Báo cáo khoa học: Akt-dependent phosphorylation negatively regulates the transcriptional activity of dHAND by inhibiting the DNA binding activity pdf

Akt-dependent phosphorylation negatively regulates the transcriptional activity of dHAND by inhibiting the DNA binding activity Masao Murakami 1,2 , Keiichiro Kataoka 1 , Shigetomo Fukuhara 1 , Osamu Nakagawa 2 and Hiroki Kurihara 1,3 1 Division of Integrative Cell Biology, Department of Embryogenesis, Institute of Molecular Embryology and Genetics, Kumamoto University, Japan; 2 Department of Molecular Biology, The University of Texas Southwestern Medical Center at Dallas, TX, USA; 3 Department of Physiological Chemistry and Metabolism, The University of Tokyo Graduate School of Medicine, Japan HAND2/dHAND is a basic helix-loop-helix transcription factor expressed in the heart and neural crest derivatives during embryogenesis. Although dHAND is essential for branchial arch, cardiovascular and limb development, its target genes have not been identified. The regulatory mech- anisms of dHAND function also remain relatively unknown. Here we report that Akt/PKB, a serine/threonine protein kinase involved in cell survival, growth and differ- entiation, phosphorylates dHAND and inhibits dHAND- mediated transcription. AU5-dHAND expressed in 293T cells became phosphorylated, possibly at its Akt phos- phorylation motif, in the absence of kinase inhibitors, whereas the phosphatidylinositol 3-kinase inhibitor wort- mannin and the Akt inhibitor NL-71-101, but not the p 70 S6 kinase inhibitor rapamycin, significantly reduced dHAND phosphorylation. Coexpression of HA-Akt augmented dHAND phosphorylation at multiple serine and threonine residues mainly located in the bHLH domain and, as a result, decreased the transcriptional activity of dHAND. Consistently, alanine mutation mimicking t he nonphos- phorylation state abolished the inhibitory effect of A kt on dHAND, whereas aspartate mutation mimicking the phos- phorylation state resulted in a loss of dHAND transcrip- tional activity. These changes in dHAND transcriptional activity were in parallel with changes in the DNA binding activity rather than in dimerization activity. These results suggest that Akt-mediated signaling may regulate dHAND transcriptional activity through the modulation of its DNA binding activity during embryogenesis. Keywords: Akt; bHLH; dHAND; phosphorylation; tran- scription. Basic helix-loop-helix (bHLH) proteins are a highly con- served superfamily of transcriptional factors that commit to various developmental processes, such as cell f ate determin- ation, differentiation and tissue-specific gene expression [1,2]. bHLH proteins form homo- or heterodimers to allow the basic region to bind to the palindromic DNA sequence, CANNTG, termed the E-box [3–5]. In general, tissue- specific bHLH proteins heterodimerize with broadly expressed bHLH proteins, named E-proteins [2]. The HAND proteins, HAND1/eHAND and HAND2/ dHAND are expressed in the heart, neural crest-derivatives and extraembryonic tissues during embryogenesis [6–8]. Gene targeting experiments have revealed that the develop- mental roles of the two HAND proteins are essential and distinct. dHAND-null mice die at embryonic day 9.5 due to defects in cardiovascular development [9,10], whereas eHAND-null mice have lethal defects in early extraembry- onic t issues and cardiovascular development [11,12]. Despite evidence for the importance of HAND proteins during embryogenesis, the regulatory mechanisms of tran- scriptional activity of HANDs have not been analyzed in detail [13], especially in terms of post-translational modifi- cations. Akt is a serine/threonine protein kinase activated down- stream of phosphatidylinositol 3-kinase (PI3K) [14,15], and plays central roles in cell survival, growth and differenti- ation. From detailed biochemical analyses, the consensus sequence of t he phosphorylation t arget for Akt w as identified as RXRXXS/T [16]. Glycogen synthase kinase- 3, BAD, and caspase-9 are known as Akt s ubstrates [1 7–21]. Insulin signaling upstream of Akt was previously shown to stimulate global protein synthesis i n the heart [22,23]. It has been reported that heart size is increased in transgenic mice that express a constitutively active form of Akt from a cardiac specific promoter [23]. In some cases Akt translo- cated into nucleus after stimulation [24,25]. Although it has been reported that Akt can phosphorylate transcription factors such as FKHR [26– 28], CREB [29], and Nur77 [30,31], the physiological importance of Akt phosphoryla- tion in the transcriptional control has not been fully established. In this study, we investigated regulatory mechanisms of dHAND t ranscriptional activity by Akt-mediated s ignaling. dHAND is phosphorylated by Akt at serine/threonine Correspondence to H. Kurihara, D epartment of Physiological Chem- istry and Metabolism, The University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Fax: + 81 3 5684 4958, Tel.: + 81 3 5841 3495, E-mail: kuri-tky@umin.ac.jp Abbreviations: bHLH, basic helix-loop-helix; PI 3K, phosphatidyl- inositol 3-kinase; GST, glutathione S-transferase; MBP, maltose binding protein. (Received 29 January 2004, revised 14 June 2004, accepted 23 June 2004) Eur. J. Biochem. 271, 3330–3339 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04267.x residues mainly w ithin t he bHLH domain. Although phosphorylation of dHAND did not affect its dimerization with E-proteins, the phosphorylation decreased the tran- scriptional activity due to inhibition of its DNA binding. This fact may present a novel model for the regulatory mechanism of bHLH transcription factors. Materials and methods Plasmid construction pCEFL-HA-Akt, that encodes HA-tagged Akt [32], was a gift from J. S. Gutkind ( National I nstitutes o f H ealth, USA). Mouse dHAND, eHAND, E47, and ME2/ITF2a cDNAs were cloned by PCR using the mouse E 10.5 cDNA library (Gibco BRL) as a template. Deletion and point mutants o f dHAND were made by PCR using Quick- Change TM XL Site-Directed Mutagenesis Kit (Stratagene). They were inserted into pGEX (Amersham Pharmacia Biotech) or pMAL (NEB) for g lutathione S-transferase (GST)- or maltose binding protein (MBP)-fusion proteins, respectively. These cDNAs were also cloned into a mam- malian e xpression vector, pCEFL-AU5 [32], w hich contains the AU5-tag at the N-terminus, or pCMV-DBD (Strata- gene) f or fusion proteins with a Gal4 DNA binding domain. Cell culture and preparation of cell lysates 293T cells were cultured in DMEM containing 10% (v/v) fetal bovine serum, 100 UÆmL )1 penicillin G and 100 lgÆmL )1 streptomycin at 37 °Cin5%(v/v)CO 2 . T he C2C12 myoblast cell line w as cultured in DMEM containing 20% fetal bovine serum, 100 UÆmL )1 penicillin G and 100 lgÆmL )1 streptomycin at 37 °Cin5%(v/v)CO 2 . Wortmannin ( a PI3K inhibitor), NL-71-101 (an Akt inhibitor), and rapamycin (a p70 S6 kinase inhibitor) were purchased from Calbiochem. Cells we re lysed in NaCl/P i containing 1% (v/v) Triton X -100 and protease i nhibitors, and whole cell lysates were prepared for immunoprecipitation or pull-down assay. For the immunocomplex kinase assay, cells were lysed in kinase-lysis buffer [ 50 m M Tris/HCl (pH 7.5), 137 m M NaCl, 1 m M EDTA, 1 m M Na 3 VO 4 ,20m M b-glycerophosphate, 5 0 m M NaF, 1% (v/v) Triton X-100, and 10% (v/v) glycerol] containing protease inhibitors. Preparation of GST- or MBP-fusion proteins GST- or MBP-fusion proteins were expressed in Escherichia coli DH5a and were induced by adding 0.4 m M isopropyl thio-b- D -galactoside. The cells were suspended in lysis buffer [NaCl/P i containing 1% (v/v) Triton X-100 and 1 m M phenylmethanesulfonyl fluoride]. After sonication, crude cell l ysates were centrifuged at 4 °C and the supernatant was used for the experiments. GST-fusion and MBP-fusion protein was purified using GSH-Sepharose beads and Amirose resin, and eluted by 20 m M glutathione or 30 m M maltose, respectively. Immunocomplex kinase assay Cell lysates were prepared from 293T cells transfected with pCEFL-HA-Akt. Twenty-four hours after transfec- tion, the c ells were starved for 24 h a nd stimulated by adding fetal bovine serum at a final concentration of 20% (v/v) for 40 min. HA-Akt was immunoprecipitated using a monoclonal antibody raised against HA, and HA-Akt bound protein-G conjugated sepharose beads were washed by kinase reaction buffer [ 10 m M Tris/HCl (pH 7.5 ), 25 m M MgCl 2 ,10m M b-glycerophosphate, 0.33 m M Na 3 VO 4 , and 0.33 m M dithiothreitol not con- taining ATP]. The substrate, GST- or MBP-dHAND and its mutants, and 5 lCi [ 32 P]ATP[cP] were added t o the beads. The kinase r eaction was performed a t 37 °Cfor 30 min and the reaction was stopped by adding 2.5· sam- ple buffer followed by boiling for 5 min. These samples were applied on 10% (w/v) SDS/PAGE followed by autoradiography. Pull-down assay GST- or MBP-fusion proteins bound to the beads were incubated at 4 °C for 6 h with cell lysates prepared from 293T cells expressing HA-Akt. The beads were w ashed three times with NaCl/P i containing 1% (v/v) Triton X-100 and protease inhibitors, and bound proteins were eluted by adding 2.5· sample buffer followed by boiling for 5 min. These samples were applied on 10% (v/v) SDS/PAGE a nd proteins were detected by Western blotting. Immunoprecipitation, Western blotting, and antibodies The procedure of immunoprecipitation and Western blot- ting was described previously [33]. Cell lysates were incubated with protein G sepharose beads and the mono- clonal antibody (anti-HA or anti-AU5) at 4 °C for 6 h. The beads were washed with NaCl/P i containing 1% (v/v) Triton X-100 and protease inhibitors three times an d bound proteins were eluted by adding 2.5· sample buffer followed by boiling for 5 min. Antibodies were purchased from Covance (anti-HA and anti-AU5), Zymed (anti-phospho- serine and anti-phosphothreonine), and Cell Signaling [anti-(phospho-Akt substrate) (RXRXXS/T)]. Gel shift analysis Double strand o ligonucleotide probes w ere end-labeled with [ 32 P]ATP[cP] by T4 polynucleotide kinase. The binding sequence o f eHAND/E47 heterodimer, reported previously [34], was used as the probe. The sequence of each probe was as shown below. Underlining indicates the E-box sequence: CANNTG. Probe E¢ (E¢); 5¢-TTTGCAAGGGG CATCTGGCATTCGCCC-3¢, and mutant probe E¢ (m E¢); 5¢-TTTGCAAGGGG CGAACAGCATTCGCCC-3¢.Two microliters of cell l ysate prepared from 293T c ells was incubated with 1 lg poly ( dI-dC) and labeled probe at room temperature for 30 min. These samples were applied on 4% (w/v) native PAGE followed by autoradiography. Transfection and reporter assay Transfection into 293T or C 2C12 cells was carried out using Lipofectamine Plus reagent (Life Technologies). G al4-DBD or Gal4-DBD constructs, pFR-Luc (Stratagene), pRL- SV40 (Promega), and pCEFL-HA-Akt were transfected Ó FEBS 2004 dHAND phosphorylation by Akt (Eur. J. Biochem. 271) 3331 and firefly luciferase units were determined 48 h after transfection. Transfection efficiency was normalized on the basis of r enila luciferase activity. All these a ssays were performed in triplicate. Results and Discussion Phosphorylation of dHAND by Akt Although g ene targeting analyses in mice demonstrated the importance of HAND proteins during development, the regulatory mechanisms of HAND transcriptional activity are not well understood. In this study, we examined the possibility that Akt might regulate dHAND activity, because d HAND has a consensus motif for A kt phos- phorylation (RXRXXS) at amino acids 108–114. First, we analyzed the phosphorylation status of dHAND in 293T cells in the absence or presence of several kinase inhibi- tors. AU5-dHAND was immunoprecipitated by anti- AU5 Ig and the phosphorylation status of dHAND was analyzed b y Western b lotting using antibody raised against the phospho-Akt substrate. As shown in Fig. 1A, AU5- dHAND expressed in 293T cells became phosphorylated possibly at i ts Akt phosphorylation motif in the a bsence of kinase inhibitors. This signal was decreased by treatment with 200 n M wortmannin, a PI3K inhibitor (Fig. 1A). Fifty micromoles of NL-71-101, an Akt inhibitor, also signifi- cantly reduced dHAND phosphorylation (Fig. 1A). In contrast, 20 n M rapamycin, a p70 S6 kinase inhibitor, did not largely affect the phosphorylation status of dHAND (Fig. 1A). These results suggest that dHAND is phosphory- lated in a PI3K/Akt-dependent, but p70 S6 kinase-inde- pendent manner. To determine whether dHAND was phosphorylated by Akt in vivo, we transiently expressed AU5-dHAND and HA-tagged Akt in 293T cells. Western blotting with anti- (phospho-Akt substrate) Ig revealed that overexpression of HA-Akt upregulated dHAND phosphorylation (Fig. 1B). Fig. 1. Phosphorylation status of dHAND in 293T cells. (A) Effec t of kinase inhibitors on the b asal phosphorylation of d HAND. 293T cells expressing AU5-dHAND w ere t reated with 200 n M wortmannin, 50 l M NL-71-101 or 20 n M rapamycin fo r 1 20 min. Dimethylsulfoxide w as used as a vehic le. AU5-dHAND was imm unopre cipitated us ing a nti-AU5 Ig and applied on 10% SDS/PAGE. The phos phorylation status of dHAND was a nalyzed by Western blotting using anti-(phospho-Akt substrate) Ig (top). The same memb rane was reblotted by an ti-AU 5 Ig (bottom). (B) Effects of Akt overexpression on dHAND phosphorylation. Cell lysates were prepared from 293T cells expressing AU5-dHAND and HA-Akt, and the phos- phorylation status of d HAND was analyzed by immunoblotting with anti -(p hospho-Ak t substrate) Ig (top). The same membrane was reblotted by anti-AU5 Ig (bottom). ( C) Effects of Akt overexpression on dHAND phosphorylation on serine/threon ine residues. Cell lysates were prepared from 293T cell expressing AU5-dHAND and/or HA-Akt. Phosphorylation status of dHAND was an alyzed b y Western blotting using anti-phosphoserine Ig (top ) or anti-p hosphot hreonine Ig (m iddle). The prot ein amount o f AU5-dHAND was estimated by Western blotting using anti-AU5 Ig (b ottom). Fig. 2. Akt phosphorylated dHAND in vitro. HA-Akt was immuno- precipitated from the cell lysate prepared from 293T cells expressing HA-Akt, and immunocomplex kinase assay was performed in the presence of [ 32 P]ATP[cP] usin g GST or G ST-dHAND as a su bstrate. Protein sample w as applied on 1 0% SDS/PAGE and incorporation of 32 P was detected by autoradiography. 3332 M. Murakami et al.(Eur. J. Biochem. 271) Ó FEBS 2004 We also analyzed the phosphorylation status of AU5- dHAND by immunoprecipitation and Western blotting methods using antibodies specific to phosphoserine or phosphothreonine. In the cells expressing AU5-dHAND alone, the anti-AU5 immunoprecipitants gave signals for both anti-phosphoserine and anti-p hosphothreonine immu- noreactivity, suggesting that dHAND may be basally phosphorylated at serine and threonine residues in the cells (Fig. 1C). These results indicate that the Akt signal m ay pos- itively modulate the phosphorylation status of dHAND. We further examined whether Akt directly phosphory- lates dHAND using immunocomplex kinase assay. HA- Akt was transiently expressed in 293T cells and cells were starved a nd stimulated by 20% (v/v) fetal bovine serum for 40 min to activate HA-Akt. HA-Akt was immunoprecip- itated using anti-HA Ig and the kinase reaction was started by adding GST-dHAND to the immunoprecipitants in the reaction buffer containing [ 32 P]ATP[cP]. As shown in Fig. 2, strong incorporation of 32 P into GST-dHAND was detected when incubated with HA-Akt immunoprecip- itates. Autophosphorylation of Akt was also detected as described previously [35]. In contrast, GST was not phosphorylated at all, and immunoprecipitates from the mock transfected cell lysates did not phosphorylate GST- dHAND, indicating that Akt specifically phosphorylates dHAND. To locate the dHAND phosphorylation sites by Akt, we constructed MBP-dHAND mutants in w hich the C-terminal region was serially deleted (dHAND-(1–102),-(1–132), and -(1–196) (Fig. 3A). Comparable phosphorylation was detected in dHAND-(1–196) and dHAND-(1–132), in which the bHLH domain was preserved and partially deleted, respectively (Fig. 4A). In contrast, no significant phosphory- lation was d etected in dHAND(1–102) (Fig. 4A). This result indicates that the major phosphorylation sites reside in the region of amino acids 103–132 of dHAND. The region of amino acids 103–132 contains several serine and threonine residues and Ser114 matches the consensus motif for Akt phosphorylation ( RXRXXS/T) (Fig. 3B). T hus, t o i dentify the phosphorylation site of dHAND, we first substituted Ser114 with alanine and examined whether this mutant (named dHAND-3SA; Fig. 3A) c an be phosphorylated by Akt. Unexpectedly, the phosphorylation signal was comparable with that of Fig. 3. Structures and putative phosphorylation sites of dHAND mutants. (A) Schematic representation of the structure of MBP- or GST-fused dHAND mutants. Serine (S) and threonine (T) residues that were substituted with alanine (A) or aspart ic acid (D) are indicated by circles. Hatched box, black boxes, and gray box indicate basic region, helix motif, and loop structure, respectively. (B) Amino acid sequence of mouse dHAND. The posi- tion of the bHLH domain is indicated. The mutated sites in this study are shown by asterisks (Ser114, Thr103, Thr112, Thr140, Thr145, and Thr204). Ó FEBS 2004 dHAND phosphorylation by Akt (Eur. J. Biochem. 271) 3333 wild type dHAND (dHAND-WT) (Fig. 4B, lanes 1 and 4), leading us to speculate that additional serine/threonine residues are pho sphorylated by A kt. There are n o sites other than Ser114 that match completely to the canonical consensus motif (Fig. 3B). However, five additional threonine residues in the C-terminal half of dHAND fit the R/KXXS/T motif, which is similar to the consensus of Akt phosphorylation (Fig. 3). Therefore we substituted Ser114 and all of the five threonines with alanine or aspartic acid (dHAND-Ala and dHAND-Asp, respect- ively; Fig. 3A) and examined whether these mutants can be phosphorylated by Akt. No phosphorylation was detected in these mutants, suggesting that these six residues contain the targets for Akt phosphorylation (Fig. 4B). Then we substituted each of the six serine/ threonine residues one by one with alanine (dHAND- 1TA, 2TA, 3SA, 4TA, 5TA and 6TA; Fig. 3A). Among these mutants, dHAND-1TA demonstrated partial but substantial decrease in phosphorylation (Fig. 4B). In addition, further d ecreases in phosphorylation were detected in mutants dHAND-Ala-X and dHAND-Ala- XZ (Fig. 4B). These results indicated that the major phosphorylation sites of dHAND were located within three sites (Thr102, Thr112, and Ser114) mutated in dHAND-Ala-X. Akt binds to dHAND in vitro and in vivo To examine whether Ak t c an dire ctly inte ract with dHAND, we performed pull-down assay using GST- Fig. 4. In vitro kinase assay using MBP-dHAND mutants. Imm uno- complex kinase a ssay was performed using dHAND deletion mutants (A) or dHAND point mutants (B) as described in Fig. 2. Each protein used as a substrate was subjected to 10% SDS/PAGE and b ands were detected by stain ing with Coomassie Brilliant B lue (bottom). Arro w- heads indicate the positions of dHAND mutants. Fig. 5. Analysis o f the interaction between d HAND and Akt. (A) GST- pull down assay was pe rformed using ce ll extract of 293T cells expressing HA-Akt (top). GST or GST-dHANDs used in this assay wereappliedon10%SDS/PAGEandproteinwasdetectedbystaining with Coomassie Brilliant Blue (bottom). Arrowheads indicate the positions of dHAND mutants. (B) Coimmunoprecipitation assay of dHAND and Akt. AU5-dHAND was immunoprecipitated from the cell lysate of 293T c ells e xpressing AU5-dHAND and HA-Akt. These samples were applied on 10% SDS/PAGE and bound proteins were detected by Western blotting using anti-HA Ig. 3334 M. Murakami et al.(Eur. J. Biochem. 271) Ó FEBS 2004 dHAND and cell extracts of 293T cells expressing HA-Akt. As shown in Fig. 5A, HA-Akt interacted with GST- dHAND but not with GST. To identify the domain of dHAND interacting with Akt, we performed pull-down assay using dHAND deletion mutants. As shown in Fig. 5A, HA-Akt was also found to interact with GST- dHAND-(1–132) and -(1–196). In contrast, Akt did not bind to GST-dHAND-(1–102) efficiently, indicating that dHAND may bind to Akt via its bHLH domain. To detect the interaction between dHAND and Akt in viv o, we coexpressed AU5-dHAND and HA-Akt in 293T cells and performed a coimmunoprecipitation experiment. As shown i n Fig. 5B, interaction b etween HA-Akt and AU5-dHAND was detected. This result indicated that A kt could bind to dHAND in the cells. Transcriptional activity of dHAND was decreased by Akt phosphorylation Interestingly, major phosphorylation sites of dHAND are located within the bHLH domain. Therefore, we speculated that phosphorylation of these sites might affect the DNA binding and/or dimerization activities of dHAND. We first analyzed the effect of phosphorylation on the transcriptional activity of dHAND using reporter assay (Fig. 6A). AU5- dHAND, -dHAND-Ala-X, -dHAND-Ala, and -E47 were cotransfected into 293T cells together with reporter plasmid driven by three E-box sequences. Although AU5-dHAND, -dHAND-Ala-X, -dHAND-Ala, and E47 alone showed relatively low transcriptional activity, coexpression of E47 significantly enhanced the activity of dHAND-WT, -Ala-X, or -Ala (Fig. 6A a nd data not shown). I t w as of interest that coexpression of Akt reduced the transcriptional activity of dHAND-WT/E47. The activity of dHAND-Ala-X/E47 was p artially reduced by coexpression of Akt. But this inhibitory effect of Akt was not detected in the case of dHAND-Ala/E47 (Fig. 6A). Luciferase assay was performed using dHAND mutants to estimate the effect of phosphorylation. As shown in Fig. 6B, dHAND-WT, dHAND-Ala-X, and dHAND- Ala showed strong transcriptional activity in the presence of E47. But dHAND-Asp-X, that mimics a phosphoryla- tion form, showed weak transcriptional activity (Fig. 6B). The expression level of dHAND-Asp-X was lower than those of others in the absence of E47. But in the presence of E47, the expression level of this mutant was equivalent to wild type dHAND or other mutants (Fig. 6B, lower panel). This indicated the transcriptional activity of dHAND- Asp-X/E47 was much lesser than those of wild type. These results indicated that phosphorylation of dHAND decreased the transcriptional a ctivity of dHAND/E47 heterodimer. Fig. 6. The effects of A kt expression on the transcriptional activity of dHAND. (A) AU5-d HAND, -dHAND-Ala-X, -dHAND-Ala, and -E 47 were cotransfected into 293T cells with reporter plasmid encoding luciferase g ene, whose expression is d riven by three E-box sequences ( CATCTG). The effect of Akt expression was monitored by luciferase assay. The expression levels of dHAND and its mutants were an alyzed b y West ern blotting using anti-AU5 Ig (bott om). Note t hat dHAND transcriptional activity requires the presence of E47. (B) Transcriptional activity of dHAND mutants was analyzed by luciferase assay using 293T cells. The expression levels of dH AND, its mutants, and E47 were analyzed by Western blotting using anti-AU5 Ig (bottom). Ó FEBS 2004 dHAND phosphorylation by Akt (Eur. J. Biochem. 271) 3335 Phosphorylation of dHAND did not affect the dimerization activity with E47 To clarify the mechanisms of the inhibition of dHAND activity by Akt phosphorylation, we examined whether dimerization activity of dHAND with E47 could be affected by Akt phosphorylation or not. We selected the Gal4-fusion system because we can estimate the dimeri- zation activity with E47 and the transcriptional activity of the dHAND/E47 h eterodimer independently of their DNA binding activity (Fig. 7, upper p anel). As shown in Fig. 7, coexpression of E47 enhanced the transcrip- tional activity of Gal4-dHAND-WT, indicating that E47 bound to Gal4-dHAND and functioned as a strong transcriptional activator. Gal4-dHAND-Ala-X and Gal4- dHAND-Ala also showed similar results. The transcrip- tional activity of dHAND-Asp-X was also enhanced by coexpression of E47. This result suggested that phosphorylated dHAND could dimerize with E47 and function as a transcriptional activator complex in a Gal4- fusion system. DNA binding activity of dHAND was decreased by Akt phosphorylation Because we did not d etect the effects of Akt phosphoryla- tion on dimerization activity of dHAND, we next examine whether the DNA binding activity of dHAND could be affected by Akt phosphorylation. The influence of phos- phorylation by Akt on the DNA binding activity was examined by gel shift analysis. We included the E-protein Fig. 7. Transcriptional activity of dHAND/E47 was analyzed using Gal4-fusion system. Gal4 DNA bindin g domain (DBD)-fused dHAND and reporter plasmid co ntaining t he luc iferase ge ne, whose expression was driven by five G al4 binding sites, were transfecte d t o C2C12 cells, a nd luciferase assay w as performed. Fold activation is the ratio of the luciferase activity in cells transf ected with E 47 construct to cells transfected with empty vector. Fig. 8. Gel shift analysis of dHAND. (A) DNA binding activity of dHAND/ME2 or dHAND-Ala/ME2 to probe E w as compared after in vitro phosphorylation procedure in t he presence (+) or absence (–) o f Akt. Protein samples of A U5-dH AND or AU5-ME2 were mixed with 32 P-labeled oligonucleotide probe containing E-box seq uence, probe E. After incubation at room temperature for 30 min, samples were applied on 4% native PAGE followed by autoradiography. (B) Protein amounts of dHAND and dHAND-Ala used in this experiment were estimated by Western blotting using anti-AU5 Ig. (C) DNA binding activities of the heterodimers between dHAND mutants and ME2 were compared to that of wild type dHAND using probe E. Protein amounts of dHAND, dHAND- Asp-X, and dHAND-Asp used in this experiment were estimated by Western blotting using anti-AU5 Ig (bottom). 3336 M. Murakami et al.(Eur. J. Biochem. 271) Ó FEBS 2004 ME2/ITF2a tagged with AU5 as a heterodimeric partner in this experiment. We phosphorylated dHAND b y Akt in vitro,mixedwith ME2, and performed gel shift analysis. As shown in Fig. 8A, DNA binding activity of wild type dHAND/ ME2 was reduced after phosphorylation by Akt. In contrast, dHAND-Ala showed a shifted band with higher intensity, and this was not affected by the phosphorylation procedure (Fig. 8A,B). The higher DNA binding activity o f dHAND-Ala may be due to escaping endogenous phos- phorylation. To investigate the effect of phosphorylation on DNA binding activity, we performed the same assay using phosphorylation mimicking mutants. As shown in Fig. 8C, dHAND-Asp-X and -Asp almost lost their DNA binding activity. These results indicated that the phosphorylation of Thr102, Thr112, and Ser114 inhibited DNA binding activity of dHAND. In this study, we showed that Akt could bind t o d HAND and phosphorylate i t at s erine and threonine residues within the bHLH domain. Phosphorylation by Akt inhibited the transcriptional activity of dHAND. We analyzed the mechanisms of this inhibitory effect of Akt-dependent phosphorylation on the transcriptional activity of dHAND focused o n t hese th re e a spects: ( a) dimeriza t ion a ctivity w ith E-protein, (b) DNA binding activity of dHAND/E-protein, and (c) transcriptional a ctivity of dHAND/E-protein heterodimer. Our results indicated that the effect of Akt- dependent p hosphorylation acted on (b), but not (a) and (c). After phosphorylation by Akt, dHAND/E-protein hetero- dimer can be present, but this heterodimer can not bind to DNA, leading to a decrease in the transcriptional activity (Fig. 9) . Functional modulation by phosphorylation has been reported in some bHLH proteins. For example, the amino acid sequences of the various myogenic regulatory factors contain co nsensus pho sphorylation motifs for cAMP- dependent protein kinase, protein kinase C and casein kinase II [36–42]. The activation of cAMP-dependent protein kinase or protein kinase C can inhibit muscle differentiation, whereas casein kinase II stimulates myogen- esis by increasing the transcriptional activities of MRF4 and MyoD [43]. In these cases, however, phosphorylation or mutagenesis of each site does not affect the transcriptional activity with an exception that phosphorylation of the protein kinase C site of myogenin blocks its DNA binding activity [40,43], suggesting that the effects of protein kinases are likely to b e indirect. Our present study is unique in that the phosphorylation sites exist w ithin the bHLH domain a nd their phosphorylation can directly affect the function as a transcriptional factor. Although p70 S6 kinase can phos- phorylate the R/KXXS/T motif, rapamycin did not inhibit dHAND phosphorylation. This result and t he evidence that Akt can bind to dHAND and phosphorylate it strongly suggest that Akt directly regulates dHAND activity. Akt signaling is known to be involved in not only many developmental processes but also homeostasis in adults [44– 47]. It has b een reported that h eart size is increased i n transgenic mice that express a constitutively active form of Akt [23], and that down regulation o f HAND gene expression is observed in rodent hypertrophy and human cardiomyopathy [48]. The modulation of dHAND tran- scriptional activity by Akt phosphorylation may explain a part of this involvement b ecause there is the report that Akt is expressed in developing heart [49]. Interestingly, the sequence of phosphorylation sites in the bHLH domain is well conserved within the Twist-related bHLH subfamily, to which dHAND and eHAND belong [6]. The activity of some of these bHLH family proteins might also be r egulated by Akt signaling in vivo. In conclusion, dHAND was phosphorylated by Akt and the phosphorylation i nhibited the transcriptional activity of dHAND. Although the phosphorylated form of dHAND could dimerize with E-protein, this complex could not activate the transcription due to a loss of the DNA binding activity. The present finding may be an important clue of the regulation mechanisms of HAND proteins. Acknowledgements WethankDrJ.S.Gutkindforplasmids,Ms.S.Okamura,Ms.K. Shin-Fukuhara, Ms. M. Yonemitsu, and Dr M . N akagawa f or technical assistance, Dr H. Saya and Dr T. Hirota for advice and discussion on the phosp horylation analysis, and D r E. N. Olson and Dr H. Iba f or critical comments on this paper. Th is work was supported by JSPS (Japan Society for t he Promotion of Science) Research for the Future Pro gram; Grants-in-Aid for Scientific R esearch from th e Ministry of Education, Science and Culture, Japan; and the Research Grant for Cardiovascular Diseases (14C-1) from t he Ministry of Health, Japan. References 1. Atchley, W. & Fitch, W . (1997) A natural classification of the b asic helix-loop-helix class of transcription factors. Proc. Natl Acad. Sci. USA 94, 5172–5176. 2. Massari, M.E. & Murre, C. (2000) Helix-Loop-Helix Proteins: Regulators of TranscriptioninEucaryoticOrganisms.Mol. Cell. Biol. 20, 429–440. Fig. 9. A model of the regulation mechanisms of dHAND activity. Phosphorylation of dHAND by Akt does not affect on the dimeriza- tion with E-proteins. Phosphorylated dHAND can dimerize with E-protein (1) and this complex can function as a transcriptional acti- vator if recruited to DNA (3). But, th e heterodimer of phospho rylated dHAND can not bind to the E-box sequence (2). As a result, the transcriptional activity of dHAND is inhibited by Akt signaling. Ó FEBS 2004 dHAND phosphorylation by Akt (Eur. J. Biochem. 271) 3337 3. Ephrussi, A., Church, G.M., Tonegawa, S. & Gilbert, W. (1985) B-lineage-specific interactions of an immunoglobulin enhancer with cellular factors in vivo. Science 227, 134–140. 4. Murre, C., McCa w, P.S. , V aessin, H., Caudy, M., Jan, L.Y., Jan, Y.N., Cabrera, C .V., Buskin, J.N., Hauschka, S .D., Lassar, A.B., Weintraub, H. & Baltimore, D. (1989) Interactions between het- erologous h elix-loop-helix proteins generate co mplexes that bind specifically to a common DNA sequen ce. Cell 58, 537–544. 5. Ellenberger, T., Fass, D., Arnaud, M. & Harrison, S. (1994) Crystal structure of transcription fac tor E47: E-box recognitio n by a basic region helix-loop-helix dimer. Genes Dev. 8, 970–980. 6. Srivastava, D., Cserjesi, P. & Olson, E.N. (1995) A subclass of bHLH proteins required for cardiac morphogenesis. Science 270, 1995–1999. 7. McFadden, D.G., Charite, J., R ichardson, J.A., S rivastava, D., Firulli, A.B. & Olson, E.N. (2000) A GATA-dependent right ventricular enhancer controls dHAND transcription in the developing heart. Developmen t 127, 5331–5341. 8. Yamagishi, H., Olson, E.N. & Srivastava, D. (2000) The basic helix-loop-helix transcription factor, dHAND, is required for vascular development. J. Clin. Invest. 105, 261–270. 9. Srivastava, D., Thomas, T., Lin, Q., Kirby, M.L., Brown, D. & Olson, E.N. (1997) Regulation of cardiac mesodermal and neural crest development by the bHLH transcription facto r, dHAND. Nat. Genet. 16, 154–160. 10. Thomas, T., Kurihara, H., Yamagishi, H., Kurihara, Y ., Yazaki, Y., Olsen, E.N. & Srivastava, D. (1998) A signaling cascade in- volving endothelin-1, dHAND and Msx1 regulates development of neural-crest-derived branchial arch mesenchyme. Developm ent 125, 3005–3014. 11. Firulli, A.B., McFadden, D.G., Lin, Q., Srivastava, D. & Olson, E.N. (1998) Heart and extra-embryonic mesodermal defects in mouse embryos lackin g the bHLH transcription factor Hand1. Nat. Genet. 18, 266–270. 12. Riley, P., Anson-Cartwright, L. & Cross, J.C. (1998) The Hand1 bHLH transcription factor is essential for placentation and car- diac morphogenesis. Nat. Genet. 18, 271–275. 13. Dai, Y.S. & Cserjesi, P. (2002) The basic helix-loop-helix factor, HAND2, functions as a transcriptional activator by binding to E-boxes as a heterodimer. J. Biol. Chem. 277, 12604–12612. 14. Burgering, B.M. & Coffer, P .J. (1995) Protein k inase B (c-Akt) in phosphatidylinositol-3-OH kinase s ignal t ransduction. Nature 376, 599–602. 15. Franke, T.F., Yang, S.I., Chan, T.O., Datta, K., Kazlauskas, A., Morrison, D.K., Kaplan, D.R. & T sichlis, P.N. (1995) The protein kinase encoded by th e Akt proto- oncogene is a target o f the PD GF-activated phosphatidylinositol 3-kinase. Cell 81, 727– 736. 16. Obata, T., Yaffe, M.B., Leparc, G.G., Piro, E.T., Maegawa, H., Kashiwagi, A., Kikkawa, R. & Cantley, L.C. (2000) Peptide and protein library screening defines optimal substrate motifs for AKT/PKB. J. Biol. Chem. 275, 36108–36115. 17. Cross, D.A., Alessi, D.R., Cohen, P., Andjelkovich, M. & Hem- mings, B.A. (1995) Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378, 785–789. 18. Alessi, D.R., Caudwell, F.B., Andjelkovic, M., Hemmings, B.A. & Cohen, P. (1996) Molecular basis for the substrate specificity of protein k inase B; comparison with MAPKAP kinase-1 a nd p70, S6 kinase. FEBS Lett. 399, 333–338. 19. Datta,S.R.,Dudek,H.,Tao,X.,Masters,S.,Fu,H.A.,Gotoh,Y. & Greenberg, M.E. (1997) Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91, 231– 241. 20. delPeso,L.,Gonzalez-Garcia,M.,Page,C.,Herrera,R.&Nunez, G. (1997) Interleukin-3-induced phosphorylation of BA D through the protein kinase Akt. Science 27 8, 687–689. 21. Cardone, M.H., R oy, N ., S ten nicke, H.R ., S alv esen, G .S., F ranke, T.F., Stan bridge, E., Frisch, S. & Reed, J .C. (1998) Regulation of cell death protease caspase-9 by phosphorylation. Science 282, 1318–1321. 22. Shiojima, I., Yefremashvili, M., Luo, Z., Kureishi, Y., Takahashi, A., T ao, J., Rosenzweig, A ., Kahn, C.R., Abel, E.D. & Walsh, K. (2002) Akt signaling mediates postnatal heart growth in response to insulin and nutritional status. J. Biol. Chem. 277, 37670–37677. 23. Shioi, T., McMullen, J.R., Kang, P.M., D ouglas, P.S., Obata, T., Franke, T.F., Cantley, L.C. & Izumo, S. (2002) Akt/protein kinase B promotes organ growth in transgenic mice. Mol. Cell. Biol. 22, 2799–2809. 24. Andjelkovic, M., Alessi, D.R., Meier, R., Ferna ndez, A., Lamb, N.J., F rech, M., Cron, P., Coh en, P., Lucocq, J.M. & Hemmings, B.A. (1997) R ole of translocation in the activation and function of protein kinase B. J. Biol. Che m. 272, 31515–31524. 25. Meier, R., Alessi, D.R., Cron, P., Andjelkovic, M. & Hemmings, B.A. (1997) Mitogenic activation, phosphorylation, and nuclear translocation of protein kinase Bbeta. J. Biol. Chem. 272, 30491– 30497. 26. Brunet, A., Bonni, A., Zigmond,M.J.,Lin,M.Z.,Juo,P.,Hu, L.S., Anderson, M.J., Arden, K.C., Blenis, J. & Greenberg, M.E. (1999) Akt promotes cell survival by phosphorylating and- inhibiting a Forkhead transcription factor. Cell 96, 857–868. 27. Brownawell, A.M., Kops, G.J., Macara, I.G. & B urgering, B.M. (2001) Inhibition of nuclear import by protein kinase B (Akt) regulates the subcellular distribution and activity of the forkhead transcription factor AFX. Mol. Cell. Biol. 21, 3534– 3546. 28. Tang. E.D., Nune z, G., Barr, F.G. & Guan, K .L. (1999) Negative regulation of the forkhead transcription f actor FKHR by Akt. J. Biol. Chem. 274, 16741–16746. 29. Du, K. & Montminy, M. (1998) CREB is a regulatory target for the protein kinase Akt/PKB. J. Biol. Chem. 273, 32377–32379. 30. Masuyama, N., Oishi, K., Mori, Y., Ueno, T., Takahama, Y. & Gotoh, Y. (2001) Akt inhibits the orphan nuclear rec ept or Nu r77 and T-cell apoptosis. J. Biol. Chem. 276, 32799–32805. 31. Pekarsky. Y., Hallas, C., Palamarchuk, A., Koval, A., B ullrich, F., Hirata, Y., Bichi, R., Letofsky, J. & Croce, C.M. (2001) Akt phosphorylates and regulates the o rphan n uclear receptor Nur77. Proc.NatlAcad.Sci.USA98, 3690–3694. 32. Murga, C., Laguinge, L ., Wetzker, R., Cuadrado, A. & Gutkind, J.S. (1998) A ctivation of Akt/protein kinase B by G protein-cou- pled receptors. A role for alpha and beta gamma subunits of heterotrimeric G proteins acting thro ugh ph osphatidylin ositol-3- OH kinasegamma. J. Biol. Chem. 273, 19080–19085. 33. Murakami, M., Sonobe, M.H., Ui, M., Kabuyama, Y., Watanabe, H., Wada, T., Handa, H. & Iba, H. (1997) Phosphorylation and high level expression of Fra-2 in v-src transformed cells: a pathway of activation of endogenous AP-1. Oncogene 14, 2435–2444. 34. Hollenberg, S.M., Sternglanz, R., Cheng, P.F. & Weintraub, H. (1995) Identification of a new family of tissue-specific basic helix- loop-helix proteins with a two-hybrid system. Mol. Cell. Biol. 15, 3813–3382. 35. Ravichandran, L.V., Chen, H., Li, Y. & Quon, M.J. (2001) Phosphorylation of PTP1B at Ser(50) by Akt impairs its ability t o dephosphorylate the insulin receptor. Mol. Endocrinol. 15, 1768– 1780. 36. Winter, B., Braun, T. & Arnold, H.H. (1993) cAMP-dependent protein kinase represses m yogenic differentiation and the ac tivity of the muscle-specific helix-loop-helix transcription facto rs Myf-5 and MyoD. J. Biol. Chem. 268, 9869–9878. 37. Tapscott, S.J., Davis, R.L., Thayer, M.J., Cheng, P.F., Weintraub, H. & Lassar, A.B. (1988) MyoD1: a nuclear phosphoprotein requiring a Myc homology region to convert fibroblasts to myo- blasts. Science 242, 405–411. 3338 M. Murakami et al.(Eur. J. Biochem. 271) Ó FEBS 2004 38. Lassar, A.B., Davis, R.L., Wright, W.E., Kadesch, T., M urre, C., Voronova, A., Baltimore, D. & Weintraub, H. (1991) Functional activity of myogenic H LH proteins requires hetero-oligomeriza- tion with E12/E47-like proteins in vivo. Cell 66, 305–315. 39. Li, L., Heller-Harrison, R., Czech, M. & Olson, E .N. (1992) Cyclic AMP-dependent protein kinase inhibits the activity of m yogenic helix-loop-helix proteins. Mol. Cell. Biol. 12, 4478–4485. 40. Li,L.,Zhou,J.,James,G.,Heller-Harrison,R.,Czech,M.P.& Olson, E.N. (1992) FGF inactivates myogenic helix-loop-helix proteins through phosphorylation of a conserved protein kinase C site in their DNA-binding domains. Cell 71, 1181–1194. 41. Hardy, S., Kong, Y. & Konieczny, S.F. (1993) Growth factor inhibits MRF4 activity independently of the phosphorylation status of a conserved threonine residue within the DNA-binding domain. Mol. Ce ll . Bi ol . 13, 5943–5956. 42. Mitsui, K., Shirakata, M. & Paterson, B.M. (1993) Ph osphory- lation inhibits the DNA-binding activity of MyoD homodimers but not MyoD-E12 heterodimers. J. Biol. Chem. 268, 24415– 24420. 43. Johnson, S.E., Wang, X., Hardy, S., Taparowsky, E.J. & Konieczny, S.F. (1996) Casein k inase II increases the transcrip- tional activities of MRF4 and MyoD independently of t heir direct phosphorylation. Mol. Cell. Biol. 16, 1604–1613. 44. Datta,S.R.,Brunet,A.&Greenberg, M.E. (1999) Cellular sur- vival: a play in three Akts. Genes Dev. 13, 2905–2927. 45. Galetic, I., Andjelkovic, M., Meier, R., Brodbeck, D., Park, J. & Hemmings, B.A. (1999) Me chanism of protein kina se B activation by insulin/insulin-like growth factor-1 revealed by specific inhibi- tors of phosphoin ositide 3-k inase – significance for diab etes and cancer. Pharmacol. Ther. 82, 409–425. 46. Kandel, E .S. & Hay, N. (1999) The regulation and activities of the multifunctional/threonine kinase Akt/PKB . Exp. Cell Res. 253, 210–229. 47. Brazil, D.P. & Hemmings, B.A. (2001) Ten years of protein kinase B signaling: a hard Akt to follow. Trends Biochem. Sci. 26, 657– 664. 48. Thattaliyath, B.D., Livi, C.B., Steinhelper, M.E., Toney, G.M. & Firulli, A.B. (2002) HAND1 and HAND2 are expressed in the adult-rodent heart and a re mo dulated during cardiac hypertrophy. Biochem. Biophys. Res. Commun. 297, 870–875. 49. Altomare, D.A., Lyons, G.E., Mitsuuc hi, Y., Ch eng, J .Q. & T esta, J.R. (1998) Akt2 mRNA is highly expressed in embryonic brown fat and the AKT2 kinase is activated by insulin. Oncogene 16, 2407–2411. Ó FEBS 2004 dHAND phosphorylation by Akt (Eur. J. Biochem. 271) 3339 . Akt-dependent phosphorylation negatively regulates the transcriptional activity of dHAND by inhibiting the DNA binding activity Masao Murakami 1,2 ,. binding activity of dHAND could be affected by Akt phosphorylation. The influence of phos- phorylation by Akt on the DNA binding activity was examined by