in concert (for review, see Ref.7Hershko A. Ciechanover A. Annu. Rev. Biochem. 1998; 67: 425-479Crossref PubMed Scopus (6793) Google Scholar). The E1 protein activates ubiquitin and then transfers it to the E2 protein. The ubiquitin forms an adduct to the E2 protein via a thiol ester linkage between the active site cysteine of E2 and the carboxyl terminus of ubiquitin. The E2 then donates the ubiquitin to the target protein, either directly or in conjunction with the E3 activity. In some instances, the same protein possesses both the E2 and E3 activity. Ultimately, a polyubiquitin-target protein conjugate is formed that then is recognized by the proteasome. The proteasome hydrolyzes the target protein and releases free ubiquitin. Whereas ubiquitin and E1 are highly conserved proteins, each eukaryotic organism contains several different E2 and E3 activities. The various E2 and E3 proteins function in cognate pairs and provide specificity in target protein ubiquitination.In the case of mitotic destruction of securin and cyclin B, the same E2 and E3 activities are thought to be responsible for the destruction of both proteins. The E3 activity is contained in a large multisubunit complex, termed the anaphase promoting complex or cyclosome (APC/C). The target protein specificity for ubiquitination seems to be conferred by different particular subunit compositions of the APC/C. The mitotic E2 proteins have been identified in several organisms, including human (UbcH10) (8Townsley F.M. Aristarkhov A. Beck S. Hershko A. Ruderman J.V. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2362-2367Crossref PubMed Scopus (191) Google Scholar), clam (E2-C) (9Aristarkhov A. Eytan E. Moghe A. Admon A. Hershko A. Ruderman J.V. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4294-4299Crossref PubMed Scopus (121) Google Scholar), mouse (mE2-C) (10Arvand A. Bastians H. Welford S.M. Thompson A.D. Ruderman J.V. Denny C.T. Oncogene. 1998; 17: 2039-2045Crossref PubMed Scopus (78) Google Scholar), Xenopus(UbcX) (11Yu H. King R.W. Peters J.M. Kirschner M.W. Curr. Biol. 1996; 6: 455-466Abstract Full Text Full Text PDF PubMed Scopus (158) Google Scholar), Schizosaccharomyces pombe (Ubc4) (12Osaka F. Seino H. Seno T. Yamao F. Mol. Cell. Biol. 1997; 17: 3388-3397Crossref PubMed Scopus (44) Google Scholar), and goldfish (E2-C) (13Tokumoto M. Nagahama Y. Tokumoto T. FEBS Lett. 1999; 458: 375-377Crossref PubMed Scopus (12) Google Scholar). These mitotic proteins are essential for cell cycle progression since mutation of the active site cysteine confers a dominant negative phenotype (8Townsley F.M. Aristarkhov A. Beck S. Hershko A. Ruderman J.V. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2362-2367Crossref PubMed Scopus (191) Google Scholar, 14Yamanaka A. Hatakeyama S. Kominami K. Kitagawa M. Matsumoto M. Nakayama K. Mol. Biol. Cell. 2000; 11: 2821-2831Crossref PubMed Scopus (46) Google Scholar).The E2 protein is remarkable in that, despite its relatively small size (typically ∼20 kDa), it must interact with 3 or 4 different proteins, namely ubiquitin, E1, E3, and perhaps the target protein. Therefore, the E2 protein must maintain structural features that allow interactions with the common elements of the system, ubiquitin and E1, and yet specify interactions with its cognate E3 and target protein. Although the crystal structures of several Ubc proteins have been determined and examination of their structures has given much insight into the function of this family of proteins (15Cook W.J. Jeffrey L.C. Sullivan M.L. Vierstra R.D. J. Biol. Chem. 1992; 267: 15116-15121Abstract Full Text PDF PubMed Google Scholar, 16Cook W.J. Jeffrey L.C., Xu, Y. Chau V. Biochemistry. 1993; 32: 13809-13817Crossref PubMed Scopus (83) Google Scholar, 17Cook W.J. Martin P.D. Edwards B.F. Yamazaki R.K. Chau V. Biochemistry. 1997; 36: 1621-1627Crossref PubMed Scopus (60) Google Scholar, 18Tong H. Hateboer G. Perrakis A. Bernards R. Sixma T.K. J. Biol. Chem. 1997; 272: 21381-21387Abstract Full Text Full Text PDF PubMed Scopus (117) Google Scholar, 19Worthylake D.K. Prakash S. Prakash L. Hill C.P. J. Biol. Chem. 1998; 273: 6271-6276Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 20Giraud M.F. Desterro J.M. Naismith J.H. Acta Crystallogr. D Biol. Crystallogr. 1998; 54: 891-898Crossref PubMed Scopus (55) Google Scholar, 21Jiang F. Basavappa R. Biochemistry. 1999; 38: 6471-6478Crossref PubMed Scopus (22) Google Scholar, 22Moraes T.F. Edwards R.A. McKenna S. Pastushok L. Xiao W. Glover J.N. Ellison M.J. Nat. Struct. Biol. 2001; 8: 669-673Crossref PubMed Scopus (136) Google Scholar, 23VanDemark A.P. Hofmann R.M. Tsui C. Pickart C.M. Wolberger C. Cell. 2001; 105: 711-720Abstract Full Text Full Text PDF PubMed Scopus (266) Google Scholar), much still remains to be understood about the stereochemical basis of E2 function. Here we report the 1.95-Å crystal structure determination of the mitotic specific E2 protein from humans, UbcH10, an essential protein for cell cycle progression. We also report complementary functional analysis of select mutants. These studies provide new insight into E2 function.RESULTS AND DISCUSSIONWe have determined the crystal structure of a mutant UbcH10, in which the active site cysteine has been changed to a serine. Extensive efforts to obtain well diffracting crystals of wild-type UbcH10 were not successful. However, the C114S mutant (designated mUbcH10) should be nearly isosteric with the wild-type, and therefore can provide insight into the stereochemical basis of Ubc function in general and UbcH10 function in particular.Activity of Mutant UbcH10We tested the ability of mutant UbcH10 to accept ubiquitin as an adduct. The linkage between mUbcH10 and ubiquitin should be an ester bond rather than the thiol ester bond that occurs with wild type. In our assay, as shown in Fig.1A, mUbcH10 indeed can form an adduct with ubiquitin, although the formation of adduct is much slower with mUbcH10 than with wild-type UbcH10. As is expected of an ester bond, the ubiquitin linkage to mUbcH10 is resistant to β-mercaptoethanol reduction but not to alkaline hydrolysis (Fig. 1, B and D). The control reaction with wild-type UbcH10 indicates that such treatment with β-mercaptoethanol is sufficient to cleave a thiol ester bond (Fig. 1B). Our observation of the ability of the cysteine to serine Ubc mutant being able to accept ubiquitin is consistent with most previous reports of similarly mutated Ubcs being able to accept ubiquitin or ubiquitin-like proteins (35Sullivan M.L. Vierstra R.D. J. Biol. Chem. 1993; 268: 8777-8780Abstract Full Text PDF PubMed Google Scholar, 36Sung P. Prakash S. Prakash L. J. Mol. Biol. 1991; 221: 745-749Crossref PubMed Scopus (52) Google Scholar, 37Wada H. Yeh E.T. Kamitani T. J. Biol. Chem. 2000; 275: 17008-17015Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar, 38Miura T. Klaus W. Gsell B. Miyamoto C. Senn H. J. Mol. Biol. 1999; 290: 213-228Crossref PubMed Scopus (85) Google Scholar, 39Banerjee A. Deshaies R.J. Chau V. J. Biol. Chem. 1995; 270: 26209-26215Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar). However, one published report indicates that a similarly mutated C114S in the mouse homolog of UbcH10 isnot able to accept ubiquitin (14Yamanaka A. Hatakeyama S. Kominami K. Kitagawa M. Matsumoto M. Nakayama K. Mol. Biol. Cell. 2000; 11: 2821-2831Crossref PubMed Scopus (46) Google Scholar). This discrepancy could be due to differences in the sensitivities of ubiquitin detection methods used (streptavidin-horseradish peroxidase/ECL detection of biotinylated ubiquitin used here versus125I-labeled ubiquitin detected by autoradiography used in the previous report). Whereas the mUbcH10 is active in accepting ubiquitin as demonstrated here, it is a dominant-negative mutant (8Townsley F.M. Aristarkhov A. Beck S. Hershko A. Ruderman J.V. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2362-2367Crossref PubMed Scopus (191) Google Scholar). The dominant-negative nature of the mutant can be explained in part by the low free energy of hydrolysis of the ester bond compared with that of the thiol ester bond, thereby making transfer of the ubiquitin from the mUbcH10 to the target protein relatively less favored. In such a case, the mUbcH10-ubiquitin adduct may remain bound unproductively to the APC/C. In addition, the slow rate of mUbcH10-ubiquitin adduct formation suggests that perhaps the mUbcH10, even without the linked ubiquitin, is able to sequester components of the ubiquitination machinery.An intriguing observation we note is that after longer incubation times, we detect what apparently are polyubiquitin chains being formed on the UbcH10 (Fig. 1A). The linkage to UbcH10 is resistant to reduction by β-mercaptoethanol treatment (Fig. 1B). This suggests that the polyubiquitin chain is linked to UbcH10 by an isopeptide bond and not with a thiol ester linkage. Autoubiquitination of other Ubcs has been reported before (40Banerjee A. Gregori L., Xu, Y. Chau V. J. Biol. Chem. 1993; 268: 5668-5675Abstract Full Text PDF PubMed Google Scholar, 41Arnold J.E. Gevers W. Biochem. J. 1990; 267: 751-757Crossref PubMed Scopus (10) Google Scholar). Although this autoubiquitination may represent nonspecific transfer of ubiquitin to a nearby primary amine (42Pickart C.M. Rose I.A. J. Biol. Chem. 1985; 260: 1573-1581Abstract Full Text PDF PubMed Google Scholar), it may be relevant to the recently reported findings that cellular levels of UbcH10 are cell cycle dependent (10Arvand A. Bastians H. Welford S.M. Thompson A.D. Ruderman J.V. Denny C.T. Oncogene. 1998; 17: 2039-2045Crossref PubMed Scopus (78) Google Scholar,14Yamanaka A. Hatakeyama S. Kominami K. Kitagawa M. Matsumoto M. Nakayama K. Mol. Biol. Cell. 2000; 11: 2821-2831Crossref PubMed Scopus (46) Google Scholar), and that UbcH10 apparently is destroyed in an APC-facilitated manner (14Yamanaka A. Hatakeyama S. Kominami K. Kitagawa M. Matsumoto M. Nakayama K. Mol. Biol. Cell. 2000; 11: 2821-2831Crossref PubMed Scopus (46) Google Scholar).Overall Fold of UbcH10We have solved and refined the crystal structure of the active site C114S mutant of the mitotic-specific ubiquitin-conjugating enzyme from human using data to 1.95-Å resolution. The data quality and model refinement statistics are presented in Table I. The UbcH10 protein is an α+β protein with one 4-stranded antiparallel β sheet and 4 α-helices (Fig. 2). The topology of the β sheet falls into group B (up-and-down meander motif) as defined by Zhang and Kim (43Zhang C. Kim S.H. J. Mol. Biol. 2000; 299: 1075-1089Crossref PubMed Scopus (48) Google Scholar). The residues forming the β sheet are located in the primary sequence between the residues forming the first and second helices. The NH2-terminal helix lays diagonally across one broad face of the sheet, whereas the other three helices flank two opposite edges of the sheet. Almost two turns of a 310-helix (residues 115–119) are located between the fourth strand of the sheet and the second α-helix. The active site residue (114) is situated in a set of β turns and adjacent to the NH2 terminus of the 310-helix. The overall shape of the UbcH10 protein is roughly that of an elongated triangular prism. The side chain of the active site residue is situated on one long edge of the prism.FIG. 2The overall fold of mUbcH10 (divergent stereoview). The α-helices, 310-helix, and β-strands are shown in red, green, and blue, respectively. The active site residue (114) is depicted in ball-and-stick rendering. This and all other structure figures were prepared using the Swiss-PDB Viewer (53Guex N. Peitsch M.C. Electrophoresis. 1997; 18: 2714-2723Crossref PubMed Scopus (9472) Google Scholar) and in certain cases post-processed with POV-Ray (www.povray.org) or MegaPOV (nathan.kopp.com/patched.htm).View Large Image Figure ViewerDownload (PPT)Comparison with Other Ubc StructuresThe secondary structural elements are highly conserved in all known Ubc structures (Fig.3, A and B). In particular, β strands 3 and 4 are strictly conserved in length, whereas deviations in the lengths of β strands 1 and 2 lead to a local breakdown of structural equivalence (Fig. 3A). All four α-helices display small variations in terms of length, whereas the 310 helix is extremely well conserved. When the various Ubc structures are superimposed, it is seen quite distinctly that the continuous polypeptide segment consisting of β strands 2, 3, 4, the 310 helix, helix 2, and the intervening turns are relatively highly conserved in backbone position. The remaining regions display more variability in disposition. As can be seen from Fig.3B, these more variable regions flank the more highly conserved region. Moreover, the active site residue, although situated in the highly conserved segment, is near a set of turns connecting helices 2 and 3 that are relatively poorly conserved. The variable regions on the face opposite to that containing the active site could be involved in interactions with E3 activities that function by providing a scaffolding for interaction of the Ubc-ubiquitin adduct. The variable regions near the active site could represent sites of interaction with E3 activities that involve ubiquitin transfer to the E3 prior to the target protein or could be important for target protein recognition. The rather striking structural conservation of β strands 2–4 and helix 2 is not reflected in the sequence (Fig. 3C). Overall, this region is not better conserved in the sequence than the other parts of the protein. The strong evolutionary pressure to maintain structure and the relatively low pressure to maintain sequence suggest that this region is more important for intrinsic structural reasons than for specific protein-protein interactions. This comparison makes clear that the Ubc structure is exquisitely structurally tuned to perform its function.FIG. 3Comparison of Ubc structures.A, structure-based sequence alignment. The structures of the Ubc proteins were superimposed and structurally equivalent residues were identified using the program STAMP (54Russell R.B. STAMP (Version 4.2) manual. 1999; barton.ebi.ac.ukGoogle Scholar, 55Russell R.B. Barton G.J. Proteins. 1992; 14: 309-323Crossref PubMed Scopus (572) Google Scholar). Thebar graphs above the sequences indicate the root mean square deviation (rmsd) of the structurally equivalent α-carbon atoms from their average position. Helical regions and β strand regions are indicated in gray and blackbackgrounds, respectively, and are labeled. B, divergent stereoview of superimposed Ubc structures. Only the α-carbon atoms are shown. The loop extending from a single structure near the 310 helix is in the Ubc7 structure. The secondary structural elements are labeled. The most highly structurally conserved region corresponds to strands S2, S3, S4, and helix 2.C, divergent stereoview of the UbcH10 structure depicted as a ribbon. The ribbon is color coded according to sequence conservation with blue being the most conserved and red the least conserved. The most highly conserved region in the structure (panel B) is not uniformly the most highly conserved region in sequence. The sequence alignment was performed using the ConSurf server (Ref. 56Armon A. Graur D. Ben-Tal N. J. Mol. Biol. 2001; 307: 447-463Crossref PubMed Scopus (379) Google Scholar, bioinfo.tau.ac.il/ConSurf/).View Large Image Figure ViewerDownload (PPT)NH2-terminal ExtensionThe full-length UbcH10 contains 179 residues. This Ubc belongs to the class III Ubc proteins, characterized by an NH2-terminal extension followed by the "core" Ubc fold. However, electron density is absent for the residues in the NH2-terminal extension (1–29) and residues 176–179. To determine whether the absence of density is due to disorder or proteolytic degradation, mass spectrometric analysis was performed on a sample prepared from a dissolved crystal. The mass spectrometric analysis yielded results that are consistent with cleavage at residue Arg12 (data not shown). Therefore, the first 12 residues have been removed by proteolytic cleavage, whereas residues 13–29 and 176–179 are present but disordered.To determine whether or not the N-terminal extension is necessary for UbcH10 to accept ubiquitin, we prepared a mutant lacking most of the NH2-terminal extension (up to residue 27). The activity assay of this deletion mutant reveals that these residues are not important for ubiquitin-adduct formation (Fig.4). Moreover, substitution of the NH2-terminal sequence with that derived from the pET28 cloning construct also does not impede ubiquitin-adduct formation (Fig.4). The sequence of the NH2-terminal extension is fairly well conserved among mitotic E2, indicating some function, for example, interaction with the APC or other components of the network regulating mitotic destruction. Experiments to pursue these ideas are underway.FIG. 4Effect of selected mutations of UbcH10 on ubiquitin-adduct formation.A, sequences of varying NH2-terminal constructs tested for ubiquitin-adduct formation. To determine the effect of the NH2-terminal extension of UbcH10, a mutant form lacking much of the NH2-terminal extension without and with the removal of the His tag region (His-NΔ and NΔ, respectively) was compared with a His-tagged construct of wild-type UbcH10 (His-WT). The letters in bold represent residues that correspond to the UbcH10 sequence and the letters in normal type correspond to residues arising from cloning into the pET28 vector. The letters underlinedindicate the thrombin cleavage site. The residues "MAS" in the deletion mutants are present due to cloning into theNheI site of the contiguous NdeI-NheI sites in the vector. The correspondence of MAS to the first three residues of UbcH10 is coincidental. B, activity assay of the NH2-terminal deletion mutants and lysine 119 mutant. The activity assay was performed as described under "Experimental Procedures" with biotinylated ubiquitin (bt-Ub) for an incubation time of 10 min. Biotinylated lysozyme (bt-Lys) was included in the reaction as a gel loading control.View Large Image Figure ViewerDownload (PPT)Oligomeric State of UbcH10An intriguing question still not answered satisfactorily concerns the oligomeric state of E2 proteins. This is an important question since the oligomeric state can have fundamental implications in the stereochemistry of the ubiquitination reaction. For example, the artificially induced dimerization of a E2–25K mutant produced by expressing as a glutathioneS-transferase fusion dramatically alters its activity (44Haldeman M.T. Xia G. Kasperek E.M. Pickart C.M. Biochemistry. 1997; 36: 10526-10537Crossref PubMed Scopus (120) Google Scholar). Sometimes the same E2 protein, such as yeast Ubc4, has been reported to be a monomer as assessed by gel filtration chromatography (45Gwozd C.S. Arnason T.G. Cook W.J. Chau V. Ellison M.J. Biochemistry. 1995; 34: 6296-6302Crossref PubMed Scopus (24) Google Scholar) and crystal packing analysis (16Cook W.J. Jeffrey L.C., Xu, Y. Chau V. Biochemistry. 1993; 32: 13809-13817Crossref PubMed Scopus (83) Google Scholar) or as a dimer also by gel filtration chromatography (44Haldeman M.T. Xia G. Kasperek E.M. Pickart C.M. Biochemistry. 1997; 36: 10526-10537Crossref PubMed Scopus (120) Google Scholar) and chemical cross-linking (45Gwozd C.S. Arnason T.G. Cook W.J. Chau V. Ellison M.J. Biochemistry. 1995; 34: 6296-6302Crossref PubMed Scopus (24) Google Scholar). In the case of UbcH10, analysis of the crystal packing reveals a rather large interaction surface that results in a total surface area of 1270 Å2 becoming solvent excluded. The interaction surface is formed by residues 36, 39, 40, 43, 44, 51, 53–57, 63, 65, 78, and 80. The presence of this large contact surface might suggest that the UbcH10 acts as a dimer. Moreover, a very similar interaction surface is seen in the crystal packing of the clam mitotic E2 protein, E2-C (21Jiang F. Basavappa R. Biochemistry. 1999; 38: 6471-6478Crossref PubMed Scopus (22) Google Scholar). To determine the quarternary structure of UbcH10 in solution, analytical ultracentrifugation studies were performed at protein concentrations of 0.94, 0.5, and 0.11 mg/ml. The results indicate a molecular weight of 19,260 (data not shown), which corresponds very well to the theoretical molecular weight of 19,652. Since the intracellular concentration of UbcH10 certainly is less than 1 mg/ml, it is likely that UbcH10 functions as a monomer and that the significant interactions observed in the crystal lattices are not functionally important.Active Site EnvironmentThe active site residue (114) is next to the NH2 terminus of a 310 helix. Unlike most 310 helices, which are situated at the terminus of an α-helix (46Barlow D.J. Thornton J.M. J. Mol. Biol. 1988; 201: 601-619Crossref PubMed Scopus (952) Google Scholar), this 310 helix is entirely separate from any α-helix. This 310 helix is formed by residues 115–119 and has the sequence LDILK. The 310 helix and the identity of residues forming this helix are highly homologous among E2 proteins (21Jiang F. Basavappa R. Biochemistry. 1999; 38: 6471-6478Crossref PubMed Scopus (22) Google Scholar). Why 310 helix formation in proteins is favored in certain instances instead of α-helices is not well understood. In isolation, the α-helix is less strained energetically than the 310 helix due to more favorable main chain hydrogen bonding geometry in the α-helix as well as slight steric hindrance of the side chains in the 310 helix. The presence of 310 helices in proteins might be explained in part by interactions of the side chains in the helix with the rest of the protein that would disfavor the α-helix and favor the 310helix. Such steric constraints are indicated in the case of the 310 helix in the E2 proteins. The 3 residue per turn geometry of the 310 helix places the charged residues Asp116 and Lys119 in-phase with each other, whereas the hydrophobic residues are oriented in different directions. In the context of the neighboring parts of the E2 structure, this allows the charged residues to be relatively solvent exposed and allows the hydrophobic residues to pack primarily against hydrophobic atoms (Fig. 5). If these residues instead were in an α-helical conformation, then these side chains would be in a much less favorable environment. The formation of the 310helix may have functional significance since it places the last residue of the helix (Lys119) in proximity to the active site residue (Fig. 5). The positive charge of the ε-amino group of this residue could be important for reducing the pKa of the active site cysteine to make it more reactive. To test whether this lysine residue indeed is important for ubiquitin-adduct formation, a mutant (K119A-UbcH10) was prepared in which this residue has been changed to an alanine. Ubiquitin-adduct formation assay with this mutant indicates greatly diminished activity when compared with wild-type. Therefore, lysine 119 seems to play an important role in the mechanism of ubiquitin-adduct formation in UbcH10. It is probable that this role is more electrostatic (as proposed above) than purely structural in nature since the lysine is highly solvent exposed and therefore mutation to an alanine would be unlikely to cause significant structural perturbation of the active site region.FIG. 5The 310 helix situated near the active site (divergent stereoview). The molecular surface was calculated using all residues except those in the 310 helix (residues 115–119). The residues forming the 310 helix are shown in ball-and-stick representation. The geometry of the 310 helix allows the hydrophobic residues in the helix to pack primarily against non-polar interior residues, whereas the hydrophilic residues are substantially solvent exposed. The polar residue Lys119 is seen to be proximal to the active site thiol (indicated by the dark patch). The smallinset α-carbon trace shows the overall perspective from which the main figure was made. The 310 helix and active site residue are indicated by thicker segments.View Large Image Figure ViewerDownload (PPT)In addition to the 310 helix, the active site is situated in the neighborhood of four β turns (Fig.6). These turns are important since they (together with the 310 helix) provide almost all the contacts with the active site residue. The only other residues contacting the active site residue are Leu138 and Ile113 (which is just NH2-terminal to the active site residue). Moreover, the residues in these turns give rise to much of the surface features surrounding the active site. In UbcH10, the turns are formed by residues 104–107 (turn A), 108–111 (turn B), 143–146 (turn C), and 147–150 (turn D). These turns are present in other known Ubc structures (Fig. 6). In all cases, turn A is nearly a canonical type I turn, turns with (φ, ψ) of residue i + 1 = ∼(−60,−30) and (φ, ψ) of r