JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M610821200 on January 23, 2007

J. Biol. Chem., Vol. 282, Issue 13, 9983-9995, March 30, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
282/13/9983    most recent
M610821200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ishiguro, A.
Right arrow Articles by Aruga, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ishiguro, A.
Right arrow Articles by Aruga, J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

ZIC2-dependent Transcriptional Regulation Is Mediated by DNA-dependent Protein Kinase, Poly(ADP-ribose) Polymerase, and RNA Helicase A*

Akira Ishiguro{ddagger}1, Maki Ideta{ddagger}, Katsuhiko Mikoshiba§, David J. Chen, and Jun Aruga{ddagger}2

From the {ddagger}Laboratory for Comparative Neurogenesis and §Laboratory of Developmental Neurobiology, RIKEN Brain Science Institute, Wako-shi, Saitama 351-0198, Japan and the Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, Texas 75390

Received for publication, November 22, 2006 , and in revised form, January 22, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The Zic family of zinc finger proteins is essential for animal development, as demonstrated by the holoprosencephaly caused by mammalian Zic2 mutation. To determine the molecular mechanism of Zic-mediated developmental control, we characterized two types of high molecular weight complexes, including Zic2. Complex I was composed of DNA-dependent protein kinase catalytic subunit (DNA-PKcs), Ku70/80, and poly(ADP-ribose) polymerase; complex II contained Ku70/80 and RNA helicase A; all the components interacted directly with Zic2 protein. Immunoprecipitation, subnuclear localization, and in vitro phosphorylation analyses revealed that the DNA-PKcs in complex I played an essential role in the assembly of complex II. Stepwise exchange from complex I to complex II depended on phosphorylation of Zic2 by DNA-PK and poly-(ADP-ribose) polymerase. Phosphorylated Zic2 protein made a stable complex with RNA helicase A, and complex II could interact with RNA polymerase II. Phosphorylation-dependent transformation of Zic2-containing molecular complexes may occur in transcriptional regulation.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Zinc finger (ZF)3 proteins belonging to the Zic family play important roles in animal development (1, 2). Chordate Zic genes control the development of neural and mesodermal tissue, and ecdysozoan Zic homologues control ectodermal and mesodermal development. In vertebrates, there are five Zic genes, each of which has a common or specific function in development. The biological and clinical importance of Zic2, a member of the vertebrate Zic gene family, has been demonstrated by several studies. In humans and mice, mutations of Zic2 result in holoprosencephaly, a forebrain anomaly characterized by loss of midline structures (3, 4). In addition, Zic2 mutant mice show various abnormalities, including spina bifida, axial/limb skeletal malformation, retinal axon mis-projection, and reduced production of neural crest cells (4). In Xenopus, overexpression of Zic2 induces expansion of the neural crest and neuroectoderm markers, and depletion of maternal Zic2 results in a gastrulation defect (5).

Although these facts elucidate the versatility of the Zic2 protein, its molecular function is not yet fully understood. Its nuclear localization and the altered expression status of various genes in Zic2-manipulated model animals strongly suggest that Zic2 has a critical role in gene expression, particularly in the transcriptional regulation. In fact, Zic2 overproduction causes marked activation of several promoters, including those of apolipoprotein E (apoE), SV40, adenovirus major late, HSV thymidine kinase, and Zic1 (6, 7). An intriguing point of the transcriptional activation capacity of Zic protein is its broad range of target genes (6, 7). This feature is consistent in vertebrate Zic1, Zic2, and Zic3. A similar capacity seems to be possessed by the urochordate Zic homologue Macho-1, because Macho-1 overproduction causes enhanced expression of a broad range of genes (8).

In structural terms, Zic2 protein possesses two functional domains known as Zic Opa-conserved (ZOC) and ZF. The ZF domain is composed of a tandem repeat of five C2H2 ZF motifs that is very similar to those of the Gli and Glis families (1). A crystallographic study of Gli1 has shown that it binds DNA with the last four fingers (9), where the similarity is particularly high between Gli and Zic families. ZOC is a stretch of around 10 amino acid residues, initially found to be conserved between the mouse and fly Zic homologues. A recent molecular phylogenic study has indicated that both ZOC and ZF have been retained in several animal phyla, suggesting the importance of these domains (10).

Previous structure-function analyses have indicated that ZF can bind to GC-rich DNA target sequences (1, 7) whereas the N-terminal region, including ZOC, plays a role in transcriptional regulation (11). However, the function of ZF is not limited to DNA binding, because ZF has significant transcriptional ability, even in the absence of clear binding sequences (11). In the case of ZOC, it is known that I-mfa, a repressor of myogenic basic helix-loop-helix type transcription factor, can bind to the region, including ZOC, and can inhibit the transcriptional activation by exporting the Zic2 protein from cell nuclei (11).

The above circumstances led us to investigate the molecular machinery that mediates transcriptional regulation by the Zic2 protein. We report here the isolation and characterization of two novel types of molecular complexes that include Zic2 protein. Our results suggest that the assemblies of these two complexes are essential for Zic2-dependent transcriptional activation. Our results also present a new role for DNA-dependent protein kinase in transcriptional regulation.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Lines—293T cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). For protein production, expression vectors were transfected with Lipofectamine and Plus reagent (Invitrogen). The DNA-PKcs knock-out cell line PK33S and the control PK34S line were established from the lung fibroblasts of knock-out or wild-type mice, as described previously (12). Cells were maintained in DMEM supplemented with 10% FBS and 5 mM L-glutamine. Ku80-deficient Xrs-6 hamster cells (13) were cultured in DMEM supplemented with 10% FBS, 5 mM L-glutamine, and nonessential amino acids (Invitrogen). A40 (mouse cerebellum granule cell line) cells were maintained in F-12/DMEM 1:1 medium (Sigma) containing 10% FBS, 5 mML-glutamine (14).

Plasmid Construction—The single epitope-tagged Zic2 expression vectors (pcDNA3-HA-Zic2 and pCMV-FLAG-Zic2), pcDNA3-HA coding terminal truncated Zic2 series, and the intact Zic2 expression vector (pEFBOS-Zic2) were described previously (7). The N-terminal double-tagged expression plasmids pCMV-F-HA-Zic2, pCMV-F-HA-RHA, and pCMV-F-HA-PARP were constructed by insertion of a full-length ORF into the EcoRI site of pCMV-F-HA-a or pCMV-F-HA-b. The cassette plasmids pCMV-F-HA-a and pCMV-F-HA-b were constructed by insertion of annealed oligonucleotide DNA fragments (5'-GGTACCCATACGACGTCCCAGACTACGCTGCG-3', 5'-GATCCGCAGCGTAGTCTGGGACGTCGTATGGGTACC-3', or 5'-GTACCCATACGACGTCCCAGACTACGCTGCG-3', 5'-GATCCGCAGCGTAGTCTGGGACGTCGTATGGGTAC-3') between the SrfI and BamHI sites of pCMVtag2a or pCMVtag2b (Stratagene). FLAG-HA-tagged DNA-PKcs expression vector was constructed by inserting an annealed oligonucleotide DNA fragment (5'-ATGTACCCATACGACGTCCCAGACTACGCTGATTACAAGGATGACGACGATAAGGCGGAGGAGGGAACCGGC-3' and 5'-GTACGGCCGGTTCCCTCCTCCGCCTTATCGTCGTCATCCTTGTAATCAGCGTAGTCTGGGACGTCGTATGGGTACATACGT-3') between the AatII and BsiwI sites of pME-PK7 that contained an entire ORF of DNA-PKcs (a gift from Dr. Masumi Abe). pHM6-HA-Ku70, pCMV-FLAG-Ku70, and pCMV-FLAG-Ku80 were constructed by insertion of PCR-cloned cDNA containing intact ORF of mouse Ku70 and Ku80 into the EcoRI site of pHM6 (Roche Applied Science) and/or pCMVtag2a. pcDNA3-HA-PARP was constructed by insertion of a PCR-cloned cDNA containing intact ORF of the mouse PARP1 into the pcDNA3HA EcoRI site. The luciferase reporter plasmid pGL4-apoE was constructed by insertion of a mouse apoE promoter (-190 to +51) between the KpnI and HindIII sites of pGL4 (Promega). The wild-type RHA and its mutant expression vectors were provided by Dr. Toshihiro Nakajima.

The Escherichia coli expression plasmids pGEX-3T-Zic2, pGEX-2T-Ku70, and pGEX-2T-Ku80 were constructed by insertion of a full-length ORF into the EcoRI site of pGEX plasmid (Amersham Biosciences). pET-TBP and pET-TFIIB were constructed by insertion of a mouse full-length ORF between BamHI and XhoI sites of pET21(+) (Novagen).

For baculovirus-mediated expression, pFastBac-His6-PARP was constructed by insertion of a full-length ORF of PARP cDNA into the NotI site of a pFastBac HTa donor plasmid. pFastBac-His6-GST-TEV-RHA was constructed by insertion of an EcoRI full-length ORF fragment of RHA from pcDNA3-HA-RHA and a GST ORF DNA fragment into the pFastBac-HTa plasmid.

Purification of Zic2-containing Complexes—293T cells were transfected with pcDNA3-HA-Zic2. After 24 h, cells were washed with PBS(-), including 1 mM phenylmethylsulfonyl fluoride (PMSF), and harvested by centrifugation. The cell pellet was suspended in lysis wash buffer 150 (40 mM HEPES-KOH (pH 7.8), 150 mM KCl, 10% glycerol, 0.5 mM dithiothreitol (DTT), 0.1 mM EDTA, 0.5% Nonidet P-40, 1 mM PMSF, 0.1% benzamidine HCl and protease inhibitor mixture (Roche Applied Science)) and then stored on ice for 60 min. A total cell extract was prepared after centrifugation (20,000 x g for 15 min) and mixed with anti-HA antibody-conjugated agarose beads for 6 h at 4 °C. Bound proteins were eluted by addition of HA peptide (100 µg/ml). The protein mixture was loaded on Bio-Rex70 (Bio-Rad) (1 ml) and washed five times with the column volume of lysis wash buffer, eluted with buffer containing 0.3, 0.6, and 1.2 M KOAc, stepwise. Free HA-tagged Zic2 was eluted with 0.6 M KOAc buffer, and the majority of the Zic2-containing protein complexes was eluted with 0.3 M KOAc buffer. To concentrate it, the eluted 0.3 M fraction was dialyzed against dialysis buffer (40 mM HEPES-KOH (pH 7.8), 50 mM KCl, 10% glycerol, 0.5 mM DTT, 0.1 mM EDTA, and 0.1% Nonidet P-40), loaded onto DE52 (0.1 ml), and eluted with buffer containing 0.2 M KCl. Eluted proteins were loaded onto a HiLoad 16/60 Superdex gel filtration column (Amersham Biosciences), and samples (0.5 ml) were collected after 40 ml had been eluted.

Identification of Proteins by Mass Spectrometry—Peaks of Zic2-containing complexes were concentrated and separated by SDS-PAGE. Coomassie Brilliant Blue-stained proteins were cleaved by trypsin, and these fragments were used in the analysis by ion trap mass spectrometer, liquid chromatography/tandem mass spectrometry system, LCQ-DECA XP (Thermo-Quest). The data were analyzed by using the MASCOT search program (Matrix Science, London, UK).

In Vitro Phosphorylation Assay—We performed in vitro phosphorylation using affinity-purified protein FLAG-HA-Zic2 (200 ng), baculovirus recombinant RHA (100 ng), PARP (N-terminal His6-tagged) (100 ng), and active DNA-PK (100 units) (Promega). The proteins were mixed into the reaction buffer (final concentration: 20 mM HEPES-KOH (pH 7.8), 75 mM NaCl, 2.5% glycerol, 5 mM MgCl, 1 mM DTT, and 0.025% Nonidet P-40) containing 100 ng of pGL4-apoE plasmid, 0.2 mM ATP, and 4.6 µCi of [{gamma}-32P]ATP in 14-µl reaction mixture, and the reaction was started by the addition of 1 µl (100 units) DNA-PK. Reaction mixtures were incubated at 30 °C for 30 min, and the reactions were stopped by the addition of 6x SDS sample buffer containing 10 mM EDTA. Proteins were separated on 7.5% SDS-PAGE, and the dried gel was exposed to imaging plates (Fuji) and analyzed by a PhosphorImager (BAS2500, Fuji).

Immunoblotting and Antibodies—Proteins were separated by SDS-PAGE, transferred on to polyvinylidene difluoride membranes, and detected with the following antibodies using ECL Western blotting detection reagent (Amersham Biosciences).

Anti-RHA antibody was generated in rabbits by using an E. coli recombinant C-terminal His6-tagged protein fragment corresponding to amino acids 1–299. The following antibodies were purchased: anti-HA rabbit polyclonal antibody (Sigma); anti-HA rat monoclonal antibody 3F10 (Roche Applied Science); anti-HA mouse monoclonal antibody 12CA5 (Medical and Biological Laboratories); anti-FLAG rabbit polyclonal antibody (Sigma); anti-FLAG mouse monoclonal antibody M2 (Sigma); anti-DNA-PK (StressGen); anti-Ku70 (Santa Cruz Biotechnology); anti-Ku86 (Santa Cruz Biotechnology); anti-acetylhistone H3 [Ac-Lys9] (Sigma); anti-dimethylhistone H3 [Me-Lys9] (Upstate); anti-TBP (Santa Cruz Biotechnology); anti-TFIIB (Pharmingen), anti-pol II (8WG16: CTD of RPB1) (QED Bioscience); anti-PARP (Sigma), anti-mouse IgG horseradish peroxidase-conjugated (Jackson ImmunoResearch); anti-rabbit IgG horseradish peroxidase-conjugated (Jackson ImmunoResearch); anti-rat IgG horseradish peroxidase-conjugated (Jackson ImmunoResearch); anti-mouse IgG Alexa 488 (Invitrogen); anti-rabbit IgG Alexa 488 (Invitrogen); anti-mouse IgG Alexa 594 (Invitrogen); and anti-rabbit IgG Alexa 594 (Invitrogen).

Coimmunoprecipitation Assay—Transfected cells were washed and harvested in PBS(-) containing 1 mM PMSF, and total cell extracts were prepared with lysis wash buffer 150 (20 mM HEPES-KOH (pH 7.8), 10% glycerol, 150 mM NaCl, 0.5 mM DTT, 0.1 mM EDTA, 0.5% Nonidet P-40, and 1 mM PMSF). The extracts were incubated with 5 µl of anti-HA or anti-FLAG affinity beads at 4 °C for 6 h. E10.5 mouse whole embryo and adult mouse cerebellum protein extracts were prepared by homogenization in the buffer. After centrifugation of the samples, the proteins in the supernatants were quantified, and 0.5 ml of diluted 0.5 mg/ml extracts were used for this assay. The protein extracts were mixed at 4 °C for 6 h with 5 µl of anti-Zic2 serum or preimmune serum, together with 5 µl of protein A affinity beads, and analyzed by Western blotting using specific antibodies.

For preparation of de-phosphorylated Zic2 protein, HA-Zic2 was first produced in 293T cells by transfection and was then immobilized on beads by incubating the cell lysate with anti-HA affinity beads. The beads were washed with lysis wash buffer containing 500 mM NaCl and then incubated for 30 min at 37 °C with 20 units of active or inactivated CIP (heat-treated at 65 °C for 40 min with 1 mM EDTA). Beads were washed with high salt buffer (the buffer containing 500 mM NaCl) two times and then used in this binding assay. For immunoprecipitation, recombinant RHA, PARP, and purified DNA-PK (50 ng each) were mixed with the beads in lysis wash buffer containing 0.1 mg/ml bovine serum albumin.

GST Pulldown Assay—FLAG-HA-Zic2 and Zic2-interacting proteins (FLAG-HA-DNA-PKcs, FLAG-HA-RHA, FLAG-HA-PARP, and HA-Ku70/FLAG-Ku80 dimer) were overproduced in 293T cells by transfection of double-tagged expression plasmids or cotransfection of FLAG- and HA-tagged protein expression plasmids. Cells on 2-1–cm dishes were harvested and lysed in 4 ml of lysis wash buffer 300 (20 mM HEPES-KOH (pH 7.8), 10% glycerol, 300 mM NaCl, 0.5 mM DTT, 0.1 mM EDTA, 0.5% Nonidet P-40, and 1 mM PMSF), and total cell extracts were incubated with anti-HA affinity beads (Sigma), 40 µl at 4 °C for 6 h. These beads were washed three times, and the bound proteins were eluted with 0.8 µl of the lysis wash buffer 300 containing 100 µg/ml HA peptide. Eluted proteins were incubated with anti-FLAG M2 affinity beads (Sigma), 10 µl for 12 h, and after the beads had been washed three times, the bound proteins were eluted with 0.1 ml of FLAG peptide (100 µg/ml). Any free peptides were eliminated by use of a centrifugal filter (Microcon YM-50; Millipore), and purified proteins were used for GST pulldown assays.

Glutathione S-transferase (GST) fusion proteins GST-Ku70 and GST-Ku80 were produced in E. coli (BL21) and purified with glutathione-Sepharose beads, as described (15). GST-Zic2 was additionally purified by heparin-Sepharose chromatography with a NaCl gradient. C-terminal His6-tagged proteins TBPch6 and TFIIBch6 were overproduced in E. coli BL21(DE3) strain by 0.5 mM isopropyl 1-thio-beta-D-galactopyranoside induction with an A600 of 0.4 at 37 °C (in LB media containing 200 µg/ml ampicillin). Harvested cells were disrupted in the buffer (40 mM HEPES-KOH (pH 7.8), 500 mM NaCl, 10% glycerol, 10 mM beta-mercaptoethanol, 0.5% Nonidet P-40, and 0.1 mM PMSF) containing 0.5 mg/ml lysozyme. After sonication, the extracts were centrifuged and loaded onto nickel-nitrilotriacetic acid (Ni-NTA)-agarose column (Qiagen). The column was washed with the buffer containing 1% Nonidet P-40, and protein was eluted by the buffer containing 300 mM imidazole. After dialysis against dialysis buffer (20 mM HEPES-KOH (pH 7.8), 50 mM NaCl, 10% glycerol, 0.5 mM DTT, and 0.1 mM EDTA), samples were loaded onto a heparin CL6B column, and proteins were eluted with the buffer containing 400–600 mM NaCl.

GST and GST fusion proteins immobilized on glutathione-Sepharose beads were incubated together with the purified proteins for 2 h at 4 °C in binding buffer containing 0.1 mg/ml bovine serum albumin. Interacting proteins were eluted with 1x SDS sample buffer and detected by Western blotting analysis using anti-HA (3F10) antibody.

Immunofluorescence Staining—PK34S and PK33S cell lines were transfected with pcDNA3-HA-Zic2 alone or together with pCMV-FLAG-Ku70 or pCMV-FLAG-Ku80. After 24 h, the cells were used for the immunofluorescence staining and analyzed by laser scanning confocal microscopy as described (11).

Preparation of Baculovirus Recombinant Proteins—A Bac-to-Bac baculovirus expression system (Invitrogen) was used for producing recombinant PARP and RHA proteins. Recombinant bacmids were generated by transformation of E. coli strain DH10bac by pFastBac-His6-PARP and pFastBac-His6-GST-TEV-RHA. Recombinant bacmid DNA was transfected to Sf9 insect cells, and the viruses generated were used for protein production. Two hundred milliliters of infected culture kept for three overnights was washed with PBS(-) containing 1 mM PMSF and harvested; the cells were then disrupted in lysis buffer (20 mM HEPES-KOH (pH 7.8), 10% glycerol, 150 mM NaCl, 10 mM beta-mercaptoethanol, 0.5% Nonidet P-40, and 1 mM PMSF). His6-PARP was bound to Ni-NTA-agarose beads (20 µl) in 1.5-ml microtubes and washed three times with 0.5 ml of lysis buffer containing 20 mM imidazole and 500 mM NaCl. The proteins were eluted with buffer containing 200 mM imidazole and dialyzed against dialysis buffer A (20 mM HEPES-KOH (pH 7.8), 10% glycerol, 50 mM NaCl, 0.5 mM DTT, 0.1 mM EDTA, and 0.1% Nonidet P-40) for use in the in vitro assay. GST-His6-RHA protein was bound to glutathione-Sepharose beads (20 µl). After the beads had been washed with buffer containing 300 mM NaCl, the protein was eluted with buffer containing 10 µM reduced glutathione. Purified protein was dialyzed against 50 mM Tris-HCl (pH 8.0) and 5 mM EDTA and then digested by addition of an enhanced form of tobacco etch virus protease (Invitrogen) and 1 mM DTT (final concentration). After dialysis of the sample against dialysis buffer B (20 mM HEPES-KOH (pH 7.8), 10% glycerol, 50 mM NaCl, 10 mM DTT, 10 mM beta-mercaptoethanol, and 0.1% Nonidet P-40), the digested GST protein and an enhanced form of tobacco etch virus protease were removed by using Ni-NTA-agarose beads and glutathione-Sepharose beads.


Figure 1
View larger version (46K):
[in this window]
[in a new window]

  		FIGURE 1.
Isolation of Zic2 complexes. A, strategy used for Zic2 complex purification. B, SDS-PAGE. Left panel shows control and HA-Zic2-containing samples that were eluted by HA peptide from anti-HA beads. Right panel is the result of Western blotting (IB) using anti-HA antibodies. C, gel filtration analysis. Two complexes (complex I (I) and complex II (II)) were observed. D, Zic2-associated proteins were identified by mass spectrometry. Panels show silver-stained SDS-polyacrylamide gels containing samples from fractions 16 and 44 in C, which contain complex I and complex II, respectively. Asterisks indicate contaminants that exist even without HA-tagged Zic2 expression (see B). E, three fractions were analyzed by Western blotting using antibodies against the proteins indicated. F and G, immunoprecipitation (IP) assay using anti-Zic2 antiserum or its control (preimmune serum), and tissue lysates prepared from E10.5 mouse whole embryos (F) and adult cerebellum (G). Input lanes contain 5 µg (5%) of tissue lysates subjected to immunoprecipitation. DNA-PKcs, RHA, PARP, Ku70, and Ku80 were analyzed by Western blotting.

 
Gel Shift Assay—A promoter containing the DNA fragment was prepared by PCR amplification of the apoE promoter region -78 to +40 by using the IRD700 (LI-COR) end-labeled primers 5'-IRD700-GCACAGCCCGCCCTAGCCCTGAGG-3', 5'-IRD700-GCTCCTTCCGAAACAAGTCCTTAGCC-3', and pGL4-apoE as a template. Protein-DNA complexes (100 fmol of DNA, 100 ng of F-HA-Zic2, and 100 ng of RHA for one reaction) were formed as described (15) and analyzed using 4% polyacrylamide gel and an infrared imaging system (LI-COR, Odyssey).

Luciferase Reporter Assay—Luciferase reporter assays were performed as described previously (7). Luciferase reporter plasmid pGL4 or pGL4-apoE (200 ng) was cotransfected with pEFBOS or pEFBOS-Zic2 (800 ng) and pEF-LacZ (200 ng). For the analysis in Fig. 7B, pcDNA3-HA or pcDNA-HA-PARP (200 ng) was also cotransfected using Lipofectamine (Invitrogen). Luciferase activity was measured according to the manufacturer's recommendations (Promega) using a luminometer, Minilumat LB 9506 (Berthold).

Chromatin Immunoprecipitation (ChIP)-PCR Assay—We performed the ChIP-PCR assay as described (16). A40 cell extracts, rabbit polyclonal antibodies against Zic2 and the components of the complexes, and normal rabbit serum were used for the immunoprecipitation. The precipitates were subjected for the PCR analysis using specific primers for apoE 5' upstream sequence (5'-AGGCTCTGTGGGCCGTGCTGTTGG-3' and 5'-TGGCCCGTGTCTCCTCCGCCACTG-3') (6) and control glyceraldehyde-3-phosphate dehydrogenase primers (5'-CCGGTGCTGAGTATGTCGTGGAGTCTAC-3' and 5'-CTTTCCAGAGGGGCCATCCACAGTCTTC-3') (17).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
To identify the essential functions of Zic2 protein, we purified Zic2-containing molecular complexes. N-terminal HA (hemagglutinin antigen)-tagged Zic2 expression vector was transfected into 293T cells. HA-Zic2 protein and interacting proteins were affinity-purified by anti-HA antibody immobilized beads. Bound proteins were eluted with HA peptide (Fig. 1, A and B), fractionated on Bio-Rex70, and concentrated by DE52 columns (Fig. 1A). The purified protein complexes were subjected to gel filtration analysis. Two high molecular weight complexes were separated by the gel filtration. One (complex I) had a molecular weight of more than 700,000 and the other (complex II) was between 232,000 and 440,000 (Fig. 1C). The components of complex I and complex II were identified by mass spectrometric analysis. Complex I contained a DNA-PKcs, poly(ADP-ribose) polymerase (PARP), Ku80, and Ku70, whereas RNA helicase A (RHA), Ku80, and Ku70 were identified in complex II (Fig. 1D). The components in each fraction were confirmed by Western blotting (Fig. 1E).

To confirm the existence of complex I and complex II in vivo, we carried out an immunoprecipitation assay using anti-Zic2 antibody. DNA-PKcs, RHA, PARP, Ku80, and Ku70 were coprecipitated with Zic2 from both E10.5 mouse whole embryo and adult cerebellum tissue lysates (Fig. 1, F and G). These results suggest that Zic2 is contained in complex I and complex II in vivo.

DNA-PKcs, Ku70, and Ku80 (Ku86) are known as subunits of a serine-threonine kinase, DNA-PK, which plays essential roles in V(D)J recombination, repair of DNA double-stranded breaks, and transcriptional regulation (18–23). PARP is a nuclear protein that catalyzes polymerization of the ADP-ribose unit to the target proteins. Recent studies have shown that PARP controls chromatin structure and association of transcriptional regulatory factors in a catalytic activity-dependent or -independent manner (24–29). RHA is an ATPase/double-stranded DNA and RNA helicase protein. RHA interacts with both CREB-binding protein and RNA polymerase II (pol II) to mediate this assembly, and it stimulates transcriptional activation (30). These facts prompted us to proceed further with the characterization of complex I and complex II.

We first performed an in vitro GST pulldown assay to examine the direct interactions between Zic2 and Zic2-associating proteins. Previous studies had shown protein-protein interactions between DNA-PKcs and PARP, Ku70/80 dimer, and RHA (31–34). GST-Zic2 fusion protein and control GST (Fig. 2A) were mixed with purified FLAG- and HA double-tagged (F-HA)-DNA-PKcs, F-HA-RHA, and F-HA-PARP or a dimer of HA-Ku70/F-Ku80 (Fig. 2B). DNA-PKcs, the Ku70/80 dimer, PARP, and RHA were co-pulled down with GST-Zic2 but not with GST (Fig. 2C). To further confirm the association between Zic2 and the Ku70/80 dimer, GST-Ku70 and GST-Ku80 were assayed with F-HA-Zic2 (Fig. 2, A and D). F-HA-Zic2 was pulled down with both GST-Ku70 and GST-Ku80. It was surprising that all Zic2-associating proteins directly interact with the Zic2 protein. We confirmed the absence of interaction with other nuclear proteins TBP and TFIIB by the same pulldown assay (Fig. 2E, right). These results indicated that all of the protein components in complex I and complex II were capable of binding Zic2.

We next examined the complex I and complex II component-binding domains in Zic2 by the coimmunoprecipitation assay. We transfected 293T cells with expression vectors for Zic2 and a truncated Zic2 mutant with an N-terminal HA tag (11) (Fig. 3A). Endogenous proteins were coimmunoprecipitated with the Zic2 mutants and analyzed using SDS-PAGE followed by Western blotting. Although all components were coimmunoprecipitated with full-length Zic2 and the mutants 1–485, 1–419, 1–393, 1–363, 102–531, and 141–531 (Fig. 3B, lanes 1–6, 9, and 10), Zic2 mutants 1–333 and 1–255 failed to coimmunoprecipitate RHA, PARP, Ku70, and Ku80 (Fig. 3B, lanes 7 and 8). The results revealed that RHA, PARP, Ku70, and Ku80 required ZF3 of the Zic2 protein (Fig. 3A). Collectively, these results indicate that there are two binding domains in Zic2 as follows: the 141–255 domain for DNA-PKcs binding and the ZF3 domain for RHA, PARP, Ku70, and Ku80. These binding sites may be involved in complex I and complex II formation.

Our next question was how the two complexes were formed and regulated. To address this point, we performed immunoprecipitation assays using DNA-PKcs or Ku80 knock-out cell lines. The PK33S cell line lacks the DNA-PKcs gene (Fig. 4A). The PK33S cell line and the wild-type control cell line PK34S were transfected with HA-Zic2 expression vector, and proteins coprecipitated with anti-HA antibody-immobilizing beads were analyzed by Western blotting. Although the Ku70/80 dimer was coprecipitated in both cell lines, the RHA and PARP proteins were not detected in immunoprecipitates from DNA-PKcs-deficient cells (Fig. 4A, lanes 6 and 8). These results indicate that DNA-PKcs is required for association of both RHA and PARP in Zic2-including complexes. On the other hand, in Ku80-deficient Xrs-6 cells, RHA and PARP were coprecipitated with Zic2, whereas Ku70 was not (Fig. 4B, lane 8). The absence of Ku70 may reflect the fact that heterodimerization of Ku70/80 is necessary for their nuclear localization (35). Although the association between Zic2 and the Ku70/80 dimer may stabilize the formation of complex I, it is not essential for binding to Zic2 (Fig. 4B, lane 8). Taking into account the fact that DNA-PKcs per se could bind to the Ku70/80 dimer, we considered that Zic2, DNA-PKcs, and Ku70/80 form a trimeric complex based on the interactions among the three proteins.


Figure 2
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 2.
A, GST fusion proteins GST-Zic2, GST-Ku70, and GST-Ku80 were produced in E. coli BL21 and purified using glutathione-Sepharose beads. CBB, Coomassie Brilliant Blue. B, FLAG- and/or HA-tagged proteins, F-HA-Zic2, HA-Ku70, F-Ku80, F-HA-DNA-PKcs, F-HA-PARP, and F-HA-RHA were produced in 293T cells and affinity-purified by anti-FLAG and anti-HA antibody-conjugated beads. C, GST pulldown assay between GST-Zic2 and F-HA-Zic2, HA-Ku70/F-Ku80, F-HA-DNA-PKcs, F-HA-PARP, or F-HA-RHA. Coprecipitated proteins were separated by SDS-PAGE and detected by Western blotting analysis; 10% of the input (binding mixture) and the precipitates using GST are shown as references. D, direct interactions between F-HA-Zic2 and GST-Ku70 or GST-Ku80. Zic2 could bind directly with all the proteins that we identified in complex I and complex II. E, purified proteins TBPch6 and TFIIBch6 (left). Other nuclear proteins TBP and TFIIB do not interact directly with Zic2 protein (right).

 


Figure 3
View larger version (32K):
[in this window]
[in a new window]

  		FIGURE 3.
Mapping of protein-protein interaction regions in Zic2. HA-tagged Zic2 and Zic2-terminally truncated expression vectors of the mutant were transfected into 293T cells, and the cell lysates from the transfectants were subjected to immunoprecipitation using anti-HA affinity beads. A, HA-tagged Zic2 constructs for the mapping. ZOC, Zic Opa conserved domain. Gray boxes with numbers show each C2H2 motif in the ZF domains. B, upper panels show proteins immunoprecipitated (IP) by anti-HA antibody. Proteins were analyzed by Western blotting using anti-HA antibody and specific antibodies against the proteins indicated. Lower panels indicate the results of Western blotting (immunoblotting; IB) using cell lysates of the transfectants and the same antibodies.

 


Figure 4
View larger version (49K):
[in this window]
[in a new window]

  		FIGURE 4.
DNA-PKcs is required for PARP and RHA association but not for Ku70 and Ku80. A, PK34S cells (wild type (WT)) and PK33S cells ({Delta}dna-pk) were transfected with HA-tagged Zic2 expression vector (pcDNA3-HA-Zic2) or empty vector (pcDNA3) and then immunoprecipitated with anti-HA affinity beads. A and B, Western blotting analyses of cell lysates prepared from the transfectants (left panel (IB)) and of immunoprecipitates (right panels (IP)) are indicated with the antibodies used. B, Xrs-6 ({Delta}Ku80) cells were transfected with HA-tagged Zic2 expression vector (pcDNA3-HA-Zic2) or empty vector (pcDNA3) in combination with FLAG-Ku80 expression vector (pCMV-FLAG-Ku80) or its empty vector (pCMV).

 
We performed immunofluorescence staining to clarify the subnuclear distributions of Zic2 and Zic2-interacting proteins in vivo. Cells were transfected with expression vectors for HA-tagged Zic2 and then doubly stained with anti-HA antibody and several specific antibodies. In the case of Ku70 and Ku80, FLAG-tagged expression vectors were cotransfected with HA-Zic2 expression vector. Zic2 was distributed mostly in the cell nuclei (Fig. 5, indicated in red). In the nuclei, Zic2 was colocalized with K9-acetylated histone H3, which represented the active state chromatin (Fig. 5, panels a1–8), but its distribution did not overlap with that of K9-dimethylated histone H3, which is believed to repress transcription by compacting chromatin (Fig. 5, panels b1–8). The subnuclear distribution of Zic2 overlapped with that of DNA-PKcs (Fig. 5, panels g2–4). We also examined the subnuclear distributions of the other Zic2-associated proteins, namely PARP, RHA, Ku70, and Ku80, in PK34S wild-type cells and PK33S DNA-PKcs-deficient cells. PARP and RHA were localized mostly in the nucleus and slightly in the cytoplasm of the cells of both lines (Fig. 5, panels c2 and 6, and panels d2 and 6). In PK34S cells their nuclear localization overlapped to a large degree with that of Zic2 protein (Fig. 5, panels c4 and d4). However, colocalization with Zic2 was not detected in PK33S DNA-PKcs-deficient cells (Fig. 5, panels c8 and d8). Ku70 and Ku80 were located mainly in the nucleus (Fig. 5, panels e3 and 7 and f3 and 7), and their subnuclear distributions coincided well not only in PK34S cells but also in PK33S cells (Fig. 5, panels e4 and 8 and panels f4 and 8). These results confirmed the essential function of DNA-PKcs in the assembly of PARP and RHA proteins with Zic2 protein.


Figure 5
View larger version (47K):
[in this window]
[in a new window]

  		FIGURE 5.
Zic2 colocalizes with PARP, RHA, Ku70, Ku80, and DNA-PKcs in the nucleus. HA-tagged Zic2 expression vector was transfected into PK34S (wild type) and PK33S ({Delta}dna-pk) cells. This was followed by double immunofluorescence detection of HA-Zic2- and the Zic2-binding proteins. For Ku70 and Ku80, N-terminally FLAG-tagged Ku70 or Ku80 expression plasmids were cotransfected into both cell lines. Anti-acetylhistone H3 (a marker of active chromatin) and anti-dimethylhistone H3 (a marker of inactive chromatin) were used to characterize the Zic2 subnuclear localization. Detected proteins or epitopes and stained cell lines are indicated to the left of the panels. Immunopositive signals for HA-Zic2 are shown in red, and those for tested proteins indicated are shown in green; the overlapping of red and green signals is shown in yellow in merge, and control 4',6-diamidino-2-phenylindole (DAPI) staining is shown in blue. Scale bars are 10 µm.

 
DNA-PKcs is involved in complex I, interacting directly with Zic2, and is required for the binding of PARP and RHA to Zic2 (Figs. 1, 4, and 5). PARP and RHA have been identified as substrates of DNA-PKcs (34, 36). Is phosphorylation of Zic2, PARP, or RHA involved in the formation of Zic2-containing molecular complexes? To answer this question, we performed in vitro phosphorylation assays using purified FLAG-HA-tagged Zic2 (Fig. 2B), PARP, and RHA proteins produced in insect cells (Fig. 6A). Proteins were incubated with DNA-PK together with [{gamma}-32P]ATP and supercoiled plasmid DNA, which contained the Zic2 binding target sequence of the apoE promoter. Zic2 was slightly phosphorylated by DNA-PK, but the extent of the phosphorylation was greatly increased by the addition of PARP (Fig. 6B, lanes 6 and 7). PARP was also slightly phosphorylated by DNA-PK irrespective of the presence of Zic2 (Fig. 6B, lanes 5–7), but no phosphorylation of RHA was detected (Fig. 6B, lanes 2 and 4). The phosphorylations of Zic2 and PARP were DNA-dependent (Fig. 6B, lane 8), being consistent with the enzymatic activity of DNA-PK. These results suggest that the main target of DNA-PK is Zic2 and that PARP is required to increase phosphorylation efficiency. In agreement with our results, a previous study showed that the protein kinase activity of DNA-PK can be stimulated by PARP (36).


Figure 6
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 6.
Phosphorylation of Zic2 by DNA-PK is enhanced by PARP, and Zic2 phosphorylation changes the priority for protein-protein interactions. A, PARP and RHA proteins used in the experiments. Both proteins were expressed in insect cells using a baculovirus expression system. CBB, Coomassie Brilliant Blue. B, in vitro phosphorylation assay. Zic2, PARP, and RHA were incubated with [{gamma}-32P]ATP and DNA-PK in the presence or absence of supercoiled DNA (DNA). The combinations of the incubated proteins and DNA are indicated above the panel. The reaction mixture was electrophoresed, and the phosphorylated protein was detected by a PhosphorImager. The positions of phosphorylated Zic2 and PARP are indicated by arrowheads. C, coimmunoprecipitation assay was carried out using HA-Zic2 and dephosphorylated HA-Zic2. HA-Zic2 produced in 293T cells was treated with CIP and inactivated CIP before the binding assay. Western blotting of immunoprecipitates is indicated with the antibodies used. Immunoprecipitation was carried out with anti-HA affinity beads. Control indicates the experiment carried out with empty vectors instead of HA-Zic2 expression vector at transfection. Binding between Zic2 and RHA was decreased by CIP treatment, whereas that between Zic2 and PARP was not (left). Graph shows the binding efficiency of de-phosphorylated Zic2 protein (right). The averages and standard errors of the means for three independent experiments are shown. D, highly purified F-HA-Zic2 and CIP-treated dephosphorylated F-HA-Zic2 are indicated by arrows. E, interaction between Zic2 and RHA was analyzed by gel shift assay using an IRD700-labeled apoE promoter DNA fragment, recombinant RHA, purified F-HA-Zic2, and dephosphorylated form of F-HA-Zic2.

 
Because DNA-PKcs is essential for the assembly of PARP and RHA proteins (Figs. 1, 4, and 5), we speculated that phosphorylation of Zic2 might affect the binding of PARP and RHA to Zic2. To test this possibility, we next examined whether Zic2 phosphorylation was required for the assembly of DNA-PK, PARP, and RHA. A co-pulldown assay was performed using Zic2 immobilized beads that were prepared by incubating a HA-Zic2-containing cell extract with anti-HA antibody-conjugated beads. The beads were treated with CIP or inactivated CIP. After being washed, purified RHA, PARP, and DNA-PK proteins were incubated with both types of Zic2 beads, and the bound and unbound proteins were analyzed by Western blotting (Fig. 6C). Although inactivated CIP treatment did not affect the interaction between Zic2 and RHA (Fig. 6C, left third panel from top, lane 3), dephosphorylation by active CIP reduced the interaction (Fig. 6C, left third panel from top, lane 4, right graph). In contrast, PARP and DNA-PK binding was not affected by active CIP treatment (Fig. 6C, left DNA-PK, upper panel, lanes 3 and 4; PARP, fifth panel from top, lanes 3 and 4). CIP treatment of highly purified F-HA-Zic2 resulted in the disappearance of a band with slow mobility (Fig. 6D), suggesting that a fraction of Zic2 was actually phosphorylated in vivo.We subsequently performed a gel shift assay using intact or dephosphorylated F-HA-Zic2 with the apoE promoter DNA fragment in the presence or absence of purified RHA. The DNA-Zic2 complex was partly supershifted by the presence of RHA. The supershifted band was not detected when we used dephosphorylated Zic2 instead of intact Zic2 (Fig. 6E, lanes 4 and 6). This suggests that transformation of the Zic2-containing complexes, complex I and complex II, is regulated by the kinase activity of DNA-PK on Zic2.

We next investigated the functional significance of complex I and complex II in transcriptional regulation by Zic2. Both wild-type PK34S cells and DNA-PKcs-deficient PK33S cells were transfected with apoE promoter-luciferase reporter plasmid and Zic2 expression plasmid, and the luciferase activities were measured. Although the apoE promoter was activated by Zic2 in PK34S cells more than 3-fold (Fig. 7A), activation was reduced in PK33S cells.

Gain-of-function of PARP was also tested in Zic2-dependent transcriptional regulation. PK34S cells were transfected with apoE promoter-luciferase reporter and Zic2 and PARP expression vectors. PARP was increased Zic2-dependent transcriptional activation to a level about 1.5 times greater than empty vector (Fig. 7B).

To confirm the coexistence of the Zic2 and Zic2-associating proteins in the transcriptional regulation, we next analyzed gene-specific association of them on the Zic2-controlling promoter by the ChIP-PCR method. ApoE 5' Zic2 binding region (6) was coimmunoprecipitated not only with anti-Zic2 antibody but also with anti-DNA-PKcs, anti-Ku70, anti-Ku80, anti-RHA, and anti-PARP (Fig. 7C). Therefore, Zic2 and Zic2-associating proteins may act together on the specific target genes.

RHA has been identified as a component of the pol II-holo complex and regulates cAMP-response element-binding protein (CREB)-dependent transcription through recruitment of pol II (30, 37). In complex II, RHA may act as a transcription coactivator for recruitment of pol II and activate the transcription recycling of Zic2 target genes. To verify this idea, we examined the presence of pol II in Zic2-RHA complexes prepared by transfecting FLAG-Zic2 and/or HA-RHA expression vectors. The largest subunit of pol II was detected in the FLAG-Zic2 transfectant cell extract by immunoprecipitation using anti-FLAG (Fig. 7D, lane 7). Coexpression of FLAG-Zic2 and HA-RHA resulted in an increased amount of coimmunoprecipitated pol II (Fig. 7D, lane 8). However, coexpression of an RHA point mutant, HA-rhaW339A, which lacked the ability to bind pol II (Fig. 7D, lane 9) with FLAG-Zic2, failed to increase the amount of pol II precipitated. By contrast, HA-rhaK417R, an RHA mutant lacking ATP-binding ability, increased the amount of coimmunoprecipitated pol II, to a level comparable with that in wild-type RHA coexpression (Fig. 7D, lane 10). General transcription factors TBP and TFIIB were also detected in the precipitates (Fig. 7D, lanes 7–10), because both factors are associated directly or indirectly with RHA (38) but not associated with Zic2 protein (Fig. 2E). The amounts of coimmunoprecipitated TBP were not affected by overproduction of RHA or in the two mutants (Fig. 7D, lanes 7–10). In contrast, we detected increased amounts of coimmunoprecipitated TFIIB by overproduction of RHA and also in the two mutants (Fig. 7D, lanes 8–10). This suggests that there is a functional relationship between RHA and TFIIB in interacting with pol II in Zic2-dependent transcriptional regulation. These results indicate that pol II is recruited to complex II through protein-protein interaction with RHA, and that the ATP-dependent helicase activity of RHA is not essential for this recruitment. Therefore, the essential function of RHA is tightly associated with pol II recruitment and recycling, but this function may not be required in the early stages of initiation-complex formation.


Figure 7
View larger version (42K):
[in this window]
[in a new window]

  		FIGURE 7.
DNA-PK is required for Zic2-dependent transcriptional activation in vivo. A, PK34S (wild type (WT)) and PK33S ({Delta}dna-pk) cell lines were transfected with a control luciferase reporter vector without any cis-acting elements (pGL4) or apoE promoter-driven luciferase (pApoE-Luc), together with an empty vector (pEFBOS, open bar) or Zic2 expression vector (pEFBOS-Zic2, closed bar). B, PARP expression vector (pcDNA-HA-PARP, open bar) or empty plasmid (pcDNA3, closed bar) and pEFBOS-Zic2 or pEFBOS was cotransfected with pApoE-Luc reporter vector and an internal control vector (pEF-lacZ). All luciferase activities were normalized to the activities of beta-galactosidase. A and B, the averages and means ± S.E. for three independent experiments of three samples each are shown. C, ChIP-PCR analysis of the apoE promoter. The cross-linked apoE 5' DNA region was immunoprecipitated with anti-Zic2, anti-DNA-PKcs, anti-Ku70, anti-Ku80, anti-RHA, anti-PARP, and nonimmune rabbit serum. A Zic2-binding region was PCR-amplified from all samples. Glyceraldehyde-3-phosphate dehydrogenase (G3PDH) shows that negative controls of nonspecific binding with all target proteins. D, coimmunoprecipitation was performed using 293T cells that were cotransfected with Zic2 expression vector (pCMV-FLAG-Zic2, F-Zic2) or control plasmid (pCMV) and RHA (pcDNA3-HA-RHA, HA-RHA), pol II binding-defective RHA (pcDNA3-HA-rhaW339A, HA-rhaW339A), ATP binding-defective RHA (pcDNA3-HA-rhaK417R, HA-rhaK417R) expression vector, or control plasmid (pcDNA3). The combinations of the transfected plasmids are indicated above the panels. Immunoprecipitation was carried out with anti-FLAG affinity beads. Western blotting analyses of 10% of the input protein sample (left panel (immunoblotting (IB)) and of immunoprecipitates (right panel (immunoprecipitation (IP)) are indicated with the antibodies used. Coexpression of HA-RHA enhanced pol II coimmunoprecipitated and the pol II-binding-defective RHA (HA-rhaW339A) did not show increased amounts of coimmunoprecipitated pol II. ATP-binding-defective RHA (HA-rhaK417R) also showed enhanced binding with pol II.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
We have reported here two novel Zic2-containing molecular complexes that are composed of DNA-PKcs, Ku70/80 dimer, and PARP (complex I) or RHA, and Ku70/80 dimer (complex II). All protein components directly interact with Zic2 protein. The third ZF motif of Zic2 protein is essential for association with the Ku70/80 dimer, PARP, and RHA. DNA-PKcs can phosphorylate Zic2 and is important for interaction as well as subnuclear colocalization between Zic2 and RHA. Zic2 phosphorylation by DNA-PK is required for transformation from complex I to complex II (Fig. 8). The transcriptional activation of Zic2 also depends on DNA-PK and PARP in vivo.

DNA-PKcs is a large protein (~460 kDa) that belongs to the ataxia telangiectasia-mutated protein family of serine/threonine protein kinases. The functions of DNA-PK have been well investigated in the DNA repair pathway against double strand DNA breaks (39). Previous studies have revealed that besides its role in nonhomologous end-joining, DNA-PKcs phosphorylates a number of transcription factors and chromatin-associated proteins, such as p53, Oct-1, Pdx-1, histoneH1, TBP, TFIIB, and pol II, in vitro (22, 40–48).


Figure 8
View larger version (24K):
[in this window]
[in a new window]

  		FIGURE 8.
Protein-protein interactions in Zic2-dependent transcription. ZF domain of Zic2 protein binds to DNA upstream of the target gene promoter. At first, Zic2 protein recruits the Ku70/80 dimer, DNA-PKcs, and PARP. In this first step, complex I is constructed for phosphorylation of Zic2 protein by DNA-PK and PARP (top right). In the next step, RHA is recruited to the phosphorylated Zic2 protein and PARP, and DNA-PKcs are dissociated (bottom right). Finally, RHA makes complex II with the phosphorylated Zic2 (bottom left) and contacts general transcription factors TBP and TFIIB. Complex II is able to recruit pol II and then activates the Zic2-dependent transcription. However, we have not yet determined what disrupts complex II and dephosphorylates the Zic2 (top left).

 
Our study and previous ones have revealed that components of complex I or complex II, Zic2, RHA, PARP, Ku70, and DNA-PKcs, can themselves be substrates for DNA-PK-dependent phosphorylation. However, at this stage of the analysis Zic2 has particularly attracted our attention because its phosphorylation promotes binding to RHA, and Zic2 and RHA are differentially distributed in DNA-PKcs-deficient cells. This latter finding of ours is supported by a previous finding that silencing of DNA-PKcs by small interfering RNA leads to changes in the subnuclear distribution of RHA (34). The same researchers found that DNA-PKcs and Ku70/80 dimer are copurified with RHA protein. Although we did not detect a Zic2-containing molecular complex that included both DNA-PKcs and RHA, this interaction supports the idea that dissociation of DNA-PKcs/PARP and association of RHA simultaneously occur in the conversion from complex I to complex II. On the basis of the above facts, we can speculate that DNA-PKcs has a key role in the transition from complex I to complex II (Fig. 8).

Complex I includes PARP. PARP catalyzes poly(ADP-ribosylation), which plays diverse roles in many molecular and cellular processes, including DNA damage detection and repair, chromatin modification, and transcription (49). A number of chromatin-associated proteins have been identified as targets of poly(ADP-ribosylation) by PARP family proteins (50, 51). The enzymatic activity of PARP regulates core histone, transduction-like enhancer corepressor switching, and CCCTC-binding factor, a chromatin insulation protein-dependent chromatin insulation (27, 29). As a component of complex I, PARP associates with Ku70/80 in mammalian cells, and the PARP and Ku proteins synergistically bind to the nuclear matrix attachment region in the genomic sequences (52). Considering the chromatin association of the Ku70–80-PARP complex, modification of the chromatin structure is noted as a possible role of PARP in complex I. Our finding that Zic2 is localized with active chromatin raises the possibility that Zic2 actively participates in chromatin remodeling processes. However, its effect on the chromatin state is not clear from the results of our immunofluorescence staining of Zic2-overexpressing cells.

Instead, the role of PARP in complex I shown in this study was the stimulation of DNA-PK enzymatic activity in Zic2 phosphorylation. Previous studies have also shown that PARP associates with DNA-PK and stimulates kinase activity (36, 53), and conversely, DNA-PK suppresses PARP enzymatic activity by phosphorylation (53, 54). These facts seem to be in agreement with the situation in complex I, where Zic2 was phosphorylated by DNA-PK in the presence of PARP. Interestingly, although PARP can directly interact with Zic2 in the binding assay using the purified proteins, it was not coprecipitated with Zic2 in the DNA-PKcs-deficient cells. This suggests that Zic2-PARP interaction is stabilized by PARP-DNA-PKcs interaction in vivo. Besides its effects on DNA-PKcs, the enzymatic activity of PARP itself could have a role in the complex I. However, we failed to identify poly(ADP-ribose) modification of Zic2 by Western blotting using anti-poly(ADP-ribose) antibodies, although we found auto-modification of PARP protein (data not shown).

RHA is a nuclear ATP-dependent helicase that unwinds both DNA and RNA (55). A transcriptional role for RHA in mammalian cells has been indicated by its bridging activity between CREB-binding protein and pol II, and RHA has been characterized as a transcriptional coactivator in CREB-dependent transcription (30, 56). RHA interacts with other nuclear factors, NF-{kappa}B and BRCA1 (breast cancer gene 1, a breast- and ovarian-specific tumor suppressor 1), directly and also interacts with TBP and TFIIB directly or indirectly (38, 57, 58). In addition to the role of RHA in the binding to pol II, an ATP-dependent process independently acts in Zic2-dependent transcriptional regulation process. Possibly, RHA recruits pol II on the core promoter, and ATPase/helicase activity enhances the initiation or promoter clearance of the pol II. In complex II, the former mechanism is supported by the fact that pol II was present in the Zic2 coprecipitates, and the amount of pol II was reduced by the presence of an RHA mutant lacking pol II binding activity (Fig. 7). The presence of DNA-Zic2-RHA is also in agreement with this mechanism. Thus, RHA may be involved in Zic2-dependent transactivation as a coactivator that bridges Zic2 protein, pol II, and general transcription factors.

Although this study has revealed the details of Zic2-containing molecular complexes, our preliminary results indicate that Zic2-associated proteins can physically interact with Zic1 and Zic3 (data not shown). It is likely that the same molecular machineries are utilized in vertebrate Zic1 and Zic3, which show similar DNA-binding properties (7), transcription-activating capacities (2, 5, 7), and tissue differentiation-inducing abilities (5). Recent studies have shown that other members of the vertebrate Zic family (Zic4 and Zic5, which share highly conserved zinc finger domains) have different transcriptional activation and tissue differentiation-inducing abilities (15, 59). The similarities and differences between the Zic-binding proteins, including any similarities to the components of Zic2 Complexes I and II, await further investigation.

We found it surprising that DNA-PKcs, Ku70/80, PARP, and RHA were identified in the Zic2-containing molecular complexes, because in the past these proteins have come to the notice of researchers mainly as components of DNA repair systems. However, recent studies are shedding new light on their roles in the normal course of development. Both Ku80 and Ku70 mutants show reduced body size in mice, and fibroblasts derived from Ku80-deficient mice show early loss of proliferation (60). Ku80/PARP compound-mutant mice demonstrate a developmental defect that occurs as early as E6.5 (the time of onset of gastrulation) but later than E3.5 (the preimplantation stage), indicating the essential roles of these proteins in early embryonic development (61). In addition, analysis of RHA knock-out mice indicates the essential role of RHA in embryonic ectoderm differentiation around the time of gastrulation (62). These roles in early embryonic development seem to partly overlap with those of the Zic family genes. Zic1, Zic2, Zic3, and Zic5 mutant mice show regional hypoplasia of the central nervous system or the neural crest generating tissues (63–65). Collectively, the contact points between Zic family genes and Ku70/80, RHA, and PARP in a developmental context may be important enough for further evaluation. In the case of DNA-PKcs, its role in the early stage of mammalian development has not been suggested by its knock-out mice phenotypes. However, considering the presence of the structurally related phosphatidylinositol 3-kinase gene that acts so early in mammalian development, it may not be appropriate to discuss it in the current context.


   FOOTNOTES
 
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence may be addressed. Tel.: 81-48-467-9791; Fax: 81-48-467-9792; E-mail: akiraishiguro{at}brain.riken.go.jp. 2 To whom correspondence may be addressed. Tel.: 81-48-467-9791; Fax: 81-48-467-9792; E-mail: jaruga{at}brain.riken.go.jp.

3 The abbreviations used are: ZF, zinc finger; PARP, poly(ADP-ribose) polymerase; RHA, RNA helicase A; DNA-PKcs, DNA-dependent protein kinase catalytic subunit; GST, glutathione S-transferase; ORF, open reading frame; PBS, phosphate-buffered saline; PMSF, phenylmethylsulfonyl fluoride; DMEM, Dulbecco's modified Eagle's medium; FBS, fetal bovine serum; DTT, dithiothreitol; Ni-NTA, nickel-nitrilotriacetic acid; HA, hemagglutinin; CREB, cAMP-response element-binding protein; ChIP, chromatin immunoprecipitation; CIP, calf intestinal alkaline phosphatase; TBP, TATA-binding protein. Back


   ACKNOWLEDGMENTS
 
We thank Masumi Abe (National Institute of Radiological Sciences, Japan) and Toshihiro Nakajima (St. Marianna University School of Medicine) for providing both the plasmids and helpful advice. We are also grateful to Yasuhiro Tomooka (Tokyo University of Science) and Ryushin Mizuta (Tokyo University of Science) for providing cell lines (A40 and Xrs-6, respectively). We are indebted to the Research Resources Center at the RIKEN Brain Science Institute, Japan, for DNA sequencing and mass spectrometric analysis.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Aruga, J. (2004) Mol. Cell. Neurosci. 26, 205-221[CrossRef][Medline] [Order article via Infotrieve]
  2. Grinberg, I., and Millen, K. J. (2005) Clin. Genet. 67, 290-296[CrossRef][Medline] [Order article via Infotrieve]
  3. Brown, S. A., Warburton, D., Brown, L. Y., Yu, C. Y., Roeder, E. R., Stengel-Rutkowski, S., Hennekam, R. C., and Muenke, M. (1998) Nat. Genet. 20, 180-183[CrossRef][Medline] [Order article via Infotrieve]
  4. Nagai, T., Aruga, J., Minowa, O., Sugimoto, T., Ohno, Y., Noda, T., and Mikoshiba, K. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 1618-1623[Abstract/Free Full Text]
  5. Nakata, K., Nagai, T., Aruga, J., and Mikoshiba, K. (1998) Mech. Dev. 75, 43-51[CrossRef][Medline] [Order article via Infotrieve]
  6. Salero, E., Perez-Sen, R., Aruga, J., Gimenez, C., and Zafra, F. (2001) J. Biol. Chem. 276, 1881-1888[Abstract/Free Full Text]
  7. Mizugishi, K., Aruga, J., Nakata, K., and Mikoshiba, K. (2001) J. Biol. Chem. 276, 2180-2188[Abstract/Free Full Text]
  8. Yagi, K., Satoh, N., and Satou, Y. (2004) Dev. Biol. 274, 478-489[CrossRef][Medline] [Order article via Infotrieve]
  9. Pavletich, N. P., and Pabo, C. O. (1993) Science 261, 1701-1707[Abstract/Free Full Text]
  10. Aruga, J., Kamiya, A., Takahashi, H., Fujimi, T. J., Shimizu, Y., Ohkawa, K., Yazawa, S., Umesono, Y., Noguchi, H., Shimizu, T., Saitou, N., Mikoshiba, K., Sakaki, Y., Agata, K., and Toyoda, A. (2006) Genomics 87, 783-792[CrossRef][Medline] [Order article via Infotrieve]
  11. Mizugishi, K., Hatayama, M., Tohmonda, T., Ogawa, M., Inoue, T., Mikoshiba, K., and Aruga, J. (2004) Biochem. Biophys. Res. Commun. 320, 233-240[CrossRef][Medline] [Order article via Infotrieve]
  12. Fukumura, R., Araki, R., Fujimori, A., Mori, M., Saito, T., Watanabe, F., Sarashi, M., Itsukaichi, H., Eguchi-Kasai, K., Sato, K., Tatsumi, K., and Abe, M. (1998) J. Biol. Chem. 273, 13058-13064[Abstract/Free Full Text]
  13. Tzung, T. Y., and Runger, T. M. (1998) Int. J. Radiat. Biol. 73, 469-474[CrossRef][Medline] [Order article via Infotrieve]
  14. Minakawa, M., Sugimoto, T., Aizawa, S., and Tomooka, Y. (1998) Brain Res. 813, 172-176[CrossRef][Medline] [Order article via Infotrieve]
  15. Ishiguro, A., Inoue, T., Mikoshiba, K., and Aruga, J. (2004) Biochem. Biophys. Res. Commun. 324, 302-307[CrossRef][Medline] [Order article via Infotrieve]
  16. Weinmann, A. S., and Farnham, P. J. (2002) Methods (San Diego) 26, 37-47
  17. Hoshino, J., Aruga, J., Ishiguro, A., and Mikoshiba, K. (2003) Brain Res. Mol. Brain Res. 120, 57-64[Medline] [Order article via Infotrieve]
  18. Blunt, T., Finnie, N. J., Taccioli, G. E., Smith, G. C., Demengeot, J., Gottlieb, T. M., Mizuta, R., Varghese, A. J., Alt, F. W., Jeggo, P. A., and Jackson, S. P. (1995) Cell 80, 813-823[CrossRef][Medline] [Order article via Infotrieve]
  19. Gao, Y., Chaudhuri, J., Zhu, C., Davidson, L., Weaver, D. T., and Alt, F. W. (1998) Immunity 9, 367-376[CrossRef][Medline] [Order article via Infotrieve]
  20. Giffin, W., Torrance, H., Rodda, D. J., Prefontaine, G. G., Pope, L., and Hache, R. J. (1996) Nature 380, 265-268[CrossRef][Medline] [Order article via Infotrieve]
  21. Lieber, M. R. (1999) Genes Cells 4, 77-85[Abstract]
  22. Peterson, S. R., Jesch, S. A., Chamberlin, T. N., Dvir, A., Rabindran, S. K., Wu, C., and Dynan, W. S. (1995) J. Biol. Chem. 270, 1449-1454[Abstract/Free Full Text]
  23. Giffin, W., Gong, W., Schild-Poulter, C., and Hache, R. J. (1999) Mol. Cell. Biol. 19, 4065-4078[Abstract/Free Full Text]
  24. Anderson, M. G., Scoggin, K. E., Simbulan-Rosenthal, C. M., and Steadman, J. A. (2000) J. Virol. 74, 2169-2177[Abstract/Free Full Text]
  25. Cervellera, M. N., and Sala, A. (2000) J. Biol. Chem. 275, 10692-10696[Abstract/Free Full Text]
  26. Hassa, P. O., Covic, M., Hasan, S., Imhof, R., and Hottiger, M. O. (2001) J. Biol. Chem. 276, 45588-45597[Abstract/Free Full Text]
  27. Ju, B. G., Solum, D., Song, E. J., Lee, K. J., Rose, D. W., Glass, C. K., and Rosenfeld, M. G. (2004) Cell 119, 815-829[CrossRef][Medline] [Order article via Infotrieve]
  28. Pavri, R., Lewis, B., Kim, T. K., Dilworth, F. J., Erdjument-Bromage, H., Tempst, P., de Murcia, G., Evans, R., Chambon, P., and Reinberg, D. (2005) Mol. Cell 18, 83-96[CrossRef][Medline] [Order article via Infotrieve]
  29. Yu, W., Ginjala, V., Pant, V., Chernukhin, I., Whitehead, J., Docquier, F., Farrar, D., Tavoosidana, G., Mukhopadhyay, R., Kanduri, C., Oshimura, M., Feinberg, A. P., Lobanenkov, V., Klenova, E., and Ohlsson, R. (2004) Nat. Genet. 36, 1105-1110[CrossRef][Medline] [Order article via Infotrieve]
  30. Nakajima, T., Uchida, C., Anderson, S. F., Lee, C. G., Hurwitz, J., Parvin, J. D., and Montminy, M. (1997) Cell 90, 1107-1112[CrossRef][Medline] [Order article via Infotrieve]
  31. Gottlieb, T. M., and Jackson, S. P. (1993) Cell 72, 131-142[CrossRef][Medline] [Order article via Infotrieve]
  32. Ariumi, Y., Turelli, P., Masutani, M., and Trono, D. (2005) J. Virol. 79, 2973-2978[Abstract/Free Full Text]
  33. Sartorius, C. A., Takimoto, G. S., Richer, J. K., Tung, L., and Horwitz, K. B. (2000) J. Mol. Endocrinol. 24, 165-182[Abstract]
  34. Zhang, S., Schlott, B., Gorlach, M., and Grosse, F. (2004) Nucleic Acids Res. 32, 1-10[Abstract/Free Full Text]
  35. Koike, M., Awaji, T., Kataoka, M., Tsujimoto, G., Kartasova, T., Koike, A., and Shiomi, T. (1999) J. Cell Sci. 112, 4031-4039[Abstract]
  36. Ruscetti, T., Lehnert, B. E., Halbrook, J., Le Trong, H., Hoekstra, M. F., Chen, D. J., and Peterson, S. R. (1998) J. Biol. Chem. 273, 14461-14467[Abstract/Free Full Text]
  37. Aratani, S., Fujii, R., Fujita, H., Fukamizu, A., and Nakajima, T. (2003) Int. J. Mol. Med. 12, 175-180[Medline] [Order article via Infotrieve]
  38. Pellizzoni, L., Charroux, B., Rappsilber, J., Mann, M., and Dreyfuss, G. (2001) J. Cell Biol. 152, 75-85[Abstract/Free Full Text]
  39. Collis, S. J., DeWeese, T. L., Jeggo, P. A., and Parker, A. R. (2005) Oncogene 24, 949-961[CrossRef][Medline] [Order article via Infotrieve]
  40. Lees-Miller, S. P., Chen, Y. R., and Ander