Casein Kinase II-mediated Phosphorylation of the C Terminus of Sp1 Decreases Its DNA Binding Activity*

(Received for publication, December 16, 1996, and in revised form, March 13, 1997)

Susan A. Armstrong Dagger §par , Denise A. Barry Dagger **, Robert W. Leggett Dagger ** and Christopher R. Mueller Dagger §**Dagger Dagger

From the Dagger  Cancer Research Laboratories, ** Department of Biochemistry, § Department of Pathology, Queen's University, Kingston, Ontario K7L 3N6, Canada

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES


ABSTRACT

We have previously observed that Sp1, a ubiquitous zinc finger transcription factor, is phosphorylated during terminal differentiation in the whole animal, and this results in decreased DNA binding activity (Leggett, R. W., Armstrong, S. A., Barry, D., and Mueller, C. R. (1995) J. Biol. Chem. 270, 25879-25884). In this study, we demonstrate that casein kinase II (CKII) is able to phosphorylate the C terminus of Sp1 and results in a decrease in DNA binding activity. This suggests that CKII may be responsible for the observed regulation of Sp1. Mutation of a consensus CKII site at amino acid 579, within the second zinc finger, eliminates phosphorylation of this site and the CKII-mediated inhibition of Sp1 binding. Phosphopeptide analysis confirms the presence of a CKII site at Thr-579 as well as additional sites within the C terminus. No gross changes in CKII subunit levels were seen during de-differentiation associated with liver regeneration. The serine/threonine phosphatase PP1 was identified as the endogenous liver nuclear protein able to dephosphorylate Sp1 but again no gross changes in activity were observed in the regenerating liver. Okadaic acid treatment of K562 cells increases Sp1 phosphorylation and inhibits its DNA binding activity suggesting that steady state levels of Sp1 phosphorylation are established by a balance between kinase and phosphatase activities.


INTRODUCTION

Sp1 was originally characterized as a GC box binding protein (1) recognizing the consensus sequence GGGCGG. Its DNA binding domain consists of three C2H2 zinc fingers (2), and a series of four domains required for transcriptional activity of Sp1 have been characterized (3). Two of these domains, A and B, correspond to glutamine-rich regions (4-6) that interact with the transcriptional machinery by binding to TAFII110 (7) and are needed for transcriptional synergy to occur (8). Domain C contains a region of high charge and functions only weakly as an independent transactivation domain (4). Domain D is required for synergistic activation in conjunction with the A and B domains and may be involved in the formation of higher order homomeric complexes (8). The zinc fingers and domain D may also be involved in the interaction of Sp1 with other proteins as they are required for binding to proteins such as YY1 (9), GATA-1 (10), and adenovirus E1A (11). Sp1 is a member of a small multi-gene family, with Sp2 and Sp3 being ubiquitously expressed (12, 13), whereas the expression of Sp4 may be limited to the brain (13). Sp3 recognizes the same DNA sequences as Sp1 and may act as a repressor of Sp1-mediated activation (14).

Sp1 has traditionally been considered to be a constitutive transcription factor and has been implicated in the regulation of a wide variety of housekeeping genes and genes involved in growth regulation (15). It is becoming increasingly clear that Sp1 binding and transactivation is regulated by a variety of stimuli. The retinoblastoma gene product appears to be able to modulate Sp1-mediated transactivation (16-18) possibly through the release of Sp1 from a negative regulator of transactivation (19). A variety of growth factors may regulate Sp1 activity as this transcription factor has been shown to mediate the epidermal growth factor stimulation of the gastrin promoter (20), the insulin-like growth factor I regulation of the elastin gene (21), and the transforming growth factor-beta -mediated activation of both the alpha 2(I) collagen gene (22) and p15INK4B (23). Transactivation of the insulin receptor gene by E1a is thought to be mediated by Sp1 (24), and increased Sp1 activity may be involved in tumor progression of a carcinoma cell line (25). Expression of the p21CIP1/WAF1 cyclin-dependent kinase inhibitor in response to phorbol ester and okadaic acid treatments is also mediated through Sp1 sites (26). These findings suggest that modulation of Sp1 activity plays a critical role in the regulation of cellular growth and differentiation.

The Sp1 protein is subject to two different forms of post-translational modification. It is extensively glycosylated with O-linked sugars, which appear to play some role in transactivation (27). It is also phosphorylated by a DNA-dependent protein kinase (28), but this phosphorylation has not been shown to alter the activity of Sp1. Sp1 also becomes phosphorylated during the process of terminal differentiation (29). This modification results in the down-regulation of the DNA binding activity of Sp1 in rat liver nuclear extracts as well as in extracts from other organs. In this study, we demonstrate that the serine/threonine kinase, casein kinase II (CKII),1 is able to phosphorylate the C terminus of Sp1 and results in decreased Sp1 DNA binding. This suggests that CKII may be responsible for the observed phosphorylation of Sp1 during terminal differentiation in vivo. Increased phosphorylation of Sp1 in response to okadaic acid treatment suggests that phosphatases may also be involved in the regulation of Sp1 phosphorylation levels.


EXPERIMENTAL PROCEDURES

Liver Nuclear Extract Preparation

Nuclear extracts were prepared at a concentration of 5-10 µg/µl, according to Gorski et al. (30), with the modifications described in Maire et al. (31). The liver regeneration extracts were prepared by treating adult male rats with CCl4 and preparing extracts on day 2 as described in Mueller et al. (32).

Kinase Assays

Phosphorylation of full-length Sp1 produced using a vaccinia virus vector (Promega) or Sp1-maltose fusion proteins (see below) was carried out in a 15- or 20-µl reaction volume of 25 mM HEPES, pH 7.5, 34 mM KCl, 50 mM MgCl2 at 30 °C for 45 min in the presence of 1 µl of partially purified CKII, 5 µCi of [gamma -32P]ATP, and ±15 µg of bovine serum albumin. Some reactions contained 100 ng of the indicated, unlabeled oligonucleotides shown below.
<AR><R><C><UP>cSp1:</UP></C></R><R><C></C></R><R><C><UP>DBP site I:</UP></C></R><R><C></C></R><R><C><UP>DBP site II:</UP></C></R><R><C></C></R><R><C><UP>RCE:</UP></C></R><R><C></C></R><R><C><UP>site C:</UP></C></R><R><C></C></R></AR><AR><R><C><UP> 5′-CT</UP><UP>GC</UP><UP>G</UP><UP>G</UP><UP>GGCGGGGCAGAC</UP></C></R><R><C><UP>            CCCGCCCCGTCTGACG-5′</UP></C></R><R><C><UP> 5′-CTCTGGGTGCAGATTGCAAAG</UP></C></R><R><C><UP>            </UP><UP>CCCACGTCTAACGTTTCGAGA-5′</UP></C></R><R><C><UP> 5′-TTTGCTTTGGCCACGCCTCAAGGCTG</UP></C></R><R><C><UP>            </UP><UP>GAAACCGGTGCGGAGTTCCGACAAAC-5′</UP></C></R><R><C><UP> 5′-ACCGTGCCCGCGCGCCACCCCTCTGGCGCC</UP></C></R><R><C><UP>            </UP><UP>GGGCGCGCGGTGGGGAGACCGCGGTGGCAC-5′</UP></C></R><R><C><UP> 5′-GTAGGAACCAATGAAATGCGAGGTAAGTAT</UP></C></R><R><C><UP>            </UP><UP>TTGGTTACTTTACGCTCCATTCATACATCC</UP>-5′</C></R></AR>
The samples were then separated on an SDS-polyacrylamide gel. The gel was fixed or stained and subsequently dried and autoradiographed.

Phosphatase Analysis of Liver Nuclear Extracts

The Sp1 fusion protein, amino acids 521-696, was 32P-labeled using CKII as described above. The labeled substrate was then exposed to rat liver nuclear extracts in the presence or absence of phosphatase inhibitors for the indicated length of time. Samples were separated on a 10% SDSpolyacrylamide gel. The gel was subsequently fixed, dried, and autoradiographed.

Gel Mobility Shift Assays

Gel mobility shift assays were performed as described in Lichsteiner and Schibler (33) except that the binding reactions lasted for 15 min on ice before the samples were loaded on a 6% nondenaturing polyacrylamide gel. In the supershift experiments, the antibody was incubated with the sample for 30 min on ice followed by the 15-min binding reaction with the labeled probe before being loaded on a gel. The anti-Sp1 antibody is a rabbit affinity purified polyclonal antibody raised against residues 520-538 of the human Sp1 protein and was obtained from Santa Cruz Biotechnology, Inc. The cSp1 oligonucleotide shown above was used as the probe in all cases.

Western Blot Detection

Protein separated by SDS-polyacrylamide gel electrophoresis was electroblotted onto nitrocellulose, and the Sp1 protein was detected using the anti-Sp1 antibody (above) and a peroxidase-linked goat anti-rabbit IgG secondary antibody. The complexes were detected using the Renaissance chemiluminescence system according to manufacturer's instructions (DuPont NEN). The anti-alpha , anti-alpha ', and anti-beta antibodies raised against the respective CKII subunits are rabbit polyclonal antibodies and were a generous gift from David Litchfield. These were detected using a goat anti-rabbit IgG alkaline phosphatase conjugate antibody (Sigma).

Production of Sp1 Maltose Fusion Proteins

A cDNA fragment spanning nucleotides 1561-2091 of Sp1 was generated by polymerase chain reaction using Vent DNA polymerase (New England BioLabs) and pPac Sp1 (generously supplied by R. Tjian) as the template, and the Sp1 521 and Sp1 696 primers, which contain EcoRI and HindIII sites, respectively. A mutation at nucleotide 1735 (A right-arrow G) was generated via polymerase chain reaction-based site-directed mutagenesis with primers Sp1 521 and 90 used to generate the 5' portion, and primers Sp1 696 and 89 used to generate the 3' portion of the mutant Sp1 cDNA. This clone also encodes amino acids 521-696 of Sp1, but it contains a threonine to alanine mutation at codon 579 within the second zinc finger. Standard techniques were used to clone the Sp1 cDNAs into plasmid pBS+ (Stratagene) at the EcoRI and HindIII sites. The derived clones were completely sequenced.
<AR><R><C><UP>Sp1 521:</UP></C></R><R><C><UP>Sp1 696:</UP></C></R><R><C><UP>89:</UP></C></R><R><C><UP>90:</UP></C></R></AR><AR><R><C><UP> 5′-GGAATTCAAAGACAGTGAAGGAAGGGG-3′</UP></C></R><R><C><UP> 5′-CCAAGCTTCAGAAGCCATTGCCACTG-3′</UP></C></R><R><C><UP> 5′-GGAAACGCTTC<B>G</B>CACGTTCG-3′</UP></C></R><R><C><UP> 5′-CGAACGTG<B>C</B>GAAGCGTTTCC-3′</UP></C></R></AR>

Each Sp1 cDNA in pBS+ was recloned using the EcoRI and HindIII sites into pMAL-c2, a maltose binding protein fusion expression vector (New England BioLabs). Colonies were screened, and large scale protein preparations were made according to manufacturers' instructions, with the following modifications. Induced bacteria from a 150-ml culture were pelleted and resuspended in 20 ml of protein lysis buffer (10 mM Tris-HCl, pH 8.0, 50 mM NaCl, 1 mM EDTA, 5% (v/v) glycerol) containing 100 µM ZnSO4 and protease inhibitors (10 mM DTT, 1 µg/ml leupeptin, and pepstatin A, 1 mM benzamidine, 1 mM phenylmethylsulfonyl fluoride, 1% Trasylol). Protein lysis buffer (1 ml) containing 20 mg/ml lysozyme was added, and the mixture was incubated on ice for 30 min and then 1 ml of 5 M NaCl was added with a further incubation of 15 min on ice, and the mixture was centrifuged at 30,000 rpm in a 55.2Ti rotor (Beckman) for 30 min at 4 °C. Two ml of the supernatant was applied to a 0.5-ml amylose resin column and was then washed with 3 × 2 ml of NDB (25 mM HEPES, pH 7.6, 0.1 mM EDTA, 40 mM KCl, 10% (v/v) glycerol, 1 mM DTT) containing 100 µM ZnSO4. The fusion protein was eluted with 2 × 1 ml of NDB with 10 mM maltose and 100 µM ZnSO4; the protein containing fractions were pooled, and the protein was denatured by the addition of urea to 5 M and incubation on ice for 1 h. Protein samples were then dialyzed against NDB containing 10 mM DTT and 100 µM ZnSO4 once and 3 × NDB with 1 mM DTT and 100 µM ZnSO4 to refold the protein in the presence of zinc. The Sp1 521-696 wild type and Thr-579 mutant proteins used in Fig. 2 were affinity purified on a calf thymus DNA-cellulose column before use.


Fig. 2. Effect of casein kinase II phosphorylation on a Sp1 C-terminal fragment and a point mutant. A, maltose fusion proteins consisting of Sp1 protein from aa521 to aa696, as well as the same region with a mutation of threonine to an alanine residue at aa579 were successively affinity purified on amylose and calf thymus DNA columns. These proteins were phosphorylated using purified CKII and then used in a bandshift assays with the consensus Sp1 site as probe. Lanes are indicated as in Fig. 1A. B, the same proteins as in A were phosphorylated using purified CKII and [gamma -32P]ATP. Labeled proteins were then separated on a 10% SDS-PAGE gel and autoradiographed. The positions of Sp1 and the autophosphorylated CKII beta  subunit are indicated.
[View Larger Version of this Image (30K GIF file)]

Phosphopeptide Mapping

Sp1-maltose binding protein fusion proteins that had been phosphorylated as described above were separated on a 10% SDS-polyacrylamide gel and electroblotted onto polyvinylidene fluoride protein blotting membrane (Corning Costar Corp.). The membrane was re-wet in methanol for 1 min and washed 3 × 2 min in 1 liter of distilled H2O and then the region containing the labeled protein was excised.

Membrane pieces were soaked in 0.5% polyvinylpyrrolidone in 100 mM acetic acid at 37 °C for 30 min and then washed with 5 × 1 ml of distilled H2O followed by freshly made 50 mM NH4HCO3 twice. The membranes were then incubated in 200 µl of 50 mM NH4HCO3, pH 8.0 (day old), containing 10 µg of L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin (Sigma) for 3 h at 37 °C. Another 10 µg of trypsin was added, and the membrane was incubated for a further 3 h at 37 °C. The sample was diluted with 800 µl of distilled H2O, centrifuged for 5 min at 15,000 × g, and the supernatant collected and lyophilized. The pellet was resuspended in 300 µl of distilled H2O and lyophilized two more times. The dry pellet was then resuspended in 50 µl of performic acid (900 µl of formic acid (98%) + 100 µl of hydrogen peroxide (30%), 60 min at room temperature) and incubated on ice for 1 h. The sample was lyophilized after 1 ml of distilled H2O was added. The pellet was resuspended in 10 µl of pH 1.9 electrophoresis buffer, spotted on a thin layer cellulose plate, and run in the first dimension using the pH 1.9 buffer for 1 h at 1000 V on the Hunter thin layer electrophoresis unit (HTLE-7000; CSB Scientific, Inc.) The phosphopeptides were separated in the second dimension by thin layer chromatography with the following buffer: 40/50/10/40 (v/v) n-butyl alcohol/pyridine/acetic acid/water. The plates were dried and autoradiographed.

K562 Nuclear Extract Preparation

K562 cells, a human chronic myelogenous leukemia cell line, were grown in suspension in 25-ml T flasks at 37 °C, 5% CO2 atmosphere in RPMI media (Life Technologies, Inc.) supplemented with 10% fetal bovine serum and 50 units/ml penicillin/streptomycin. Induction of Sp1 phosphorylation was achieved by supplementing the media with okadaic acid (Life Technologies, Inc.) at a final concentration of 100 ng/ml. Nuclear extracts were prepared according to Schreiber et al. (34), either before treatment or 2 h after the addition of okadaic acid.


RESULTS

We have previously observed that Sp1 is phosphorylated in nuclear extracts prepared from mature rat liver as well as from a variety of other tissues (29). This phosphorylation results in an approximately 10-fold decrease in the affinity of Sp1 for its cognate site. The site or sites phosphorylated in Sp1 as well as the kinase responsible have not been identified. Examination of the primary sequence of Sp1 revealed three potential phosphorylation sites for CKII, one of which is located within the second zinc finger of the DNA binding domain of Sp1. The down-regulation of AP-1 activity by CKII during growth arrest (35) suggested a potential involvement of this kinase in Sp1 phosphorylation during terminal differentiation. To examine the role of CKII in Sp1 regulation, the effect of CKII phosphorylation on Sp1 DNA binding activity was assessed. Full-length affinity purified Sp1 prepared from a vaccinia virus vector was used in a standard bandshift assay with a consensus Sp1 site as the probe. CKII alone, Sp1 alone, Sp1 with CKII and ATP, or Sp1 with CKII but no ATP were tested (Fig. 1A). Only when both CKII and ATP were present was the binding of Sp1 inhibited. An in vitro kinase assay using [gamma -32P]ATP was also carried out with purified CKII and Sp1 to determine if Sp1 is a substrate for phosphorylation. Specific labeling of Sp1 was observed only in the presence of both proteins (Fig. 1B). Due to the presence of the putative CKII site within the second zinc finger of Sp1, it seemed possible that binding of a specific DNA to Sp1 might interfere with phosphorylation. Double-stranded oligonucleotides corresponding to high affinity Sp1 sites including elements derived from the SV40 21-base pair repeat (Sp1), from site II of the DBP gene promoter (II), and from the retinoblastoma control element (RCE), which has a lower affinity for Sp1 but which is still specific (29), were added to the kinase assay before the addition of CKII. The binding of these oligonucleotides resulted in a dramatic decrease in the phosphorylation of Sp1 by CKII (Fig. 2B). The addition of non-Sp1 binding sites had no effect on CKII phosphorylation (sites I and C).


Fig. 1. Phosphorylation of Sp1 by casein kinase II and inhibition of Sp1 DNA binding activity. A, purified full-length Sp1 derived from a vaccinia virus vector was used in a bandshift assay with a consensus Sp1 binding site as the probe. Proteins were subject to a kinase reaction before being used in the bandshift reaction. Sp1 alone (Sp1), casein kinase II alone (CKII), or Sp1 with the addition of CKII and ATP (BOTH), or Sp1 and CKII but no ATP (-ATP) were also assayed. The position of protein-DNA complexes (bound) or uncomplexed DNA (free) are indicated. B, purified CKII and Sp1 were used in a kinase assay with [gamma -32P]ATP, separated on a 7% SDS-PAGE gel, and autoradiographed. Various double-stranded oligonucleotides corresponding to the indicated sites (see text) were added to the kinase reaction before the addition of CKII. The position of Sp1 is indicated.
[View Larger Version of this Image (33K GIF file)]

To localize those Sp1 residues necessary for CKII-mediated inhibition, a maltose binding protein fusion was created that encompassed the Sp1 zinc finger domain and the C-terminal D domain, from amino acids 521 to 696 (Fig. 3A). This region contains only one consensus CKII site. A point mutant of this construct was also generated that changes the theoretical phosphate acceptor in the CKII consensus site (at aa579) from a threonine to an alanine residue which should inactivate the site. These proteins were purified on a calf thymus DNA-cellulose column before being used in the binding and kinase assays. Both proteins bound to a consensus Sp1 site with comparable affinities (Fig. 2A). Phosphorylation with purified CKII inhibited the binding of the wild type protein, and this effect was dependent on the presence of ATP. This defined the C-terminal 175 amino acids of Sp1 as being sufficient for being inhibited by CKII. In contrast, binding of the protein containing the point mutation at Thr-579 was not inhibited by treatment with CKII and ATP. Phosphorylation of Thr-579 is clearly required for CKII-mediated inhibition of Sp1 DNA binding activity. The effect of this mutation on Sp1 phosphorylation was examined using an in vitro kinase assay with the same two proteins. Both proteins are phosphorylated to a comparable degree, with the Thr-579 mutant being slightly more active (Fig. 2B). This indicates that Thr-579 is not the only, or even the predominant, CKII phosphorylation site within the region of Sp1 from aa521 to aa696. The maltose binding protein alone was not phosphorylated by CKII (data not shown). Phosphorylation of the mutant protein is also suggested by the shift in mobility of the mutant protein complex in the bandshift reaction upon CKII treatment (Fig. 2A). Overall, these results indicate that Thr-579 is essential for CKII-mediated inhibition of Sp1 DNA binding activity despite it not representing the major phosphorylation site in this protein. However, the almost complete blocking of Sp1 phosphorylation by binding of a specific oligonucleotide suggests that these other CKII sites are in some way also masked upon DNA binding.


Fig. 3. Phosphotryptic analysis of casein kinase II-phosphorylated Sp1 proteins. A, the primary amino acid sequence of the region of Sp1 from aa521 to 696 is shown. The cysteine and histidine residues believed to coordinate zinc in the finger structure are shaded, and the core zinc finger domains are indicated by the thick line and labeled as ZF 1-3. The location of the consensus casein kinase II site at Thr-579 is boxed and is the residue mutated to an alanine in the mutant protein. Thick vertical arrows indicate the location of trypsin cleavage sites. B, approximately equal weights of Sp1 fusion proteins (as indicated) were phosphorylated in vitro with purified CKII and [gamma -32P]ATP and separated on a 10% SDS-PAGE gel followed by transfer to an Immobilon membrane. Labeled proteins were subject to tryptic digestion and separated by two-dimensional chromatography, with the direction of separation indicated, and autoradiographed. Spots are numbered as referred to in the text.
[View Larger Version of this Image (43K GIF file)]

Phosphotryptic analysis of the wild type and Thr-579 mutant fusion proteins was carried out to confirm the presence of an active CKII phosphorylation site at aa579. Four spots are readily visible with the full-length protein with the most prominent having relatively low mobility in both dimensions (spot 2) (Fig. 3B). Examination of the tryptic cleavage sites in Sp1 (Fig. 3A) reveals that two large C-terminal fragments would result from protease digestion. The presence of both phosphoserine and phosphothreonine in spot 2 (data not shown) makes it likely that this spot is the aa624-672 tryptic fragment as the aa673 to aa696 peptide does not contain threonines. The large size of this fragment would be consistent with its low mobility in the two-dimensional separation. Spot 1 appears to represent a proteolytic cleavage product of spot 2. Tryptic mapping of the Thr-579 mutant reveals that both of the weaker upper spots (spots 3 and 4) are lost (Fig. 3B). Spot 3 likely represents the three-residue peptide that contains Thr-579, and its disappearance indicates that this site is phosphorylated by CKII. Spot 4, which also disappears, is likely a partial cleavage product as there is a lysine residue adjacent to the N-terminal arginine in this peptide (Fig. 3A) so that cleavage at one or the other residue by trypsin is likely. These mapping experiments indicate that Thr-579, as well as a number of serine and possibly threonine residues within the D domain, are phosphorylated by CKII.

In the regenerating rat liver Sp1 is less phosphorylated compared with the normal liver (29). The ability of CKII to alter Sp1 binding led us to examine if changes in CKII subunit levels in rat liver nuclear extracts are responsible for the differences in Sp1 phosphorylation. Western blot analysis of the alpha , alpha ', and beta  CKII subunits in nuclear extracts from normal and regenerating liver indicated that the levels of all three proteins were essentially the same in both (Fig. 4). This suggested that gross CKII subunit levels were not subject to regulation but did not rule out alteration in activity due to other interactions. The possibility that dephosphorylation of Sp1 might be regulated led to the examination of endogenous phosphatases that could dephosphorylate CKII-phosphorylated Sp1. The aa521-696 Sp1 fusion protein was phosphorylated with CKII and [gamma -32P]ATP and then mixed with normal rat liver nuclear extracts. These extracts were able to dephosphorylate Sp1 over the course of a 45-min incubation (Fig. 5A). Increasing concentrations of okadaic acid were then added, which inhibit various serine/threonine phosphatases at characteristic concentrations (Fig. 5A). At 100 nM of okadaic acid all dephosphorylation was inhibited, and this inhibition was more efficient than a mix of nonspecific phosphatase inhibitors. This indicates that it is likely PP1 that is able to dephosphorylate Sp1. This phosphatase has been suggested to be regulated under some circumstances (36, 37). However, when the phosphatase activity in normal and regenerating rat liver nuclear extracts was examined using a time course incubation, no difference in overall phosphatase levels was observed (Fig. 5B).


Fig. 4. Western blot analysis of casein kinase II subunits in rat nuclear extracts from normal or regenerating liver. Normal (N) and regenerating (R) liver extracts (40 µg) were separated on a 12% SDS-PAGE gel and electroblotted onto nitrocellulose. Anti-casein kinase II subunit alpha , alpha ', and beta  antibodies were used to detect each specific subunit of CKII. The location of the specific subunit is indicated by an arrow. As a control, 100 µg of the peptide used to generate the alpha ' peptide was added to the primary incubation with that antibody.
[View Larger Version of this Image (45K GIF file)]


Fig. 5. Dephosphorylation of casein kinase II-phosphorylated Sp1. A, the Sp1 fusion protein 521-696 was phosphorylated (phos.) using purified CKII and [gamma -32P]ATP and incubated with 20 µg of regenerating rat liver nuclear extract in the presence or absence of various phosphatase inhibitors. The general phosphatase inhibitor combination (MIX) consisted of a final concentration of 1 mM each of sodium fluoride, sodium orthovanadate, and sodium phosphate. Okadaic acid was added at the indicated concentrations. The reactions were run on a 10% SDS-PAGE gel and autoradiographed. The arrow indicates the position of the Sp1 fusion protein. B, dephosphorylation reactions were carried out as described in A except that nuclear extracts from both normal and regenerating liver were used. The reactions were carried out for the indicated times, at which point they were stopped by the addition of SDS loading buffer. The reactions were then run on a 10% SDS-PAGE gel and autoradiographed. The arrow indicates the location of the Sp1 fusion protein.
[View Larger Version of this Image (38K GIF file)]

In vitro assays are limited in that they may disrupt interactions essential for regulation of a given process. To investigate if Sp1 phosphorylation is a dynamic process involving both active phosphorylation and dephosphorylation, the effect of okadaic acid treatment on Sp1 phosphorylation in vivo was assayed. K562 cells were treated with 100 ng/ml okadaic acid for 2 h, at which time nuclear extracts were prepared. Bandshift assays with the consensus Sp1 site, using normal and okadaic acid-treated extracts, indicates that Sp1 DNA binding activity is essentially eliminated by okadaic acid treatment (Fig. 6A). Addition of an Sp1 antibody results in a supershift of approximately half of the DNA-protein complex in the untreated extract, confirming that this factor is Sp1. The remaining DNA binding activity represents Sp3 which is subject to similar phosphorylation-mediated regulation as Sp1.2 A longer exposure of these lanes is shown in Fig. 6B to illustrate the level of reduction in DNA binding activity. Dephosphorylation of these extracts using alkaline phosphatase restores Sp1 DNA binding activity in the okadaic acid-treated extract, and the protein composition was again confirmed using a supershift assay. The extracts were also compared using an NF-Y binding site (C site), whose binding activity remains relatively constant in most cells, and indicated that slightly less protein is present in the treated extracts. Western blot analysis of Sp1 protein in these extracts was also carried out (Fig. 6C). Again, slightly less Sp1 was present in the treated extracts, but the differences were not large. No differences in mobility of the protein after treatment with okadaic acid, or after dephosphorylation, were observed. This indicates that the amount of phosphorylation of these proteins is not high enough to affect their mobility. The phosphatase treatment of these extracts increases the DNA binding activity of Sp1 even in the untreated extracts, suggesting that a significant population of Sp1 exists in the phosphorylated form in these cells. These experiments do not establish which kinase is responsible for Sp1 phosphorylation in vivo. However, by inhibiting the endogenous phosphatase, okadaic acid treatment results in the phosphorylation of the remaining population of Sp1, indicating that Sp1 phosphorylation is an active process in the cell.


Fig. 6. Regulation of Sp1 phosphorylation in K562 cells treated with okadaic acid. A, nuclear extracts were prepared from control (-) or K562 cells that had been treated with 100 ng/ml okadaic acid for 2 h (OA). Bandshift assays were carried out using the consensus Sp1 site or an NF-Y site (C SITE) as the probe. Extracts were also dephosphorylated before being added to the bandshift reactions (dePhosph). Antibody specific to Sp1 was also added to the bandshift reactions where indicated (+Sp1 Ab). The location of supershifted complexes (S.S.), DNA-protein complexes (bound), and free DNA (free) are indicated. All lanes were autoradiographed for the same length of time. B, the indicated lanes were autoradiographed approximately three times longer than in A. C, nuclear extracts, as indicated, were subject to Western blot analysis using an Sp1 antibody. The location of the Sp1 protein is indicated by the arrow.
[View Larger Version of this Image (62K GIF file)]


DISCUSSION

Our initial observation that down-regulation of Sp1 DNA binding activity by phosphorylation occurs during terminal differentiation (29) led us to investigate which kinase was involved in this process. The presence of several consensus CKII phosphorylation sites in Sp1, as well as previous reports linking CKII phosphorylation to growth arrest (35), strongly suggested a role for this kinase in this process. In particular, the location of one of these CKII sites within the second zinc finger of Sp1 was suggestive of a possible mechanism by which phosphorylation would interfere with binding of Sp1 to DNA. The demonstration that CKII is able to phosphorylate full-length Sp1 produced from a vaccinia virus vector and that this phosphorylation interfered with Sp1 DNA binding activity indicates that CKII is able to regulate Sp1 activity. The ability of CKII to both phosphorylate and inhibit the binding of a fusion protein containing the C-terminal 175 amino acids of Sp1 (aa521-696) further localized the CKII phosphorylation site to this region. In addition, the ability of an Sp1 recognition site oligonucleotide to block the majority of the CKII-mediated phosphorylation of both the full-length and truncated proteins suggests that this C-terminal region contains most or all of the active sites. However, mutation of the acceptor threonine (aa579) in the consensus CKII site did not eliminate CKII phosphorylation of Sp1, suggesting that other non-consensus CKII sites exist in the C terminus of Sp1. This is consistent with phospho-amino acid analysis which indicates that the majority of phosphorylation occurs on serine and not threonine residues (data not shown). Phospho-tryptic analysis comparing the wild type and Thr-579 mutant proteins indicates that residue 579 is phosphorylated. Furthermore, the mutation eliminates the ability of CKII to inhibit the binding of Sp1 to its recognition site in vitro. Extrapolation of the three-dimensional structure of Sp1 (38) from that of Zif-268 (39) indicates that Thr-579 likely lies in a position directly opposite the phosphate backbone of the DNA helix. Phosphorylation of this residue would provide a strong charge-charge repulsion, and given that the contribution of the middle zinc finger of Sp1 to total binding activity is thought to be of relatively greater importance than the other two (40), this could produce the observed decrease in binding affinity.

The Thr-579 residue is the only consensus CKII site in the C-terminal fragment; however, phosphorylation of non-consensus sites has been observed in numerous instances (41). The majority of the CKII phosphorylation is on serine residues and appears to be localized in the C-terminal D domain, particularly within a tryptic peptide from aa624 to aa672. This region contains 5 serine and 5 threonine residues. The binding of a specific oligonucleotide to Sp1 blocks almost all CKII-mediated phosphorylation. Due to the location of Thr-579, DNA binding would be expected to block the access of CKII to this site. Inhibition of D domain phosphorylation by DNA binding suggests that some tertiary structure may exist involving both the zinc finger domain and the D domain. Either the CKII sites cluster together in a configuration that places all of these sites in a region where access of the kinase is subject to steric hindrance by the presence of DNA, or DNA binding results in a conformational change which disrupts the CKII sites in the D domain.

We have previously observed changes in the phosphorylation state of Sp1 during liver development and regeneration (29). CKII has been associated with the control of growth through its ability to phosphorylate c-jun and thus down-regulate the AP-1 complex (35). In addition, disruption of CKII function as a result of parasitic infection or in a CKII transgenic animal results in oncogenic transformation (42). All of these observations suggest a role for CKII in growth regulation. The possible involvement of Sp1 in mediating retinoblastoma function (16, 43, 44) and the ability of Sp1 to regulate critical factors in growth regulation such as p21CIP1/WAF1 (26) may imply that CKII-mediated regulation of Sp1 is a critical point in controlling withdrawal from the cell cycle as the initial stage of terminal differentiation. Examination of gross CKII subunit levels and phosphatase activity in normal and regenerating rat liver nuclear extracts has not revealed any substantial changes in either activity.

Experiments in K562 cells clearly indicate that Sp1 phosphorylation is a dynamic process, involving the active phosphorylation and dephosphorylation of Sp1 to establish steady state levels. We have not yet been able to directly establish that CKII is responsible for this in vivo phosphorylation. PP1 has recently been suggested to modulate Sp1 activity in response to the glucose-mediated activation of the acetyl-CoA carboxylase gene (45). In addition, PP1 activity may be modulated by phosphorylation (46) or through its association with proteins such as NIPP-1 (47) and p53BP2 (47). PP1 activity in the nucleus, as well as PP1alpha protein, is elevated in rat hepatoma cells (48). It has recently been noted that the subnuclear partitioning of CKII into various fractions is altered during development (49) and regeneration (50). It may be that interaction between CKII or PP1 and Sp1 is modulated by subnuclear compartmentalization or that other factors are able to restrict or enhance their interaction. Whatever the mechanism by which overall Sp1 phosphorylation levels are modulated, it does appear to be a critical point in the regulation of cellular function.


FOOTNOTES

*   This work was supported in part by a grant from the National Cancer Institute of Canada with funds provided by the Canadian Cancer Society.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
   These authors contributed equally to this work.
par    Supported by a fellowship from the Ontario Cancer Treatment and Research Foundation.
Dagger Dagger    Research Scientist of the National Cancer Institute of Canada. To whom correspondences should be addressed: Cancer Research Laboratories, 3rd Floor Botterell Hall, Queen's University, Kingston, Ontario, Canada K7L 3N6. Tel.: 613-545-6751; Fax: 613-545-6830; E-mail: MUELLERC@POST.QUEENSU.CA.
1   The abbreviations used are: CKII, casein kinase II; DTT, dithiothreitol; PAGE, polyacrylamide gel electrophoresis; aa, amino acid(s).
2   Manuscript in preparation.

ACKNOWLEDGEMENTS

We thank Kari Newcombe for excellent technical assistance and Dr. David Litchfield for providing purified casein kinase II as well as antibodies for the casein kinase II subunits. We also acknowledge the gift of pPacSp1 from Dr. Robert Tjian.


REFERENCES

  1. Briggs, M. R., Kadonaga, J. T., Bell, S. P., and Tjian, R. (1986) Science 234, 46-52
  2. Kadonaga, J. T., Carner, K. C., Masiarz, F. R., and Tjian, R. (1987) Cell 51, 1079-1090 [Medline] [Order article via Infotrieve]
  3. Courey, A. J., and Tjian, R. (1988) Cell 55, 887-898 [Medline] [Order article via Infotrieve]
  4. Buckler, A. J., Chang, D. D., Graw, S. L., Brook, J. D., Haber, D. A., Sharp, P., and Housman, D. E. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 4005 [Abstract]
  5. Kadonaga, J. T., Courey, A. J., Ladika, J., and Tjian, R. (1988) Science 242, 1566-1570 [Medline] [Order article via Infotrieve]
  6. Courey, A. J., Holtzman, D. A., Jackson, S. P., and Tjian, R. (1989) Cell 59, 827-836 [Medline] [Order article via Infotrieve]
  7. Hoey, T., Weinzierl, R. O., Gill, G., Chen, J.-L., Dynlacht, B. D., and Tjian, R. (1993) Cell 72, 247-260 [Medline] [Order article via Infotrieve]
  8. Pascal, E., and Tjian, R. (1991) Genes Dev. 5, 1646-1656 [Abstract]
  9. Lee, J. S., Galvin, K. M., and Shi, Y. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 6145-6149 [Abstract]
  10. Merika, M., and Orkin, S. H. (1995) Mol. Cell. Biol. 15, 2437-2447 [Abstract]
  11. Liu, F., and Green, M. R. (1994) Nature 368, 520-525 [CrossRef][Medline] [Order article via Infotrieve]
  12. Kingsley, C., and Winoto, A. (1992) Mol. Cell. Biol. 12, 4251-4261 [Abstract]
  13. Hagen, G., Muller, S., Beato, M., and Suske, G. (1992) Nucleic Acids Res. 20, 5519-5525 [Abstract]
  14. Hagen, G., Muller, S., Beato, M., and Suske, G. (1994) EMBO J. 13, 3843-3851 [Abstract]
  15. Kadonaga, J. T., Jones, K. A., and Tjian, R. (1986) Trends Biochem. Sci. 11, 20-23 [CrossRef]
  16. Kim, S. J., Onwuta, U. S., Lee, Y. I., Li, R., Botchan, M. R., and Robbins, P. D. (1992) Mol. Cell. Biol. 12, 2455-2463 [Abstract]
  17. Udvadia, A. J., Rogers, K. T., Higgins, P. D. R., Murata, Y., Martin, K. H., Humphrey, P. A., and Horowitz, J. M. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 3265-3269 [Abstract]
  18. Udvadia, A. J., Templeton, D. J., and Horowitz, J. M. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 3953-3957 [Abstract/Free Full Text]
  19. Murata, Y., Kim, H. G., Rogers, K. T., Udvadia, A. J., and Horowitz, J. M. (1994) J. Biol. Chem. 269, 20674-20681 [Abstract/Free Full Text]
  20. Merchant, J. L., Shiotani, A., Mortensen, E. R., Shumaker, D. K., and Abraczinskas, D. R. (1995) J. Biol. Chem. 270, 6314-6319 [Abstract/Free Full Text]
  21. Jensen, D. E., Rich, C. B., Terpstra, A. J., Farmer, S. R., and Foster, J. A. (1995) J. Biol. Chem. 270, 6555-6563 [Abstract/Free Full Text]
  22. Inagaki, Y., Truter, S., Tanaka, S., Di Liberto, M., and Ramirez, F. (1995) J. Biol. Chem. 270, 3353-3358 [Abstract/Free Full Text]
  23. Li, J.-M., Nichols, M. A., Chandrasekharan, S., Xiong, Y., and Wang, X.-F. (1995) J. Biol. Chem. 270, 26750-26753 [Abstract/Free Full Text]
  24. Kim, H. S., Lee, J. K., and Tsai, S. Y. (1995) Mol. Endocrinol. 9, 178-186 [Abstract]
  25. Gunther, M., Frebourg, T., Laithier, M., Fossar, N., Bouziane-Ouartini, M., Lavialle, C., and Brison, O. (1995) Mol. Cell. Biol. 15, 2490-2499 [Abstract]
  26. Biggs, J. R., Kudlow, J. E., and Kraft, A. S. (1996) J. Biol. Chem. 271, 901-906 [Abstract/Free Full Text]
  27. Jackson, S. P., and Tjian, R. (1988) Cell 55, 125-133 [Medline] [Order article via Infotrieve]
  28. Jackson, S. P., MacDonald, J. J., Lees-Miller, S. P., and Tjian, R. (1990) Cell 63, 155-165 [Medline] [Order article via Infotrieve]
  29. Leggett, R. W., Armstrong, S. A., Barry, D., and Mueller, C. R. (1995) J. Biol. Chem. 270, 25879-25884 [Abstract/Free Full Text]
  30. Gorski, K., Carneiro, M., and Schibler, U. (1986) Cell 47, 767-776 [Medline] [Order article via Infotrieve]
  31. Maire, P., Wuarin, J., and Schibler, U. (1989) Science 244, 343-346 [Medline] [Order article via Infotrieve]
  32. Mueller, C. R., Maire, P., and Schibler, U. (1990) Cell 61, 279-291 [Medline] [Order article via Infotrieve]
  33. Lichsteiner, S., and Schibler, U. (1989) Cell 57, 1179-1187 [Medline] [Order article via Infotrieve]
  34. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic Acids Res. 17, 6419 [Medline] [Order article via Infotrieve]
  35. Lin, A., Frost, J., Deng, T., Smeal, T., Al-Alawi, N., Kikkawa, U., Hunter, T., Brenner, D., and Karin, M. (1992) Cell 70, 777-789 [Medline] [Order article via Infotrieve]
  36. Hagiwara, M., Alberts, A., Brindle, P., Meinkoth, J., Feramisco, J., Deng, T., Karin, M., Shenolikar, S., and Montminy, M. (1992) Cell 70, 105-113 [Medline] [Order article via Infotrieve]
  37. Cohen, P. (1989) Annu. Rev. Biochem. 58, 453-508 [CrossRef][Medline] [Order article via Infotrieve]
  38. Kriwacki, R. W., Schultz, S. C., Steitz, T. A., and Caradonna, J. P. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 9759-9763 [Abstract]
  39. Pavletich, N. P., and Pabo, C. O. (1991) Science 252, 809-817 [Medline] [Order article via Infotrieve]
  40. Miner, J. H., and Wold, B. J. (1991) Mol. Cell. Biol. 11, 2842-2851 [Medline] [Order article via Infotrieve]
  41. Pinna, L. A. (1990) Biochim. Biophys. Acta 1054, 267-284 [Medline] [Order article via Infotrieve]
  42. Seldin, D. C., and Leder, P. (1995) Science 267, 894-897 [Medline] [Order article via Infotrieve]
  43. Udvadia, A. J., Rogers, K. T., Higgins, P. D., Murata, Y., Martin, K. H., Humphrey, P. A., and Horowitz, J. M. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 3265-3269 [Abstract]
  44. Chen, L. I., Nishinaka, T., Kwan, K., Kitabayashi, I., Yokoyama, K., Fu, Y.-H. F., Grunwald, S., and Chiu, R. (1994) Mol. Cell. Biol. 14, 4380-4389 [Abstract]
  45. Daniel, S., Zhang, S., DePaoli-Roach, A. A., and Kim, K.-H. (1996) J. Biol. Chem. 271, 14692-14697 [Abstract/Free Full Text]
  46. Villa-Moruzzi, E., and Puntoni, F. (1996) Biochem. Biophys. Res. Commun. 219, 863-867 [CrossRef][Medline] [Order article via Infotrieve]
  47. Van Eynde, A., Wera, S., Beullens, M., Torrekens, S., Van Leuven, F., Stalmans, W., and Bollen, M. (1995) J. Biol. Chem. 270, 28068-28074 [Abstract/Free Full Text]
  48. Saadat, M., Mizuno, Y., Takizawa, N., Kakinoki, Y., and Rikuchi, R. (1995) Cancer Lett. 94, 165-170 [CrossRef][Medline] [Order article via Infotrieve]
  49. Gruppuso, P. A., and Boylan, J. M. (1995) J. Cell. Biochem. 58, 65-72 [Medline] [Order article via Infotrieve]
  50. Pancetti, F., Bosser, R., Itarte, E., and Bachs, O. (1996) Biochem. Biophys. Res. Commun. 218, 35-39 [CrossRef][Medline] [Order article via Infotrieve]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.