Molecular Cloning of the cDNA and Chromosome Localization of the Gene for Human Ubiquitin-conjugating Enzyme 9

Abstract

We report a novel human gene whose product specifically associates with the negative regulatory domain of the Wilms' tumor gene product (WT1) in a yeast two-hybrid screen and with WT1 in immunoprecipitation and glutathione S-transferase (GST) capture assays. The gene encodes a 17-kDa protein that has 56% amino acid sequence identity with yeast ubiquitin-conjugating enzyme (yUBC) 9, a protein required for cell cycle progression in yeast, and significant identity with other subfamilies of ubiquitin-conjugating enzymes. The human gene fully complements yeast that have a temperature-sensitive yUBC9 gene mutation to fully restore normal growth, indicating that we have cloned a functionally conserved human (h) homolog of yUBC9. Transcripts of hUBC9 of 4.4 kilobases (kb), 2.8 kb, and 1.3 kb were found in all human tissues tested. A single copy of the hUBC9 gene was found and localized to human chromosome 16p13.3. We conclude that hUBC9 retains striking structural and functional conservation with yUBC9 and suggest a possible link of the ubiquitin/proteosome proteolytic pathway and the WT1 transcriptional repressor system. We report a novel human gene whose product specifically associates with the negative regulatory domain of the Wilms' tumor gene product (WT1) in a yeast two-hybrid screen and with WT1 in immunoprecipitation and glutathione S-transferase (GST) capture assays. The gene encodes a 17-kDa protein that has 56% amino acid sequence identity with yeast ubiquitin-conjugating enzyme (yUBC) 9, a protein required for cell cycle progression in yeast, and significant identity with other subfamilies of ubiquitin-conjugating enzymes. The human gene fully complements yeast that have a temperature-sensitive yUBC9 gene mutation to fully restore normal growth, indicating that we have cloned a functionally conserved human (h) homolog of yUBC9. Transcripts of hUBC9 of 4.4 kilobases (kb), 2.8 kb, and 1.3 kb were found in all human tissues tested. A single copy of the hUBC9 gene was found and localized to human chromosome 16p13.3. We conclude that hUBC9 retains striking structural and functional conservation with yUBC9 and suggest a possible link of the ubiquitin/proteosome proteolytic pathway and the WT1 transcriptional repressor system. INTRODUCTIONThe ubiquitin-dependent protein degradation system has been recognized as a complete enzymatic pathway that is responsible for the selective degradation of abnormal and short-lived proteins (1Finley D. Chau V. Annu. Rev. Cell Biol. 1991; 7: 25-69Google Scholar, 2Hershko A. Ciechanover A. Annu. Rev. Biochem. 1992; 61: 761-807Google Scholar). In this process, ubiquitin is covalently linked to target proteins prior to their degradation through the combined action of three classes of proteins, the ubiquitin-activating enzyme (E1), 1The abbreviations used are: E1ubiquitin-activating enzymeE2ubiquitin-conjugating enzymeE3ubiquitin-protein ligasePAGEpolyacrylamide gel electrophoresisHAhemagglutininkbkilobase(s)Mbmegabase(s)FISHfluorescence in situ hybridization. ubiquitin-conjugating enzymes (E2), and in some cases, the ubiquitin-protein ligases (E3) that are believed to be important in substrate recognition. E1 catalyzes the ATP-dependent formation of a thioester bond between the C-terminal glycine of ubiquitin and the active site cysteine of E1. “Activated” ubiquitin is then transferred to the active site cysteine of an E2 enzyme which itself, or in conjunction with an E3 protein, catalyzes the formation of an isopeptide bond between the C-terminal glycine of ubiquitin and a lysine ϵ-amino group in the protein targeted for degradation by the 20 S proteosome complex. The diversity and multiple protein-protein interactions of the ubiquitin-dependent proteolytic system suggests that a high degree of regulation is required to achieve the specificity needed to control the fate of the different proteins that are degraded through ubiquitination. It is likely that ubiquitin-conjugating (E2) enzymes may contribute to the determination of which groups of protein are targeted for selective degradation.The ubiquitin-conjugating enzymes (E2s) are a family of proteins characterized by a highly conserved catalytic site containing an invariant active site cysteine (3Jentsch S. Annu. Rev. Genet. 1992; 26: 179-207Google Scholar). In yeast Saccharomyces cerevisiae, at least 10 different E2s have been identified that are involved in essential cellular processes such as DNA repair, cell cycle control, and stress responses (4Ciechanover A. Cell. 1994; 79: 13-21Google Scholar), suggesting that E2 enzymes may be key players in establishing the diversity of the ubiquitin-proteolytic system. Although genetic analysis in yeast has revealed that ubiquitin conjugation is essential to cell viability (5Seufert W. Jentsch S. EMBO J. 1990; 9: 543-550Google Scholar, 6Seufert W. Futcher B. Jentsch S. Nature. 1995; 373: 78-81Google Scholar), only a few cellular targets of the ubiquitin protein ligase system have been identified. These include cyclin B and the yeast transcription factor MATa2 (7Glotzer M. Murray A.W. Kirschner M.W. Nature. 1991; 349: 132-138Google Scholar, 8Hochstrasser M. Ellison M.J. Chau V. Varshavsky A. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4606-4610Google Scholar). Recently, the ubiquitin-dependent proteolytic system has also been shown to actively participate in the degradation of p53 (9Scheffner M. Huibregtse J.M. Vierstra R.D. Howley P.M. Cell. 1993; 75: 495-505Google Scholar), in the degradation and processing of cystic fibrosis transmembrane conductance regulator (10Ward C.L. Omura S. Kopito R.R. Cell. 1995; 83: 121-127Google Scholar, 11Jensen T.J. Loo M.A. Pind S. Williams D.B. Goldberg A.L. Riordan J.R. Cell. 1995; 83: 129-135Google Scholar), and in the processing of NF-κB (12Palombella V.J. Rando O.J. Goldberg A.L. Maniatis T. Cell. 1994; 78: 773-785Google Scholar), suggesting more widespread roles of this system.The Wilms' tumor suppressor gene encodes a zinc finger DNA-binding protein (13Call K.M. Glaser T. Ito C.Y. Buckley A.J. Pelletier J. Haber D.A. Rose E.A. Kral A. Yeger H. Lewis W.H. Jones C. Housman P.E. Cell. 1990; 60: 509-520Google Scholar, 14Gessler M. Poustka A. Gavenee W. Neve R.L. Orkin S.H. Bruns G.A.P. Nature. 1990; 343: 774-778Google Scholar) that functions in transient transfection assays as a transcriptional repressor of several growth-related genes such as insulin-like growth factor II (15Drummond I.R. Madden S.L. Rohwer-Nutter P. Bell G.I. Sukhatme V.P. Rauscher III, F.J. Science. 1992; 257: 674-678Google Scholar), insulin-like growth factor IR (16Werner H. Re G.G. Drummond I.A. Sukhatme V.P. Rauscher III, F.J. Sens D.A. Garvin A.J. Leroitth D. Roberts Jr., C.T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5828-5832Google Scholar), platelet-derived growth factor A-chain (17Wang Z.Y. Madden S.L. Deuel T.F. Rauscher III, F.J. J. Biol. Chem. 1992; 268: 9172-9175Google Scholar), colony stimulating factor 1 (18Harrington M.A. Konieck B. Song A. Xia X.L. Fredericks W.J. Rauscher III, F.J. J. Biol. Chem. 1993; 268: 21271-21275Google Scholar), and transforming growth factor β1 (19Dey B.R. Sukhatme V.P. Roberts A.B. Sporn M.B. Rauscher III, F.J. Kim S. Mol. Endocrinol. 1994; 8: 595-602Google Scholar). Previously, we identified two separate domains (residues 85-179 and 250-266) of WT1 that function independently as transcriptional repressors (20Wang Z.Y. Qiu Q.Q. Huang J. Gurrieri M. Deuel T.F. Oncogene. 1995; 10: 415-422Google Scholar, 21Wang Z.Y. Qiu Q.Q. Deuel T.F. J. Biol. Chem. 1993; 268: 9172-9175Google Scholar). When expressed independently of its zinc finger domain, the proximal repressor domain (amino acids 85-179) relieved the repressor activity of wild type WT1 in a dose-dependent manner (22Wang Z.-Y. Qiu Q.-Q. Gurrieri M. Huang J. Deuel T.F. Oncogene. 1995; 10: 1243-1247Google Scholar), suggesting the presence of an interactive protein with WT1 in vivo. In order to identify and clone the gene encoding this interactive protein, we have used the yeast two-hybrid system with the proximal repressor domain of WT1 as “bait.” We report here the isolation and characterization of cDNA clones that encode the human homolog of the yeast ubiquitin-conjugating enzyme (yUBC9) that specifically interacts with the proximal repressor domain of WT1 in the yeast two-hybrid system and with WT1 in co-immunoprecipitation and glutathione S-transferase assays. The results suggest that UBC9 may be involved in the transcription regulation mediated by WT1.RESULTSTo identify the WT1 interactive protein previously described (22Wang Z.-Y. Qiu Q.-Q. Gurrieri M. Huang J. Deuel T.F. Oncogene. 1995; 10: 1243-1247Google Scholar), a human placental cDNA/Gal4 transcriptional activation domain fusion library was screened in the yeast two-hybrid system with the proximal repressor domain of WT1 as bait. Eleven of 2 × 106 clones screened were strongly positive in the yeast two-hybrid screen, and seven of these clones encoded portions of the same protein. Each of the encoded proteins of the 7 positive clones interacted strongly with the bait but failed to interact with the LexA DNA-binding domain alone, with lamin C, or with the second repressor domain of WT1 (residues 250-260) (Table I). These data suggest that the interaction of the protein products of the 7 related positive clones with the proximal repressor domain of WT1 is specific.To obtain full-length cDNAs, a placental cDNA library was probed with one of the positive clones. Eight independent clones with inserts of 0.6, 1.1, and 1.8 kb were isolated. The 1.8-kb and 1.1-kb cDNAs were fully sequenced. The two cDNA clones share the initial 15 base pairs upstream of the translation start site and the entire coding as well as 3′-untranslated regions. However, the two cDNA clones differ in the more 5′-untranslated regions, presumably the result of alternative splicing. The longest open reading frame of each clone predicts a protein product of 157 amino acids (Fig. 1A). To establish that the cDNAs encoded protein, each of the cDNA clones was tested by in vitro transcription/translation. Each cDNA encodes a protein that migrated identically in SDS-polyacrylamide gels with a mobility consistent with the predicted 157-amino acid protein products (data not shown).The predicted amino acid sequence of the two cDNA clones was compared with protein sequences deposited in GenBank™. A 56% identity was established between the predicted amino acid sequence of our clones and yeast ubiquitin-conjugating enzyme 9 (yUBC9), a nuclear protein that is required for cell growth and that is involved in the degradation of S- and M-phase cyclins (6Seufert W. Futcher B. Jentsch S. Nature. 1995; 373: 78-81Google Scholar). The predicted protein product contains the active site cysteine residue (residue 93 boxed, Fig. 1B) necessary for thioester formation and the ubiquitin conjugating activity of all E2 enzymes. To establish that our clone encodes a functional UBC9 homolog, the 1.1-kb human cDNA was expressed in a yeast carrying a temperature-sensitive (ts) yUBC9 mutation (6Seufert W. Futcher B. Jentsch S. Nature. 1995; 373: 78-81Google Scholar). The human UBC9 homolog fully restored normal growth to the yeast strain carrying the yUBC9 ts mutant at the otherwise nonpermissive temperature (Fig. 2). This result indicates that the cDNA we isolated encodes a biologically active human homolog of yUBC9 that is structurally and functionally highly conserved with yUBC9. We have designated our gene as human (h) UBC9.Fig. 2Human UBC9 complements the loss of yeast UBC9 gene function. Yeast strain W9432 carrying the temperature-sensitive ubc9-1 allele was transformed with control vectors (row 1, p416GAL1; row 4, pSE936) or constructs expressing either a human UBC9 cDNA (row 2) or the yeast UBC9 gene (row 3). To compare growth of these strains, cells were spotted in a dilution series on galactose-containing plates and incubated for 3.5 days at the permissive temperature (23°C) or 2 days at the restrictive temperature (34°C).View Large Image Figure ViewerDownload (PPT)We compared the predicted amino acid sequence of hUBC9 with other E2 enzymes. Identification in regions of amino acid sequence were established within a number of other ubiquitin-conjugating enzymes, some of which are shown in Fig. 3. The UBCs from human, pea, yeast S. cerevisiae, and Arabidopsis thaliana retain a striking degree of amino acid identity in several domains between the UBC9 and UBC4 families (Fig. 3), suggesting perhaps some degree of functional overlap among UBC9 and UBC4.Fig. 3A multiple sequence alignment of the Arabidopsis UBC9 sequence, the yeast UBC9 and UBC4 sequences, and the human UBC9 and UBC4 sequences. The sequences were aligned using the PILEUP algorithm. PILEUP is part of the Wisconsin Sequence Analysis Package. An amino acid is placed in the consensus sequence if it is located at that position in four of the five aligned sequences.View Large Image Figure ViewerDownload (PPT)To further establish the specificity of the interaction of hUBC9 with the N-terminal repressor domain of WT1 that was revealed in the yeast two-hybrid screening, glutathione S-transferase (GST) capture assays were used. Human UBC9 was expressed as a GST-fusion protein, coupled to glutathione-Sepharose beads, and incubated with extracts from 293 cells that express the full-length (or fragments) of WT1. Eluates were analyzed by Western blots with a rabbit anti-WT1 polyclonal antisera. Repeated experiments demonstrated that GST-hUBC9 “captures” WT1 from lysates of 293 cells (Fig. 4A). In control experiments, the GST-hUBC9 fusion protein failed to capture the C-terminal DNA-binding domain of WT1 that lacks the N-terminal amino acid repressor domain used as bait in the yeast two-hybrid screening. GST alone failed to capture WT1 (Fig. 4A), and GST-hUBC9 failed to capture proteins from control 293 cells that did not express WT1 (Fig. 4A). These data support the specificity of the interaction between the N-terminal repressor domain of WT1 and hUBC9.Fig. 4The interaction of WT1 and hUBC9. A, in vitro interactions of WT1 and hUBC9. The full-length hUBC9 cDNA was fused with the GST and expressed as the GST-hUBC9 fusion protein. GST and GST-hUBC9 fusion proteins were to glutathione-Sepharose from 293 cells that expressed WT1 and of and were incubated with GST and GST-fusion protein to for with were with gel and by and transferred to for with and C-terminal portions of WT1 and by co-immunoprecipitation of WT1 and hUBC9 from 293 cells expressing WT1 and of 293 cells were with either anti-WT1 or a DNA on and probed with Large Image Figure ViewerDownload also was used to establish the specificity of the interactions of hUBC9 and WT1. hUBC9 was fused at its amino with an the hUBC9 was with WT1 in 293 and extracts of cells were with anti-WT1 polyclonal and The were on SDS-polyacrylamide and analyzed in Western blots with an The hUBC9 was identified in lysates from cells that WT1 and hUBC9 (Fig. The hUBC9 was not in the same lysates a polyclonal was for the anti-WT1 in the immunoprecipitation and of lysates of cells that expressed hUBC9 alone did not proteins recognized by (Fig. the results of the yeast two-hybrid screening, GST fusion protein capture and co-immunoprecipitation strongly support the that the N-terminal repressor domain of WT1 and hUBC9 interact specifically in and in blots from different human tissues were used to identify the tissues that hUBC9. Transcripts of and 1.3 kb were identified in each of the tissues that were (Fig. and have of hUBC9 to the other tissues the is expressed in suggesting that the hUBC9 is expressed but at different in different of hUBC9 in different human A human was probed with the 1.1-kb hUBC9 Each 2 of from and The cDNA was used to the same as The are on the of the analysis of the hUBC9 10 of human DNA were with or and DNA were by on The was probed with the 1.1-kb hUBC9 cDNA Large Image Figure ViewerDownload (PPT)To the of we probed blots of human DNA with the 1.1-kb cDNA and single in and (Fig. suggesting that hUBC9 is a The human UBC9 was then in human by The hUBC9 on the of human chromosome in of on all were located on localized at (Fig. A cell with between of and a of 1 was used to further the of 20 kb and kb and of kb, kb, and kb were in the The human of 20 kb was on chromosome and was in the and but from the and with more proximal of to indicating that hUBC9 between (data not of hUBC9 by on human of to 16p13.3. The a and Large Image Figure ViewerDownload data strongly suggest that the interactive protein previously identified (22Wang Z.-Y. Qiu Q.-Q. Gurrieri M. Huang J. Deuel T.F. Oncogene. 1995; 10: 1243-1247Google Scholar) that WT1 repressor activity is hUBC9. This is the specificity of the interaction of the N-terminal but not the C-terminal repressor domain of WT1 with the of lamin to interact with WT1 in the yeast two-hybrid and the specificity of the interaction of hUBC9 and WT1 demonstrated in co-immunoprecipitation and GST capture assays. The data also significant identity of and of the invariant active site cysteine residue between the predicted amino acid sequence of our cDNA clones and yUBC9. cDNA clone functionally complements yeast carrying a temperature-sensitive UBC9 gene indicating that hUBC9 is fully functional as an E2 enzyme in yeast and that we have cloned the human homolog of the yUBC9. Human UBC9 retains a high degree of amino acid identity with other UBCs that have been cloned and is related to the UBC4 The of the highly conserved domains that we have identified among UBCs is but these domains may be important in selective interactions with target of our cDNA to fully yeast that a temperature-sensitive mutant is a striking of functional conservation from yeast to human and suggests a highly important of the In yeast, UBC9 is involved in cell cycle Yeast UBC9 is required for degradation of S- and cyclins and for viability (6Seufert W. Futcher B. Jentsch S. Nature. 1995; 373: 78-81Google Scholar). In this it is important to that WT1 cell cycle progression and that this is relieved by of cyclin and as well as cyclin T. T. M. T. Proc. Natl. Acad. Sci. U. S. A. 1995; Scholar). of the striking conservation of and hUBC9 fully complements yeast with yUBC9 temperature-sensitive to normal progression of yeast through the cell we suggest that hUBC9 may have a in The results also suggest that the interaction of WT1 with hUBC9 may be important in the cell cycle by hUBC9 is an conserved protein in from yeast to its in human cells is the of UBC9 in yeast for normal cell cycle progression and the interactions with WT1 suggest important roles of hUBC9 in cells and the of important roles of in regulation of INTRODUCTIONThe ubiquitin-dependent protein degradation system has been recognized as a complete enzymatic pathway that is responsible for the selective degradation of abnormal and short-lived proteins (1Finley D. Chau V. Annu. Rev. Cell Biol. 1991; 7: 25-69Google Scholar, 2Hershko A. Ciechanover A. Annu. Rev. Biochem. 1992; 61: 761-807Google Scholar). In this process, ubiquitin is covalently linked to target proteins prior to their degradation through the combined action of three classes of proteins, the ubiquitin-activating enzyme (E1), 1The abbreviations used are: E1ubiquitin-activating enzymeE2ubiquitin-conjugating enzymeE3ubiquitin-protein ligasePAGEpolyacrylamide gel electrophoresisHAhemagglutininkbkilobase(s)Mbmegabase(s)FISHfluorescence in situ hybridization. ubiquitin-conjugating enzymes (E2), and in some cases, the ubiquitin-protein ligases (E3) that are believed to be important in substrate recognition. E1 catalyzes the ATP-dependent formation of a thioester bond between the C-terminal glycine of ubiquitin and the active site cysteine of E1. “Activated” ubiquitin is then transferred to the active site cysteine of an E2 enzyme which itself, or in conjunction with an E3 protein, catalyzes the formation of an isopeptide bond between the C-terminal glycine of ubiquitin and a lysine ϵ-amino group in the protein targeted for degradation by the 20 S proteosome complex. The diversity and multiple protein-protein interactions of the ubiquitin-dependent proteolytic system suggests that a high degree of regulation is required to achieve the specificity needed to control the fate of the different proteins that are degraded through ubiquitination. It is likely that ubiquitin-conjugating (E2) enzymes may contribute to the determination of which groups of protein are targeted for selective degradation.The ubiquitin-conjugating enzymes (E2s) are a family of proteins characterized by a highly conserved catalytic site containing an invariant active site cysteine (3Jentsch S. Annu. Rev. Genet. 1992; 26: 179-207Google Scholar). In yeast Saccharomyces cerevisiae, at least 10 different E2s have been identified that are involved in essential cellular processes such as DNA repair, cell cycle control, and stress responses (4Ciechanover A. Cell. 1994; 79: 13-21Google Scholar), suggesting that E2 enzymes may be key players in establishing the diversity of the ubiquitin-proteolytic system. Although genetic analysis in yeast has revealed that ubiquitin conjugation is essential to cell viability (5Seufert W. Jentsch S. EMBO J. 1990; 9: 543-550Google Scholar, 6Seufert W. Futcher B. Jentsch S. Nature. 1995; 373: 78-81Google Scholar), only a few cellular targets of the ubiquitin protein ligase system have been identified. These include cyclin B and the yeast transcription factor MATa2 (7Glotzer M. Murray A.W. Kirschner M.W. Nature. 1991; 349: 132-138Google Scholar, 8Hochstrasser M. Ellison M.J. Chau V. Varshavsky A. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4606-4610Google Scholar). Recently, the ubiquitin-dependent proteolytic system has also been shown to actively participate in the degradation of p53 (9Scheffner M. Huibregtse J.M. Vierstra R.D. Howley P.M. Cell. 1993; 75: 495-505Google Scholar), in the degradation and processing of cystic fibrosis transmembrane conductance regulator (10Ward C.L. Omura S. Kopito R.R. Cell. 1995; 83: 121-127Google Scholar, 11Jensen T.J. Loo M.A. Pind S. Williams D.B. Goldberg A.L. Riordan J.R. Cell. 1995; 83: 129-135Google Scholar), and in the processing of NF-κB (12Palombella V.J. Rando O.J. Goldberg A.L. Maniatis T. Cell. 1994; 78: 773-785Google Scholar), suggesting more widespread roles of this system.The Wilms' tumor suppressor gene encodes a zinc finger DNA-binding protein (13Call K.M. Glaser T. Ito C.Y. Buckley A.J. Pelletier J. Haber D.A. Rose E.A. Kral A. Yeger H. Lewis W.H. Jones C. Housman P.E. Cell. 1990; 60: 509-520Google Scholar, 14Gessler M. Poustka A. Gavenee W. Neve R.L. Orkin S.H. Bruns G.A.P. Nature. 1990; 343: 774-778Google Scholar) that functions in transient transfection assays as a transcriptional repressor of several growth-related genes such as insulin-like growth factor II (15Drummond I.R. Madden S.L. Rohwer-Nutter P. Bell G.I. Sukhatme V.P. Rauscher III, F.J. Science. 1992; 257: 674-678Google Scholar), insulin-like growth factor IR (16Werner H. Re G.G. Drummond I.A. Sukhatme V.P. Rauscher III, F.J. Sens D.A. Garvin A.J. Leroitth D. Roberts Jr., C.T. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 5828-5832Google Scholar), platelet-derived growth factor A-chain (17Wang Z.Y. Madden S.L. Deuel T.F. Rauscher III, F.J. J. Biol. Chem. 1992; 268: 9172-9175Google Scholar), colony stimulating factor 1 (18Harrington M.A. Konieck B. Song A. Xia X.L. Fredericks W.J. Rauscher III, F.J. J. Biol. Chem. 1993; 268: 21271-21275Google Scholar), and transforming growth factor β1 (19Dey B.R. Sukhatme V.P. Roberts A.B. Sporn M.B. Rauscher III, F.J. Kim S. Mol. Endocrinol. 1994; 8: 595-602Google Scholar). Previously, we identified two separate domains (residues 85-179 and 250-266) of WT1 that function independently as transcriptional repressors (20Wang Z.Y. Qiu Q.Q. Huang J. Gurrieri M. Deuel T.F. Oncogene. 1995; 10: 415-422Google Scholar, 21Wang Z.Y. Qiu Q.Q. Deuel T.F. J. Biol. Chem. 1993; 268: 9172-9175Google Scholar). When expressed independently of its zinc finger domain, the proximal repressor domain (amino acids 85-179) relieved the repressor activity of wild type WT1 in a dose-dependent manner (22Wang Z.-Y. Qiu Q.-Q. Gurrieri M. Huang J. Deuel T.F. Oncogene. 1995; 10: 1243-1247Google Scholar), suggesting the presence of an interactive protein with WT1 in vivo. In order to identify and clone the gene encoding this interactive protein, we have used the yeast two-hybrid system with the proximal repressor domain of WT1 as “bait.” We report here the isolation and characterization of cDNA clones that encode the human homolog of the yeast ubiquitin-conjugating enzyme (yUBC9) that specifically interacts with the proximal repressor domain of WT1 in the yeast two-hybrid system and with WT1 in co-immunoprecipitation and glutathione S-transferase assays. The results suggest that UBC9 may be involved in the transcription regulation mediated by WT1.