Recent works have shown the importance of reduction/oxidation (redox) regulation in various biological phenomena. Thioredoxin (TRX) is one of the major components of the thiol reducing system and plays multiple roles in cellular processes such as proliferation, apoptosis, and gene expression. To investigate the molecular mechanism of TRX action, we used a yeast two-hybrid system to identify TRX-binding proteins. One of the candidates, designated as thioredoxin-binding protein-2 (TBP-2), was identical to vitamin D3 up-regulated protein 1 (VDUP1). The association of TRX with TBP-2/VDUP1 was observed in vitro and in vivo. TBP-2/VDUP1 bound to reduced TRX but not to oxidized TRX nor to mutant TRX, in which two redox active cysteine residues are substituted by serine. Thus, the catalytic center of TRX seems to be important for the interaction. Insulin reducing activity of TRX was inhibited by the addition of recombinant TBP-2/VDUP1 protein in vitro. In COS-7 and HEK293 cells transiently transfected with TBP-2/VDUP1 expression vector, decrease of insulin reducing activity of TRX and diminishment of TRX expression was observed. These results suggested that TBP-2/VDUP1 serves as a negative regulator of the biological function and expression of TRX. Treatment of HL-60 cells with 1α,25-dihydroxyvitamin D3 caused an increase of TBP-2/VDUP1 expression and down-regulation of the expression and the reducing activity of TRX. Therefore, the TRX-TBP-2/VDUP1 interaction may be an important redox regulatory mechanism in cellular processes, including differentiation of myeloid and macrophage lineages. Recent works have shown the importance of reduction/oxidation (redox) regulation in various biological phenomena. Thioredoxin (TRX) is one of the major components of the thiol reducing system and plays multiple roles in cellular processes such as proliferation, apoptosis, and gene expression. To investigate the molecular mechanism of TRX action, we used a yeast two-hybrid system to identify TRX-binding proteins. One of the candidates, designated as thioredoxin-binding protein-2 (TBP-2), was identical to vitamin D3 up-regulated protein 1 (VDUP1). The association of TRX with TBP-2/VDUP1 was observed in vitro and in vivo. TBP-2/VDUP1 bound to reduced TRX but not to oxidized TRX nor to mutant TRX, in which two redox active cysteine residues are substituted by serine. Thus, the catalytic center of TRX seems to be important for the interaction. Insulin reducing activity of TRX was inhibited by the addition of recombinant TBP-2/VDUP1 protein in vitro. In COS-7 and HEK293 cells transiently transfected with TBP-2/VDUP1 expression vector, decrease of insulin reducing activity of TRX and diminishment of TRX expression was observed. These results suggested that TBP-2/VDUP1 serves as a negative regulator of the biological function and expression of TRX. Treatment of HL-60 cells with 1α,25-dihydroxyvitamin D3 caused an increase of TBP-2/VDUP1 expression and down-regulation of the expression and the reducing activity of TRX. Therefore, the TRX-TBP-2/VDUP1 interaction may be an important redox regulatory mechanism in cellular processes, including differentiation of myeloid and macrophage lineages. Thioredoxin (TRX) 1The abbreviations used are: TRX, thioredoxin; ASK1, apoptosis signal-regulating kinase 1; TBP, thioredoxin-binding protein; VDUP1, vitamin D3 up-regulated protein 1; GST, glutathione S-transferase; GFP, green fluorescent protein; mAb, monoclonal antibody. is a 12-kDa ubiquitous protein that has disulfide reducing activity (1Holmgren A. Annu. Rev. Biochem. 1985; 54: 237-271Crossref PubMed Google Scholar, 2Holmgren A. Structure. 1995; 3: 239-243Abstract Full Text Full Text PDF PubMed Scopus (386) Google Scholar). TRX has two redox-active cysteine residues in its consensus sequence (Trp-Cys-Gly-Pro-Cys) and serves as a general disulfide oxido-reductase. The two cysteine residues can be reversibly oxidized to form a disulfide bond and reduced by the action of TRX reductase and NADPH (3Holmgren A. J. Biol. Chem. 1979; 254: 9113-9119Abstract Full Text PDF PubMed Google Scholar). TRX catalyzes the reduction of disulfide bonds in multiple substrate proteins. The TRX system (TRX, TRX reductase, and NADPH) is widely conserved in almost all species from bacteria to higher eukaryotes and has a wide variety of biological functions. TRX was originally identified as a hydrogen donor of ribonucleotide reductase in Escherichia coli (4Laurent T.C. Moore E.C. Reichard P. J. Biol. Chem. 1964; 239: 3436-3444Abstract Full Text PDF PubMed Google Scholar) and is an essential protein subunit of the phage T7 DNA polymerase complex (5Mark D.F. Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1976; 73: 780-784Crossref PubMed Scopus (183) Google Scholar). The simultaneous deletion of two TRX genes ofSaccharomyces cerevisiae prolonged the cell cycle (6Muller E.G.D. J. Biol. Chem. 1991; 266: 9194-9202Abstract Full Text PDF PubMed Google Scholar). The TRX homologue gene of Drosophila was identified as a gene in the deadhead locus, which is required for female meiosis and early embryonic development (7Salz H.K. Flickinger T.W. Mittendorf E. Pellicena-Palle A. Petschek J.P. Albrecht E.B. Genetics. 1994; 136: 1075-1086Crossref PubMed Google Scholar). In the human system, TRX has been cloned as adult T-cell leukemia-derived factor, produced by human T-cell leukemia virus-I-transformed T-cells, or as interleukin-1-like factor from Epstein-Barr virus transformed cells (8Wollman E.E. d'Auriol L. Rimsky L. Shaw A. Jacquot J.P. Wingfield P. Graber P. Dessarps F. Robin P. Galibert F. Bertoglio J. Fradeliz D. J. Biol. Chem. 1988; 263: 15506-15512Abstract Full Text PDF PubMed Google Scholar, 9Tagaya Y. Maeda Y. Mitsui A. Kondo N. Matsui H. Hamuro J. Brown N. Arai K. Yokota T. Wakasugi H. Yodoi J. EMBO J. 1989; 8: 757-764Crossref PubMed Scopus (529) Google Scholar). Human TRX has been reported to be involved in cell activation (10Yodoi J. Uchiyama T. Immunol. Today. 1992; 13: 405-411Abstract Full Text PDF PubMed Scopus (170) Google Scholar, 11Nakamura H. Nakamura K. Yodoi J. Annu. Rev. Immunol. 1997; 15: 351-369Crossref PubMed Scopus (1005) Google Scholar) and cell growth promotion (12Wakasugi N. Tagaya Y. Wakasugi H. Mitsui A. Maeda M. Yodoi J. Tursz T. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8282-8286Crossref PubMed Scopus (263) Google Scholar, 13Gasdaska J.R. Berggren M. Powis G. Cell Growth Differ. 1995; 6: 1643-1650PubMed Google Scholar). TRX expression seems essential for early development of the mouse embryo, because targeted disruption of the mouse TRX gene caused early embryonic lethality (14Matsui M. Oshima M. Oshima H. Takaku K. Maruyama T. Yodoi J. Taketo M.M. Dev. Biol. 1996; 178: 179-185Crossref PubMed Scopus (437) Google Scholar). Accumulating evidence has indicated the importance of the regulation of reduction and oxidation (redox regulation) in various biological phenomena (15Pahl H.L. Baeuerle P.A. Bioessays. 1994; 16: 497-502Crossref PubMed Scopus (232) Google Scholar). The TRX system and the glutathione system constitute major thiol reducing systems (16Holmgren A. J. Biol. Chem. 1989; 264: 13963-13966Abstract Full Text PDF PubMed Google Scholar). The reduced condition is preferable for DNA binding activity of various transcriptional factors such as AP-1 (17Abate C. Patel L. Rauscher III, F.J. Curran T. Science. 1990; 249: 1157-1161Crossref PubMed Scopus (1383) Google Scholar), NF-κB (18Toledano M.B. Leonard W.J. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4328-4332Crossref PubMed Scopus (598) Google Scholar), polyomavirus enhancer-binding protein 2/AML1 (19Kagoshima H. Akamatsu Y. Ito Y. Shigesada K. J. Biol. Chem. 1996; 271: 33074-33082Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar), glucocorticoid receptor (20Makino Y. Okamoto K. Yoshikawa N. Aoshima M. Hirota K. Yodoi J. Umesono K. Makino I. Tanaka H. J. Clin. Invest. 1996; 98: 2469-2477Crossref PubMed Scopus (163) Google Scholar), and estrogen receptor (21Hayashi S. Hajiro-Nakanishi K. Makino Y. Eguchi H. Yodoi J. Tanaka H. Nucleic Acids Res. 1997; 25: 4035-4040Crossref PubMed Scopus (118) Google Scholar). TRX plays an important role in the regulation of protein-nucleic acid interactions through the redox regulation of cysteine residue(s) (22Matthews J.R. Wakasugi N. Virelizier J.L. Yodoi J. Hay R.T. Nucleic Acids Res. 1992; 20: 3821-3830Crossref PubMed Scopus (741) Google Scholar,23Xanthoudakis S. Miao G. Wang F. Pan Y.C. Curran T. EMBO J. 1992; 11: 3323-3335Crossref PubMed Scopus (832) Google Scholar). The direct physical association between TRX and redox factor-1/AP-endonuclease has been demonstrated to play a key role in the AP-1 transcriptional activity (23Xanthoudakis S. Miao G. Wang F. Pan Y.C. Curran T. EMBO J. 1992; 11: 3323-3335Crossref PubMed Scopus (832) Google Scholar, 24Hirota K. Matsui M. Iwata S. Nishiyama A. Mori K. Yodoi J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3633-3638Crossref PubMed Scopus (730) Google Scholar). Cellular redox status is also important in the regulation of apoptosis (25Kroemer G. Zamzami N. Susin S.A. Immunol. Today. 1997; 18: 44-51Abstract Full Text PDF PubMed Scopus (1389) Google Scholar, 26Ueda S. Nakamura H. Masutani H. Sasada T. Yonehara S. Takabayashi A. Yamaoka Y. Yodoi J. J. Immunol. 1998; 161: 6689-6695PubMed Google Scholar). Recently, TRX has been isolated using a yeast two-hybrid system as a binding protein of a mitogen-activated protein kinase kinase kinase, apoptosis signal-regulating kinase 1 (ASK1) (27Saitoh M. Nishitoh H. Fujii M. Takeda K. Tobiume K. Sawada Y. Kawabata M. Miyazono K. Ichijo H. EMBO J. 1998; 17: 2596-2606Crossref PubMed Scopus (2112) Google Scholar). In order to investigate the molecular mechanism of TRX-dependent redox regulation in mammalian cells, we have identified TRX-binding proteins. A TRX-binding protein designated as thioredoxin-binding protein-2 (TBP-2) is identical to vitamin D3 up-regulated protein 1 (VDUP1). VDUP1 was originally reported as an up-regulated gene in HL-60 cells treated with 1α,25-dihydroxyvitamin D3 (28Chen K.S. DeLuca H.F. Biochim. Biophys. Acta. 1994; 1219: 26-32Crossref PubMed Scopus (282) Google Scholar). Although several homologous sequences of VDUP1 from mammalian species have been reported, the function of VDUP1 remains unclear. In this paper, we report how TBP-2/VDUP1 interacts with TRX and modulates the function and the expression of TRX in vitro and in vivo. Standard methods were used for DNA and RNA manipulations (29Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). TRX mutant C32S/C35S, which lacks reducing activity, was made by substituting two redox-active cysteine residues for serine residues (24Hirota K. Matsui M. Iwata S. Nishiyama A. Mori K. Yodoi J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3633-3638Crossref PubMed Scopus (730) Google Scholar, 30Kallis G.B. Holmgren A. J. Biol. Chem. 1980; 255: 10261-10265Abstract Full Text PDF PubMed Google Scholar). A cDNA of TRX or TRX C32S/C35S was fused in-frame to pGBT9 (CLONTECH) or pGEX4T-2 (Amersham Pharmacia Biotech). TRX cDNA fragment was excised byEcoRI from λgt10-TRX vector and ligated into pcDNA3 (Invitrogen) (9Tagaya Y. Maeda Y. Mitsui A. Kondo N. Matsui H. Hamuro J. Brown N. Arai K. Yokota T. Wakasugi H. Yodoi J. EMBO J. 1989; 8: 757-764Crossref PubMed Scopus (529) Google Scholar). pACT-cl.13 and pACT-cl.29 were isolated from positive colonies of yeast two-hybrid screening and contained partial coding sequences of TBP-2. The open reading frame of the TBP-2/VDUP1 was amplified by polymerase chain reaction using the following oligonucleotide primers: 5′-GGAATTCGATGGTGATGTTCAAGAAGATC-3′ and 5′-CCGCTCGAGTCACTGACAATTGTTGTTGA-3′. To prepare protein expression vectors, the TBP-2/VDUP1 open reading frame was ligated in-frame to pGADGH (CLONTECH), pcDNA3.1/His B (Invitrogen), pGEX4T-3, pRSET B (Invitrogen), or pEGFP-C1 (CLONTECH). pcDNA3.1/His B-lacZ was purchased from Invitrogen. The two-hybrid library screening was performed using the yeast MATCHMAKER two-hybrid system (CLONTECH) and human lymphocyte MATCHMAKER cDNA library (B-cell population of Epstein-Barr virus-transformed peripheral blood lymphocytes, CLONTECH) according to the manufacturer's instructions. pGBT9-TRX was used as the bait plasmid. Plasmids were introduced to yeast strains HF7c or SFY526 by a polyethylene glycol/lithium acetate method (31Bartel P. Chien C.T. Sternglanz R. Fields S. BioTechniques. 1993; 14: 920-924PubMed Google Scholar, 32Feilotter H.E. Hannon G.J. Ruddell C.J. Beach D. Nucleic Acids Res. 1994; 22: 1502-1503Crossref PubMed Scopus (226) Google Scholar). Colonies were grown on selective synthetic medium and were tested for histidine prototrophy or β-galactosidase activity using 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside. The λZAP II human placenta cDNA library (Stratagene) was screened with [α-32P]dCTP-labeled DNA probes derived from theXhoI fragment of pACT-cl.29. Positive plaques were phagemid-rescued by VCSM13 helper phage (Stratagene) according to the manufacturer's instruction. A 2.9-kilobase pair cDNA in pBluescript was sequenced on both strands. In vitrotranslated proteins were prepared using a TNT-coupled rabbit reticulocyte translation system (Promega) and [35S]methionine (Amersham Pharmacia Biotech). Bacterially expressed His6-tagged recombinant protein was prepared under denaturing conditions according to the instructions provided in the QIAexpressionist booklet (Qiagen). E. coli strain BL21 (DE3) pLysS transformed with pRSET-TBP-2/VDUP1 was treated for 4 h with 1 mm isopropyl-β-d-thiogalactoside. His6-tagged protein was purified by use of a Ni2+-nitrilotriacetic acid-agarose column. GlutathioneS-transferase (GST) fusion proteins were prepared as follows: E. coli strain XL1 Blue MRF′ transformed with each pGEX expression vector was treated for 4 h with 1 mmisopropyl-β-d-thiogalactoside. Pelleted cells were lysed in 10 mm Tris-HCl, pH 8.0, 1 mm EDTA, pH 8.0, 150 mm NaCl, and protease inhibitors (2 mmphenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 5 μg/ml aprotinin) by sonication. After centrifugation, supernatants were applied to glutathione Sepharose CL-6B columns (Amersham Pharmacia Biotech). GST fusion proteins were either eluted with the above buffer containing 10 mm reduced glutathione or immobilized on the beads and stored at −80 °C. Recombinant TRX was produced and provided by Ajinomoto Co. Inc. (Basic Research Laboratory, Kawasaki, Japan) (33Mitsui A. Hirakawa T. Yodoi J. Biochem. Biophys. Res. Commun. 1992; 186: 1220-1226Crossref PubMed Scopus (175) Google Scholar). Anti-TBP-2/VDUP1 antibody was prepared by immunization of bacterially expressed His6-TBP-2/VDUP1 protein as described before (14Matsui M. Oshima M. Oshima H. Takaku K. Maruyama T. Yodoi J. Taketo M.M. Dev. Biol. 1996; 178: 179-185Crossref PubMed Scopus (437) Google Scholar). The immune serum was further purified by affinity columns coupled with the GST-TBP-2/VDUP1 protein and protein A Sepharose CL-6B (Amersham Pharmacia Biotech) (34Harlow E. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1988Google Scholar). Anti-human TRX monoclonal antibodies (TRX-11 mAb and 21 mAb) were produced and provided by Fujirebio (Tokyo, Japan). These antibodies recognize different sites of TRX, respectively (35Kogaki H. Fujiwara Y. Yoshiki A. Kitajima S. Tanimoto T. Mitsui A. Shimamura T. Hamuro J. Ashihara Y. J. Clin. Lab. Anal. 1996; 10: 257-261Crossref PubMed Scopus (36) Google Scholar). Anti-Xpress antibody (Invitrogen) is a monoclonal antibody that recognizes the amino acid sequence Asp-Leu-Tyr-Asp-Asp-Asp-Asp-Lys, which is contained in the N terminus of His6-tagged protein that was expressed with pcDNA3.1/His. COS-7 and HEK293 cells were cultured in Dulbecco's modified Eagle's medium (Nissui Pharmaceutical) with heat-inactivated 10% fetal calf serum (Life Technologies, Inc.); HL-60 and Jurkat cells were cultured in RPMI 1640 medium (Life Technologies, Inc.) with heat-inactivated 10% fetal calf serum. COS-7 and HEK293 cells were transfected with Superfect transfection reagent (Qiagen) according to the manufacturer's instruction. After 24 h of transfection, cells were harvested and used for further studies. For the HL-60 differentiation, cells were seeded at 1 × 105 cells/ml and treated with 100 nm calcitriol (1α,25-dihydroxyvitamin D3, Wako Pure Chemical Ind.) for 3 days. In vitro translated proteins were diluted with Nonidet P-40 buffer containing 150 mm NaCl, 1.0% Nonidet P-40, and 50 mm Tris-HCl, pH 7.5. For immunoprecipitation, after preclearing by use of protein G-Sepharose (Zymed Laboratories Inc.), samples were incubated with antibody for 2 h and with protein G-Sepharose for an additional 20 min at 4 °C. For in vitro binding assay using GST fusion protein, samples were incubated with GST fusion protein. Then samples were centrifuged and washed five times with Nonidet P-40 buffer. The precipitated proteins were subjected to SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. The anti-human TRX monoclonal antibody (TRX-11 mAb) and mouse IgG1 (MOPC21, Sigma) were covalently coupled to Affi-Gel 10 beads (Bio-Rad), according to the manufacturer's instruction. Cells were washed and lysed with Nonidet P-40 buffer containing protease inhibitors (2 mm phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 5 μg/ml aprotinin). Then 10 mg of lysate was applied to each antibody column. Proteins were eluted with 100 mm glycine-HCl, pH 2.5. Eluted samples were neutralized with 110 volume of 1m Tris-HCl, pH 8.0, and dialyzed with 50 mmammonium acetate, pH 7.5. Samples were concentrated by evaporation and subjected to Western blotting analysis. GST-TRX was treated with reducing/oxidizing reagents according to previous report (27Saitoh M. Nishitoh H. Fujii M. Takeda K. Tobiume K. Sawada Y. Kawabata M. Miyazono K. Ichijo H. EMBO J. 1998; 17: 2596-2606Crossref PubMed Scopus (2112) Google Scholar). Breifly, GST-TRX immobilized on the beads was treated with Nonidet P-40 buffer containing reducing/oxidizing reagents for 15 min at room temperature. Treated beads were centrifuged and washed five times with degassed Nonidet P-40 buffer by batch method. Then in vitrotranslated TBP-2/VDUP1 protein was diluted with Nonidet P-40 buffer and added to the treated beads. After incubation for 2 h, samples were centrifuged and washed five times with Nonidet P-40 buffer. The precipitated proteins were subjected to SDS-polyacrylamide gel electrophoresis and visualized by autoradiography. Cell extracts were prepared with lysis buffer (150 mm NaCl, 1.0% Nonidet P-40, 1.0% sodium deoxycolate, 0.1% SDS, 20 mm Tris-HCl, pH 7.5, 5 mm EDTA, 2 mm phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 5 μg/ml aprotinin). SDS-polyacrylamide gel electrophoresis and Western blotting were performed as described previously (36Sasada T. Iwata S. Sato N. Kitaoka Y. Hirota K. Nakamura K. Nishiyama A. Taniguchi Y. Takabayashi A. Yodoi J. J. Clin. Invest. 1996; 97: 2268-2276Crossref PubMed Scopus (184) Google Scholar). Total RNA from cultured cells was extracted using TRIzol reagent (Life Technologies, Inc.) according to the manufacturer's instruction. Electrophoresis and Northern blotting were performed as described previously (37Maruyama T. Kitaoka Y. Sachi Y. Nakanoin K. Hirota K. Shiozawa T. Yoshimura Y. Fujii S. Yodoi J. Mol. Hum. Reprod. 1997; 3: 989-993Crossref PubMed Scopus (46) Google Scholar). To estimate the reducing activity of TRX, insulin reducing assay was performed according to a previous report (36Sasada T. Iwata S. Sato N. Kitaoka Y. Hirota K. Nakamura K. Nishiyama A. Taniguchi Y. Takabayashi A. Yodoi J. J. Clin. Invest. 1996; 97: 2268-2276Crossref PubMed Scopus (184) Google Scholar) with slight modifications. In our assay, yeast TRX reductase was used and was able to reduce recombinant human TRX. Yeast TRX reductase was provided by Oriental Yeast Co. Ltd. (Tokyo, Japan). The decrease in absorbance at 340 nm was recorded by use of a THERMOMAX micro plate reader (Molecular Devices) to detect maximal NADPH consumption rate (Vmax, millioptical density at 340 nm/min). As a control, samples were incubated with the reaction mixture without insulin. Each value was calculated according to a method previously reported (36Sasada T. Iwata S. Sato N. Kitaoka Y. Hirota K. Nakamura K. Nishiyama A. Taniguchi Y. Takabayashi A. Yodoi J. J. Clin. Invest. 1996; 97: 2268-2276Crossref PubMed Scopus (184) Google Scholar). We used the yeast two-hybrid system to clone genes encoding TRX-binding protein using a cDNA library of B cell population of Epstein-Barr virus-transformed human peripheral blood lymphocyte. Among approximately 1.8 × 106 yeast transformants screened, nine colonies showed histidine prototrophy and β-galactosidase activity. Isolated plasmids were classified by restriction enzyme excision or DNA sequencing into three groups that were designated as thioredoxin-binding proteins. Because double transformants with pGBT-TRX and each plasmid belonging to one group, TBP-2, showed strongly positive phenotypes of histidine prototrophy and β-galactosidase activity, we chose TBP-2 for further study (Fig. 1). Data base searching of pACT-cl.29 revealed that TBP-2 has homology with TRT407–2 and vitamin D3 up-regulated protein 1 (VDUP1). TRT407–2 was reported as a gene screened by an RNA fingerprinting method in mink Mv1Lu cells (38McClelland M. Ralph D. Cheng R. Welsh J. Nucleic Acids Res. 1994; 22: 4419-4431Crossref PubMed Scopus (39) Google Scholar), and VDUP1 was reported as an up-regulated gene in HL-60 cells treated with 1α,25-dihydroxyvitamin D3 (28Chen K.S. DeLuca H.F. Biochim. Biophys. Acta. 1994; 1219: 26-32Crossref PubMed Scopus (282) Google Scholar). Another candidate (TBP-1) was identical to human p40 phox, a cytosolic phagocyte oxidase component (39Nishiyama A. Ohno T. Iwata S. Matsui M. Hirota K. Masutani H. Nakamura H. Yodoi J. Immunol. Lett. 1999; 68: 155-159Crossref PubMed Scopus (39) Google Scholar). We attempted to clone the full-length cDNA of TBP-2 for further study. To isolate the full-length cDNA of TBP-2, a human placenta cDNA library was screened, using probes derived from the insert DNA of pACT-cl.29. DNA sequencing analysis of isolated 2.9-kilobase pair cDNA revealed that TBP-2 was identical to VDUP1. To examine the interaction between TRX protein and TBP-2/VDUP1 protein, anin vitro binding assay was performed.35S-Labeled proteins were prepared by in vitrotranslation. 35S-Labeled His6-TBP-2/VDUP1 protein was co-immunoprecipitated with TRX by anti-TRX monoclonal antibodies (Fig. 2A,lane 1) but not by control mouse IgG1 (Fig. 2A,lane 2). 35S-Labeled His6-β-galactosidase protein was not co-immunoprecipitated with TRX (Fig. 2A, lane 3). In addition, 35S-labeled TRX was co-immunoprecipitated within vitro translated His6-TBP-2/VDUP1 protein by anti-Xpress antibody but not by control mouse IgG1 (Fig. 2B). We then analyzed whether TRX interacts with TBP-2/VDUP1 in vivo. Cell lysates from human Jurkat cells were applied to an affinity column of either anti-TRX monoclonal antibody or control IgG1. Eluted proteins were subjected to Western blotting analysis with affinity purified anti-TBP-2/VDUP1 antibody. TBP-2/VDUP1 was detected in eluates from the anti-TRX antibody column (Fig. 2C,lane 1) but not from the control IgG1 column (lane 2). Therefore, these results demonstrated the interaction of TRX with TBP-2/VDUP1 in vitro and in vivo. Because TRX is a redox active protein, we next tested whether TRX-TBP-2/VDUP1 interaction is influenced by the redox status of TRX. GST-fused TRX was pretreated with reducing/oxidizing reagent, dithiothreitol, hydrogen peroxide, or diamide (a sulfhydryl-specific oxidant) and subjected to an in vitro binding assay. Whereas the TRX-TBP-2/VDUP1 interaction was unaffected by treatment with dithiothreitol, the interaction was markedly inhibited by treatment with hydrogen peroxide or diamide, suggesting that the reduced form of TRX is critically important for the interaction (Fig. 3). To examine whether TRX-TBP-2/VDUP1 interaction requires intact redox active sites of TRX, we used mutant TRX C32S/C35S in which redox active cysteine residues are substituted to serine residues. In an in vitro binding assay, 35S-labeled His6-TBP-2/VDUP1 protein was co-precipitated with GST-TRX but not with GST-TRX C32S/C35S (Fig. 4A). Using a yeast two-hybrid system to test the interaction in vivo, S. cerevisiae strain HF7c was transformed with pGADGH-TBP-2/VDUP1 and either pGBT9-TRX or pGBT9-TRX C32S/C35S. Transformed colonies of pGBT9-TRX grew well on synthetic medium lacking histidine, whereas transformed colonies of pGBT9-TRX C32S/C35S failed to grow (Fig. 4B). These data showed that these redox active cysteine residues are required for the TRX-TBP-2/VDUP1 interaction. TRX has a disulfide reducing activity and cleaves a disulfide bond in substrates such as insulin. We examined the effect of TBP-2/VDUP1 on TRX activity by the insulin reducing assay (36Sasada T. Iwata S. Sato N. Kitaoka Y. Hirota K. Nakamura K. Nishiyama A. Taniguchi Y. Takabayashi A. Yodoi J. J. Clin. Invest. 1996; 97: 2268-2276Crossref PubMed Scopus (184) Google Scholar, 40Holmgren A. Björnstedt M. Methods Enzymol. 1995; 252: 199-208Crossref PubMed Scopus (823) Google Scholar). In our experiments, yeast TRX reductase was used and was able to reduce recombinant human TRX (data not shown). As shown in Fig. 5, a significant decrease of the reducing activity of TRX was observed in cellular extracts of COS-7 or HEK293 cells transiently transfected with a green fluorescent protein (GFP)-fused TBP-2/VDUP1 protein expression vector (Fig. 5A), in comparison with those of cells transfected with a GFP expression vector (Fig. 5A,lane 1). Similar results were obtained in experiments using His6-TBP-2/VDUP1 expression vector (data not shown). To confirm the inhibitory effect of TBP-2/VDUP1 on the reducing activity of TRX, we tested recombinant GST-TBP-2/VDUP1 protein in the insulin reducing assay. The reducing activity of TRX was repressed to less than 50% by the addition of 1 μm GST-TBP-2/VDUP1 protein, indicating that TBP-2/VDUP1 protein inhibits the disulfide reducing activity of TRX in vitro (Fig. 5B). We examined the effect of TBP-2/VDUP1 on TRX expression. As shown in Fig. 6, Western blotting analysis demonstrated that expression of TRX protein was significantly down-regulated in COS-7 cells transiently transfected with GFP-TBP-2/VDUP1 expression vector (Fig. 6). Similar results were obtained in experiments using His6-TBP-2/VDUP1 expression vector (data not shown). Thus, TBP-2/VDUP1 protein down-regulated TRX protein expression as well. Because TBP-2/VDUP1 was originally reported as an up-regulated gene in 1α,25-dihydroxyvitamin D3 treatment in HL-60 cells (28Chen K.S. DeLuca H.F. Biochim. Biophys. Acta. 1994; 1219: 26-32Crossref PubMed Scopus (282) Google Scholar), we analyzed TBP-2/VDUP1 and TRX expression in the differentiation of 1α,25-dihydroxyvitamin D3-induced HL-60 cells. There was a gradual increase of TBP-2/VDUP1 mRNA after treatment with 1α,25-dihydroxyvitamin D3 (Fig. 7A). After 72 h, the TBP-2/VDUP1 mRNA was enhanced 9-fold over that before the treatment. In contrast, 72 h after the treatment, the TRX mRNA level markedly declined to less than 20% compared with that before the treatment. This inverted expression pattern was also observed in protein expression. The expression of TBP-2/VDUP1 protein was enhanced after treatment of 1α,25-dihydroxyvitamin D3 (Fig. 7B). In contrast, 48 h after the treatment, the amount of TRX protein was reduced to half of that before treatment (Fig. 7B). We then analyzed the reducing activity of TRX in HL-60 cells treated with 1α,25-dihydroxyvitamin D3. Insulin reducing activity of the cell lysates decreased to 70% of the control value within 24 h and to 60% by 72 h (Fig. 8). Thus, TRX expression and its reducing activity were down-regulated in HL-60 cells treated with 1α,25-dihydroxyvitamin D3, whereas TBP-2/VDUP1 expression was up-regulated.Figure 8The reducing activity of TRX in HL-60 cells treated with 1α,25-dihydroxyvitamin D3. Cell extracts (10 μg) of HL-60 cells treated with 1α,25-dihydroxyvitamin D3 were collected at indicated hours. The TRX activities were determined by use of the insulin reducing assay. Activities for samples are shown relative toVmax of control (0 h), which is assigned as 100%. Data shown are representative of two-independent experiments. The results are the means ± S.D. of three samples.View Large Image Figure ViewerDownload Hi-res image Download (PPT) In our search for interacting molecules with TRX, we isolated VDUP1 as a TBP-2 using a yeast two-hybrid system. We characterized TBP-2/VDUP1 as a TRX-binding protein. The interaction was dependent on the redox status of TRX. Moreover, the inhibitory effect of TBP-2/VDUP1 on the reducing activity of TRX and the expression was observed in TBP-2/VDUP1-overexpressed cells as well as HL-60 cells treated with 1α,25-dihydroxyvitamin D3. VDUP1 was originally reported as an up-regulated gene in HL-60 cells stimulated by 1α,25-dihydroxyvitamin D3 (28Chen K.S. DeLuca H.F. Biochim. Biophys. Acta. 1994; 1219: 26-32Crossref PubMed Scopus (282) Google Scholar). The function of VDUP1 is still unclear, although several homologous sequences from mammalian species have been reported. The rat VDUP1 homologue was isolated as a down-regulated gene byN-methyl-N-nitrosourea in rat mammary tumor (41Young L.H. Yang X. Voigt J.M. Mol. Carcinogen. 1996; 15: 251-260Crossref PubMed Scopus (24) Google Scholar). There are two transcripts homologous to VDUP1, TRT407–2 and TRT407–9, whose expressions were induced by cycloheximide and repressed by transforming growth factor-β in mink Mv1Lu cells (38McClelland M. Ralph D. Cheng R. Welsh J. Nucleic Acids Res. 1994; 22: 4419-4431Crossref PubMed Scopus (39) Google Scholar). The interaction of TRX with TBP-2/VDUP1 was demonstrated both in vitro and in vivo. TRX treated with oxidizing reagents was incapable of binding with TBP-2/VDUP1. It should be noted that the effect of oxidizing reagents was detectable at a low concentration (10 μm), in which aggregation of TRX was avoided (42Sato N. Iwata S. Nakamura K. Hori T. Mori K. Yodoi J. J. Immunol. 1995; 154: 3194-3203PubMed Google Scholar). Thus, only the reduced form of TRX appears to interact with TBP-2/VDUP1. There is a possibility that the presence of unreacted reagents has effects on the interaction between TRX and TBP-2/VDUP1. However, it seems unlikely because the GST-TRX beads used in the experiments were intensively washed after treatment with reducing/oxidizing reagents. In addition, the interaction was hardly inhibited by direct addition of the low concentration (10 μm) of reducing/oxidizing reagents to the TRX and TBP-2/VDUP1 reaction mixture. 2A. Nishiyama and J. Yodoi, unpublished data. Therefore, TRX-TBP-2/VDUP1 interaction appears highly dependent on the redox status of TRX. In addition, we used substituted TRX to analyze the involvement of the active site of TRX in the interaction with TBP-2/VDUP1. TRX C32S/C35S was not able to interact with TBP-2/VDUP1 either in the in vitro binding assay or the yeast two-hybrid system. These results strongly suggest that the TRX active site is important for the TRX-TBP-2/VDUP1 interaction. Analysis of the crystal structure of TRX has indicated that the active-site conformation of TRX C32S/C35S is very similar to that of oxidized TRX (43Weichsel A. Gasdaska J.R. Powis G. Montfort W.R. Structure. 1996; 4: 735-751Abstract Full Text Full Text PDF PubMed Scopus (335) Google Scholar). Thus, the incapability of the TRX C32S/C35S and oxidized TRX to interact with TBP-2/VDUP1 does not seem to be caused by a large scale conformational change. In addition, TBP-2/VDUP1 inhibited TRX activity in insulin reducing assay. If TBP-2/VDUP1 is a substrate for TRX, inhibitory effect was not observed in insulin reducing assay. Our result suggests that TBP-2/VDUP1 is not a substrate for TRX. Therefore, interaction of TRX with TBP-2/VDUP1 may be a direct protein-protein interaction manner. Physical interactions of TRX with other proteins have been reported. TRX has been isolated as an ASK1-binding protein using a yeast two-hybrid system (27Saitoh M. Nishitoh H. Fujii M. Takeda K. Tobiume K. Sawada Y. Kawabata M. Miyazono K. Ichijo H. EMBO J. 1998; 17: 2596-2606Crossref PubMed Scopus (2112) Google Scholar). TRX directly binds to ASK1 and inhibits the apoptosis signal conducted by ASK1. These results suggest that the direct protein-protein interaction of TRX with its binding protein may be a basic mechanism of the redox regulation of cellular processes. In the TBP-2/VDUP1-overexpressed cells, decrease of the reducing activity of TRX was observed. This result raised the possibility that TBP-2/VDUP1 affects enzymatic action of TRX or TRX expression. Therefore, we examined the effect of TBP-2/VDUP1 on insulin reducing activity of TRX and TRX protein expression. The insulin reducing assay using recombinant GST-TBP-2/VDUP1 protein clearly demonstrated the inhibitory effect of TBP-2/VDUP1 on the reducing activity of TRX. Structural interference caused by GST seems unlikely, because experiments using cell lysates transfected with His6-TBP-2/VDUP1 expression vector also showed the decreased TRX activity. Although recombinant TBP-2/VDUP1 protein demonstrated an inhibitory effect, the effect was not complete, probably because the TBP-2/VDUP1 protein concentration was insufficient. Additionally, suppression of TRX protein expression was observed in cells transiently transfected with TBP-2/VDUP1 expression vector. In addition to its inhibitory effect, TBP-2/VDUP1 may be involved in TRX protein expression, and the interaction of TBP-2/VDUP1 with TRX might be required for the suppression mechanism of TRX expression. Based on these findings, we hypothesize that TBP-2/VDUP1 inhibits the redox-regulatory action of TRX. We observed the decrease of TRX expression and the reducing activity in HL-60 cells treated with 1α,25-dihydroxyvitamin D3 whose TBP-2/VDUP1 expression was enhanced as described in a previous report (28Chen K.S. DeLuca H.F. Biochim. Biophys. Acta. 1994; 1219: 26-32Crossref PubMed Scopus (282) Google Scholar). An intriguing possibility is that up-regulation of TBP-2/VDUP1 expression is involved in the decrease of TRX expression and reducing activity. The decrease of TRX mRNA also was observed in HL-60 cells treated with 1α,25-dihydroxyvitamin D3. 1α,25-Dihydroxyvitamin D3 treatment or action of TBP-2/VDUP1 may be involved in the decrease of TRX mRNA. This inverted pattern of TRX and TBP-2/VDUP1 expression also was observed in cell cycle-synchronized cells.2 The suppression mechanism of TRX expression in HL-60 cells treated with 1α,25-dihydroxyvitamin D3 and the involvement of TBP-2/VDUP1 should be further analyzed. 1α,25-Dihydroxyvitamin D3 is an essential biologically active molecule and is important for regulation of calcium homeostasis and secretion of hormone (44Suda T. Shinki T. Takahashi N. Annu. Rev. Nutr. 1990; 10: 195-211Crossref PubMed Scopus (101) Google Scholar, 45Darwish H. DeLuca H.F. Crit. Rev. Eukaryotic Gene Expression. 1993; 3: 89-116PubMed Google Scholar). 1α,25-Dihydroxyvitamin D3 also is a potent inducer of myeloid leukemic cell differentiation (46Miyaura C. Abe E. Kuribayashi T. Tanaka H. Konno K. Nishii Y. Suda T. Biochem. Biophys. Res. Commun. 1981; 102: 937-943Crossref PubMed Scopus (442) Google Scholar, 47Mangelsdorf D.J. Koeffler H.P. Donaldson C.A. Pike J.W. Haussler M.R. J. Cell Biol. 1984; 98: 391-398Crossref PubMed Scopus (422) Google Scholar) and can inhibit the growth of cancer cells from several different tissues (48Frampton R.J. Omond S.A. Eisman J.A. Cancer Res. 1983; 43: 4443-4447PubMed Google Scholar, 49Dokoh S. Donaldson C.A. Haussler M.R. Cancer Res. 1984; 44: 2103-2109PubMed Google Scholar). TRX has been reported to have cell-growth promoting effects (12Wakasugi N. Tagaya Y. Wakasugi H. Mitsui A. Maeda M. Yodoi J. Tursz T. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 8282-8286Crossref PubMed Scopus (263) Google Scholar, 13Gasdaska J.R. Berggren M. Powis G. Cell Growth Differ. 1995; 6: 1643-1650PubMed Google Scholar), although TRX is involved in growth inhibitory mechanism in some system (50Deiss L.P. Kimchi A. Science. 1991; 252: 117-120Crossref PubMed Scopus (179) Google Scholar). Human TRX was discovered as adult T-cell leukemia-derived factor from human T-cell lymphotropic virus-I transformed T-cells (9Tagaya Y. Maeda Y. Mitsui A. Kondo N. Matsui H. Hamuro J. Brown N. Arai K. Yokota T. Wakasugi H. Yodoi J. EMBO J. 1989; 8: 757-764Crossref PubMed Scopus (529) Google Scholar). Elevated levels of TRX expression has been observed in some human tumors (51Fujii S. Nanbu Y. Nonogaki H. Konishi I. Mori T. Masutani H. Yodoi J. Cancer. 1991; 68: 1583-1591Crossref PubMed Scopus (104) Google Scholar, 52Nakamura H. Masutani H. Tagaya Y. Yamauchi A. Inamoto T. Nanbu Y. Fujii S. Ozawa K. Yodoi J. Cancer. 1992; 69: 2091-2097Crossref PubMed Scopus (149) Google Scholar). In addition, TRX transfected cells exhibited severalfold increased colony formation in soft agarose, whereas a redox-inactive mutant TRX C32S/C35S acts in a dominant negative manner to inhibit proliferation (53Gallegos A. Gasdaska J.R. Taylor C.W. Paine-Murrieta G.D. Goodman D. Gasdaska P.Y. Berggren M. Briehl M.M. Powis G. Cancer Res. 1996; 56: 5765-5770PubMed Google Scholar). Accordingly, the decrease of TRX function due to the TRX-TBP2/VDUP1 interaction may be one of the mechanisms through which 1α,25-dihydroxyvitamin D3 exerts its growth inhibitory effect (54Yang X. Young L.H. Voigt J.M. Breast Cancer Res. Treat. 1998; 48: 33-44Crossref PubMed Scopus (43) Google Scholar). In conclusion, we identified TBP-2/VDUP1 as a new TRX-binding protein. Stable interaction of TRX with TBP-2/VDUP1 was detected in vitro and in vivo. TBP-2/VDUP1 had an inhibitory effect on TRX-dependent reducing activity. Suppression of TRX protein expression also was observed in cells transfected with TBP-2/VDUP1 expression vector. Our results suggest the possibility that TBP-2/VDUP1 acts as an endogenous negative regulator of TRX, although the precise mechanism of this regulation is not clear yet. TRX-TBP-2/VDUP1 interaction may play an important role in the redox regulation of various cellular processes such as growth and differentiation of the cells sensitive to a variety of inducers, including 1α,25-dihydroxyvitamin D3 responses. We thank S. Toyama for helpful technical advice, W. Brown for review of the manuscript, H. Takei and H. Yamanaka for communicating their results before publication, Y. Yamaguchi for technical assistance, and Y. Kanekiyo for secretarial help.
