Redox Regulation of the Embryonic Stem Cell ... - Wiley Online Library

4 downloads 0 Views 178KB Size Report
YING GUO,a LAWRENCE EINHORN,a MARK KELLEY,b KIICHI HIROTA,c JUNJI YODOI,c ROLLAND. REINBOLD,d HANS SCHOLER,d HEATHER RAMSEY,e ...
Stem C ells R C

® This material is protected by U.S. Copyright law. Unauthorized reproduction is prohibited. For reprints contact: [email protected]

apid ommunication

Redox Regulation of the Embryonic Stem Cell Transcription Factor Oct-4 by Thioredoxin YING GUO,a LAWRENCE EINHORN,a MARK KELLEY,b KIICHI HIROTA,c JUNJI YODOI,c ROLLAND REINBOLD,d HANS SCHOLER,d HEATHER RAMSEY,e ROBERT HROMASe a

The Walther Oncology Center, bDepartment of Pediatrics and Wells Center for Pediatric Research, James Whitcomb Riley Hospital for Children, Indiana University Medical Center, Indianapolis, Indiana; cDepartment of Anesthesia, Kyoto University Hospital, and the Institute for Virus Research, Kyoto University, Kyoto, Japan; d The Center for Animal Transgenesis and Germ Cell Research, New Bolton Center, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; eDepartment of Internal Medicine, and the Cancer Treatment and Research Center, the University of New Mexico, Albuquerque, New Mexico Key Words. Embryonic stem cells · Transcription factors · Differentiation · Oxidation

A BSTRACT Oct-4 is a transcriptional regulator required to maintain the totipotentiality of embryonic stem (ES) cells. Downregulation of its activity is required for proper differentiation of the blastocyst during uterine implantation. Uterine implantation and subsequent vascularization increase oxygen exposure of the developing embryo, thereby altering the intracellular reduction-oxidation status. We tested whether Oct-4 could be regulated by these changes in reduction-oxidation status. We found that Oct4 DNA binding was exquisitely sensitive to abrogation by oxidation but that the DNA binding of another ES cell transcription factor, FoxD3, was much less sensitive to

oxidation. The reducing enzyme Thioredoxin (but not Ape-1) could restore DNA-binding activity of Oct-4. Thioredoxin was less effective at restoring the DNA-binding ability of FoxD3. It was also found that Thioredoxin (but not Ape-1) could physically associate with cysteines in the POU domain of Oct-4. Finally, overexpressing normal Thioredoxin increased the transcriptional activity of Oct4, while overexpressing a mutant Thioredoxin decreased the transcriptional activity of Oct-4. These data imply that ES cell transcription factors are differentially sensitive to oxidation and that Thioredoxin may differentially regulate ES cell transcription factors. Stem Cells 2004;22:259-264

INTRODUCTION Previous studies have shown that high oxygen concentrations are deleterious to early mammalian embryonic development [1-3]. Furthermore, blastocysts cultured under decreased O2 tension correlate more closely with in-vivorecovered blastocysts than in vitro blastocysts cultured

under normal O2 tension [3-6]. However, during uterine implantation, which occurs at gastrulation, the exposure of the developing embryo to oxygen increases. These changes in the reduction-oxidation (redox) state within the embryo may affect the activity of transcription factors, which may alter gene-expression patterns.

Correspondence: Robert Hromas, M.D., Department of Internal Medicine, and the Cancer Treatment and Research Center, University of New Mexico, 900 Camino de Salud, Albuquerque, New Mexico 87131, USA. Telephone: 505-272-5837; Fax: 505272-5865; email: [email protected] Received September 16, 2003; accepted for publication December 10, 2003. ©AlphaMed Press 1066-5099/2004/$12.00/0

STEM CELLS 2004;22:259-264

www.StemCells.com

Guo, Einhorn, Kelley et al. Some transcription factors are known to be regulated by changes in the redox state [7-10]. These include PEBP2, AP-1, p53, and NF-κΒ. The redox regulation of these transcription factors occurs through conserved cysteine residues in the DNA-binding domains of these proteins [7-12]. There are two important intracellular enzymes, Thioredoxin and Ape-1/Ref-1, which can restore DNA binding of oxidized transcriptional regulators by reducing critical cysteines [11, 12]. Lineage commitment and stage progression during embryonic stem (ES) cell development is controlled by transcription factors. Oct-4 and FoxD3 are two transcription factors that are preferentially expressed and active in ES cells. Downregulation of Oct-4 is ultimately essential for mammalian ES cells to differentiate, and, conversely, expression of Oct-4 is required for maintenance of the undifferentiated state of ES cells [13-17]. FoxD3 is required for appropriate survival of ES cells, but it remains after ES differentiation to define the neural crest lineage [18]. There is evidence that continued FoxD3 expression after downregulation of Oct-4 may be important in stimulating early lineage commitment [18]. This study found that there was an exquisite redox regulation of Oct-4 by Thioredoxin but less redox regulation of FoxD3, implying that intracellular redox states governed by Thioredoxin could play a novel specific regulatory role in ES cell totipotentiality and lineage decisions. MATERIALS AND METHODS Recombinant Protein Production Recombinant protein was expressed in BL21 E. coli using pET24-Oct-4, pQE30-Thioredoxin, and pGEX-Ape1 and purified as we previously described [7-9, 11, 18]. Briefly, purified expression vector was transduced into BL21 bacteria, transduced bacteria selected in ampicillin, and induced to express the desired protein with isopropylβ-d-thiogalactopyranoside (IPTG). Bacteria were disrupted by sonication, and the sonicate was run over the appropriate Ni or glutathione column. After the column was washed three times, recombinant protein was eluted with either Ni or glutathione and then was tested for purity by silver staining after SDS-PAGE and for specificity by Western analysis. Electrophoretic Mobility-Shift DNA-Binding Assays (EMSA) EMSA for Oct-4 and FoxD3 were performed as we previously described [18, 19]. Both Oct-4 and FoxD3 protein used the 6PORE DNA binding sequence, 5′-AAGTTAA AATCACATTTGAAATGCAAATGG. Briefly, 10 ng of radiolabeled duplex oligonucleotide were incubated at

260

room temperature with 30 ng of recombinant FoxD3 or Oct-4 protein for 20 min in 20 µl of 10 mM Tris (pH 7.5), 100 mM KCl, 10% glycerol, 2 mM MgCl2, and 1 µg poly(dI-dC). This was then electrophoresed on a nondenaturing 4% polyacrylamide gel (30:1 bis) in recirculating 0.25 × TBE at 10 V/cm for 2 hours. The gel was then dried and autoradiographed. Depending on the experiment, varying concentrations of dithiothreitol (DTT), Diamide, Thioredoxin, or Ape-1 were added to the binding buffer and incubated with Oct-4 or FoxD3 protein on ice for 15 min before radiolabeled duplex oligonucleotide and binding buffer were added. Both Thioredoxin and Ape-1/Ref-1 were reduced in 2 mM of DTT prior to use, and then the DTT was dialyzed away in 10 mM of Tris at pH 7.5. The relative amount of bound to free DNA can be established using densitometric analysis of the EMSA autoradiogram. In Vitro Disulfide Cross-Linking Interaction Assay The ability of pure Thioredoxin to physically interact with pure Oct-4 was tested as we previously described [7-9]. Pure recombinant Thioredoxin (300 ng) was incubated with purified Oct-4 (150 ng) for 30 min at room temperature with 1 mm DTT. Cysteines interacting between the proteins were then cross-linked by adding 2 mm Diamide at room temperature in 20 µl PBS, pH 7.3, for 30 minutes. The reaction mixture was denatured for 5 min at 90ºC in SDSPAGE buffer with or without 1% 2-mercaptoethanol as a reducing agent. Each reaction was applied to a 4%-12% SDS-PAGE solution and electrophoresed. The Oct-4 with the oxidoreductase enzyme Thioredoxin covalently bound to free cysteines was defined by Western blot using specific antisera against Oct-4. Cotransfection Assays All transfections were performed three times in triplicate in the human embryonic kidney 293 cell line using calcium phosphate as we previously described [18]. One million cells were plated in 100-cM culture dishes 24 hours before transfection. Equal amounts of DNA, normalized with empty vector, were precipitated using calcium phosphate, then incubated with the 293 cells for 18 hours, and then washed off. Cells were harvested 48 hours after transfection, and luciferase activity was examined. Transfection efficiencies were normalized using RSV/β-galactosidase to obtain relative luciferase units (RLU). The expression vectors for pCMV-Oct-4, pCDNA3-TRX/wt (wild type), and pCDNA3-TRX/dm (negatively competing mutant) have been previously described [7-9, 18, 19]. Student’s t-test was used to establish statistically significant differences.

261

Redox Regulation of Factor Oct-4 by Thioredoxin

RESULTS Effect of Oxidation on ES Cell Transcription Factor DNA Binding Ability We showed previously that Oct-4 and FoxD3 can bind to the 6PORE sequence and activate transcription through that sequence-specific DNA binding [18, 19]. The 6PORE sequence is the core of the Osteopontin enhancer, which is active in ES cells, and in the blastocyst, but is downregulated as Oct-4 expression decreases during gastrulation [14-16]. The DNA binding of Oct-4 to 6PORE was analyzed under oxidizing and reducing conditions [18, 19]. Treatment with increasing concentrations of the oxidative agent Diamide abolished Oct-4 DNA-binding ability to 6PORE (Fig. 1A). The effect on Oct-4 was observed at concentrations of Diamide that were equimolar to that of DTT present. Increasing concentrations of DTT reversed the effect of Diamide on Oct-4 DNA binding (Fig. 1A). To test whether embryonic transcription factors were differentially sensitive to oxidation, FoxD3 was analyzed for its ability to bind DNA in increasing concentrations of Diamide (Fig. 1B). FoxD3 required fivefold more Diamide than did Oct-4 to abolish DNA binding, indicating that it is far less sensitive to oxidation than is Oct-4. Thioredoxin but Not Ape-1 Restores DNA Binding of Oxidized ES Cell Transcription Factors Thioredoxin has an important role in regulating the redox state of some transcription factors [7-10, 12]. To test the effects of Thioredoxin on oxidized Oct-4, increasing concentrations of Thioredoxin were added to the EMSA DNA-binding reactions containing Diamide (Fig. 2A). Full DNA-binding ability of oxidized Oct-4 (as compared with the control, reduced Oct-4) was restored when 50 ng of Thioredoxin was added. Ape-1 is another important reductase that regulates many transcription factors during changes in redox status [11]. We tested whether Ape-1 could reverse the oxidation of Oct-4 in the DNA-binding assay (Fig. 2B). The binding ability of Oct-4 was lost at a concentration of 2 mM Diamide. In an attempt to reverse the effect of Diamide, increasing concentrations of reduced Ape-1 were added, but no change in Oct-4 DNA binding was observed, indicating that Thioredoxin (but not Ape-1) could restore the DNA-binding ability of oxidized Oct-4. To test whether the effect of Thioredoxin was specific for Oct-4 or general for other ES cell-transcription factors, the ability of Thioredoxin to restore the DNA-binding ability of FoxD3 was examined (Fig. 2C). Full DNA-binding ability of oxidized FoxD3 (compared with a reduced

Figure 1. Redox regulation of Oct-4 versus FoxD3 DNA binding ability. A) Pure recombinant DTT-reduced Oct-4 protein can bind to the Osteopontin enhancer sequence 6PORE. Without DTT present, Oct-4 does not bind DNA [18, 19]. However, the DNA-binding ability of Oct-4 was abolished by increasing the concentration of Diamide, an agent that oxidizes free cysteine groups. The minimum concentration to abrogate DNA binding for DTT-reduced Oct-4 is 2 mM Diamide, equimolar to the DTT present. Under a Diamide-oxidizing environment, increasing concentrations of DTT reversed the oxidation of Oct-4 and restored DNA binding. B) The effect of oxidation on FoxD3 DNA binding. Pure recombinant-reduced FoxD3 is significantly less sensitive than Oct-4 to oxidation by Diamide, as analyzed by EMSA DNA-binding assays. FoxD3 requires a much higher concentration of Diamide to block DNA binding than does Oct-4. The lower band represents unbound radiolabeled duplex 6PORE oligonucleotide, and the upper, more slowly migrating band represents protein bound to the 6PORE sequence in these autoradiograms of electrophoretic mobility shift assays. Each experiment in this figure was performed four times with identical results.

control) was not restored until 400 ng of Thioredoxin were added. Thus, oxidized FoxD3 required eightfold more Thioredoxin than did Oct-4 to restore full DNA binding, indicating that Thioredoxin has a greater specificity for Oct-4 than do other ES cell transcription factors. Direct Association between Thioredoxin and Oct-4 Thioredoxin reversed the effects of Diamide on Oct-4, but the mechanism of the reversal of Oct-4 oxidation was not clear. Thioredoxin has been previously shown to interact

Guo, Einhorn, Kelley et al.

262

Figure 3. Thioredoxin physically associates with dimeric Oct-4 through disulfide linkages. Pure recombinant Thioredoxin protein (TRX, 300 ng) was incubated with pure recombinant Oct-4 (150 ng) in the presence of Diamide, which will cross-link cysteines that are physically interacting by oxidizing their sulfhydryl groups. After incubation, the complexes were resolved by denaturing polyacrylamide gel electrophoresis under reducing (2-ME) or oxidizing conditions (Diamide); the complexes were then detected by Oct-4 antisera. Oct4 commonly dimerizes in order to bind DNA, as is seen here. This experiment was performed three times with identical results.

with various target molecules through disulfide linkages [710, 12]. Disulfide cross-linking experiments were conducted in order to assess whether pure recombinant Thioredoxin formed direct intermediates with free cysteines of purified Oct-4 through disulfide linkages (Fig. 3). In these experiments, Oct-4 antisera recognized a band of a molecular weight that could only be consistent with the Oct-4 dimer cross-linked to Thioredoxin.

Figure 2. Thioredoxin restores oxidized Oct-4 DNA binding. A) Oct-4 binding ability can be abolished by the oxidizing agent Diamide. Increasing concentrations of pure recombinant Thioredoxin (TRX) could reverse the Diamide effect and restore DNA binding of Oct-4. B) Ape-1 cannot restore oxidized Oct-4 DNA binding. Increasing concentrations of pure recombinant Ape-1 cannot reverse the Diamide effect on Oct-4. In addition, Ape-1 also could not restore the DNA-binding ability of FoxD3. C) More Thioredoxin is required to restore oxidized FoxD3 DNA binding. Once FoxD3 is oxidized by Diamide, it is less responsive to Thioredoxin restoring its ability to bind DNA than is Oct-4. Eightfold more Thioredoxin is required before oxidized FoxD3 DNA binding nears that of the reduced control. The lower band represents unbound radiolabeled duplex 6PORE oligonucleotide, and the upper, more slowly migrating band represents protein bound to the 6PORE sequence in these autoradiograms of electrophoretic mobility shift assays. Each experiment in this figure was performed at least three times with identical results.

Regulation of Oct-4 Transcriptional Activity by Thioredoxin In Vivo Since the DNA binding capability of oxidized Oct-4 was restored by Thioredoxin, we postulated that increasing Thioredoxin levels would lead to increased transcriptional activity of Oct-4. Oct-4 transfection activated transcription from the 6PORE-luciferase reporter by 2.9-fold over baseline 6PORE activity (Fig. 4). Cotransfecting wt Thioredoxin increased Oct-4 activation of 6PORE by 4.6-fold over baseline, a 1.6-fold increase over Oct-4 alone. Cotransfecting a mutant Thioredoxin (termed dm), without reductase activity, decreased Oct-4 activation of 6PORE to 2.2-fold above baseline, a decrease of 1.3-fold from Oct-4 alone. Thus, increasing wt Thioredoxin levels increased Oct-4 transcriptional activity, but expressing a mutant Thioredoxin reduced Oct-4 transcription. DISCUSSION Thioredoxin (but not Ape-1) was able to restore the DNAbinding activity of Oct-4 by physically interacting with the oxidized cysteines of Oct-4 dimers, the transcriptionally active form of Oct-4. This is significant because there is evidence that

Redox Regulation of Factor Oct-4 by Thioredoxin

263

*

5 4

RLU

3

*

2 1 6PORE-Luc Oct-4 TRX wt TRX dm

1 0 0 0

1 2 0 0

1 2 2 0

1 2 0 2

Figure 4. Thioredoxin overexpression increases the activity of Oct-4 transcriptional activity in vivo. When an expression vector containing Thioredoxin (TRX wt) is cotransfected with Oct-4 and the 6PORE-luciferase reporter, the activity of Oct-4 to activate this reporter is increased. When a dominant negative mutant of Thioredoxin (TRX dm) is overexpressed, Oct-4 transcriptional activation ability decreased. RLU (relative luciferase units), normalized to transfection efficiency. Numbers represent µg of vector per million cells. Asterisks indicate statistically significant differences when compared with Oct-4 alone. Average of three experiments.

monomeric Oct-4 has transcriptional corepressor activity [18]. Thus, the interaction of Thioredoxin with only the dimeric transcriptional activator form of Oct-4 indicates that the co-repressor activity of Oct-4 is not regulated by Thioredoxin. In addition, cotransfection of Thioredoxin significantly increased the transcriptional activity of Oct4, while cotransfection of a mutant Thioredoxin, without reductase activity, decreased Oct-4 transcriptional activation. The competition of endogenous native Thioredoxin by the mutant Thioredoxin produced a decrease in Oct-4 activity. This implies that endogenous Thioredoxin is necessary for full Oct-4 activity and is consistent with the Thioredoxin knockout mouse [20]. Homozygous deletion of Thioredoxin results in death of the embryo immediately after implantation, before gastrulation is complete. When Thioredoxin-deleted preimplantation blastocysts are placed in culture, the inner cell mass, where Oct-4 is preferentially expressed, fails to proliferate [20]. This is consistent with the hypothesis that increased levels of Thioredoxin protect the survival of implanting murine embryos [21]. Without Thioredoxin present, even small amounts of oxygen will reduce Oct-4 activity, preventing the inner cell mass of the blastocyst from forming proper lineages. Thus, one mechanism by which Thioredoxin may promote proper early embryonic development is by maintaining appropriate Oct-4 activity.

Besides Thioredoxin, several other antioxidants, such as catalase and superoxide dismutase, also appear to enhance normal early embryonic development [3-6]. However, there is a paucity of published evidence that these antioxidants directly regulate embryonic transcription factors by reducing cysteines in the DNA-binding domain rather than simply by absorbing oxidative intermediates generally. The best evidence is that Thioredoxin and Ape-1/Ref-1 are the major regulators of transcription factor activity by directly reducing cysteines in the DNA-binding domain [7-12]. Thus, there may be many general reducing proteins that can serve as oxygen sinks and indirectly assist in maintaining proper redox status of transcription factors. However, there may be only a few specific enzymes that directly reduce specific DNA-binding domain cysteines to maintain proper DNA binding. The data presented here, showing that Thioredoxin (but not Ape-1) restores Oct-4 DNA binding, indicate that there may be some specificity to this process and that individual redox regulators have specific targets. Conversely, the finding here that oxidized FoxD3, another embryonic transcription factor, requires significantly more Thioredoxin to restore DNA binding indicates that there also may be specificity of Thioredoxin targets. In these experiments, Thioredoxin has more activity for one transcription factor as opposed to another. Thus, the intracellular redox state may produce differential effects on distinct transcription factors, which have specific interacting partners required to restore their activities. While it is always difficult to extrapolate findings in an in vitro system such as used here to in vivo physiology, an interesting model can be postulated from these data. Oct-4 is preferentially expressed in the inner cell mass of the blastocyst, which is destined to form the embryo [13-16]. This inner cell mass would have a lower oxygen tension than the surrounding cells, thereby increasing Oct-4 DNAbinding activity [1-6]. Furthermore, the blastocyst is also the location of Thioredoxin requirements during gastrulation [20, 21], and thus the redox state of the individual cells within the blastocyst could define which cells become embryonic tissue versus extraembryonic trophoectoderm. The exposure to increased oxygen tension upon uterine implantation and vascularization could downregulate the inner cell mass of Oct-4 DNA-binding activity, allowing other less-sensitive transcription factors to specify gastrulation and subsequent lineage differentiation. Therefore, changing redox states could also trigger embryonic lineage decisions, such as the induction of endoderm by FoxA1 and FoxA2 by decreasing Oct-4 activity [18]. Thus, the environmental oxygen levels could play important and yet unexplored roles in regulating lineage differentiation by the blastocyst.

Guo, Einhorn, Kelley et al.

264

R EFERENCES 1 Thompson JG, Simpson AC, Pugh PA et al. Effect of oxygen concentration on in-vitro development of preimplantation sheep and cattle embryos. J Reprod Fertil 1990;89:573-578. 2 Li J, Foote RH. Culture of rabbit zygotes into blastocysts in protein-free medium with one to twenty per cent oxygen. J Reprod Fertil 1993;98:163-167. 3 Dumoulin JC, Meijers CJ, Bras M et al. Effect of oxygen concentration on human in-vitro fertilization and embryo culture. Hum Reprod 1999;14:465-469. 4 Yuan YQ, Van Soom A, Coopman FO et al. Influence of oxygen tension on apoptosis and hatching in bovine embryos cultured in vitro. Theriogenology 2003;59:1585-1596. 5 Johnson MH, Nasr-Esfahani MH. Radical solutions and cultural problems: could free oxygen radicals be responsible for the impaired development of preimplantation mammalian embryos in vitro? Bioessays 1994;16:31-38. 6 Guerin P, El Mouatassim S, Menezo Y. Oxidative stress and protection against reactive oxygen species in the pre-implantation embryo and its surroundings. Hum Reprod Update 2001;7:175-189.

11 Evans AR, Limp-Foster M, Kelley MR. Going APE over ref-1. Mutat Res 2000:461:83-108. 12 Masutani H, Yodoi J. Thioredoxin. Overview. Methods Enzymol 2002;347:279-286. 13 Pesce M, Scholer HR. Oct-4: gatekeeper in the beginnings of mammalian development. STEM CELLS 2001;19:271278. 14 Kirchhof N, Carnwath JW, Lemme E et al. Expression pattern of Oct-4 in preimplantation embryos of different species. Biol Reprod 2000;636:1698-1705. 15 Pesce M, Scholer HR. Oct-4: control of totipotency and germline determination. Mol Reprod Dev 2000;55:452-457. 16 Nichols J, Zevnik B, Anastassiadis K et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 1998; 95:379-391. 17 Tomilin A, Remenyi A, Lins K et al. Synergism with the coactivator OBF-1 (OCA-B, BOB-1) is mediated by a specific POU dimer configuration. Cell 2000;103:853-864.

7 Hirota K, Matsui M, Iwata S et al. AP-1 transcriptional activity is regulated by a direct association between thioredoxin and Ref-1. Proc Natl Acad Sci USA 1997:94:3633-3638.

18 Guo Y, Costa R, Ramsey H et al. The embryonic stem cell transcription factors Oct-4 and FoxD3 interact to regulate endodermal-specific promoter expression. Proc Natl Acad Sci USA 2002;99:3663-3667.

8 Hirota K, Murata M, Sachi Y et al. Distinct roles of thioredoxin in the cytoplasm and in the nucleus: a two-step mechanism of redox regulation of transcription factor NF-κΒ. J Biol Chem 1999:274:27891-27897.

19 Botquin V, Hess H, Fuhrmann G et al. New POU dimer configuration mediates antagonistic control of an osteopontin preimplantation enhancer by Oct-4 and Sox-2. Genes Dev 1998;12:2073-2090.

9 Ueno M, Masutani H, Arai R et al. Thioredoxin-dependent redox regulation of p53-mediated p21 activation. J Biol Chem 1999;274:35809-35815.

20 Matsui M, Oshima M, Oshima H et al. Early embryonic lethality caused by targeted disruption of the mouse thioredoxin gene. Dev Biol 1996;178:179-185.

10 Akamatsu Y, Ohno T, Hirota K et al. Redox regulation of the DNA binding activity in transcription factor PEBP2: the roles of two conserved cysteine residues. J Biol Chem 1997;272:1449714500.

21 Kobayashi-Miura M, Nakamura H, Yodoi J et al. Thioredoxin, an anti-oxidant protein, protects mouse embryos from oxidative stress-induced developmental anomalies. Free Radic Res 2002;36:949-956.