Báo cáo Y học: Nuclear proteins that bind to metal response element a (MREa) in the Wilson disease gene promoter are Ku autoantigens and the Ku-80 subunit is necessary for basal transcription of the WD gene ppt

Nuclear proteins that bind to metal response element a MREain the Wilson disease gene promoter are Ku autoantigens and the Ku-80 subunit is necessary for basal transcription of the WD ge

Nuclear proteins that bind to metal response element a (MREa)

in the Wilson disease gene promoter are Ku autoantigens and the Ku-80 subunit is necessary for basal transcription of the WD gene

Won Jun Oh1, Eun Kyung Kim1, Jung Ho Ko1, Seung Hee Yoo1, Si Houn Hahn2and Ook-Joon Yoo1

1

Department of Biological Sciences, Korea Advanced Institute of Science and Technology (KAIST), Taejon Korea;

2 Department of Pediatrics, Ajou University School of Medicine, Suwon, Korea

Wilson disease (WD), an inherited disorder affecting copper

metabolism, is characterized by hepatic cirrhosis and

neur-onal degeneration, which result from toxic levels of copper

that accumulate in the liver and brain, respectively We

reported previously that the 1.3-kb promoter of the WD

gene contains four metal response elements (MREs) Among

the four MREs, MREa plays the most important role in the

transcriptional activation of the WD promoter

Electro-phoretic mobility shift assays (EMSAs) using synthetic

MREa and an oligonucleotide containing the binding site

for transcription factor Sp1 revealed the presence of nuclear

factors that bind specifically to MREa Two MREa-binding

proteins of 70 and 82 kDa were purified using avidin–biotin

affinity chromatography Amino acid sequences of peptides

from each protein were found to be highly homologous to

the Ku proteins Immunoblot analysis and EMSAs showed

that the MREa-binding proteins are immunologically

rela-ted to the Ku proteins To study further the functional

significance of these Ku-related proteins in transcriptional regulation of the WD gene, we performed RNA interference (RNAi) assays using a Ku-80 inverted-repeat gene to inhibit expression of the Ku-80 gene in vivo Results of the RNAi assays showed that expression of the Ku-80 protein was suppressed in transfected cells, which in turn led to the suppression of the WD gene In addition, a truncated Ku-80 (DKu-80) mutant inhibited WD promoter activity in HepG2 cells in a dominant-negative manner We also found that

WD promoter activity was decreased in Xrs5 cells, which, unlike the CHO-K1 cells, are defective in the Ku-80 protein When Ku-80 cDNA was transfected into Xrs5 and CHO cells, WD promoter activity was recovered only in Xrs5 cells Taken together, our findings suggest that the Ku-80 subunit

is required for constitutive expression of the WD gene Keywords: ATP7B gene; Ku antoantigen; RNA interference; site-directed mutagenesis

Wilson disease (WD) is an autosomal recessive disorder

characterized by defect in copper transport Hepatic

cirrhosis and neuronal degeneration are the most

debilita-ting symptoms of WD and are caused by the impairment of

biliary copper excretion and the accumulation of toxic

concentration of copper The WD gene product shares a

high degree of sequence similarity with the

cation-trans-porting P-type ATPases [1–5] and functions in the binding

and translocation of copper [6]

Interestingly, multiple copies of metal-response elements

(MREs) are located in the 1.3-kb promoter of the WD gene

[7] Five or more nonidentical MREs are present in the

5¢-flanking region of the vertebrate metallothionein (MT)

gene [8,9] and are believed to mediate the transcriptional

activation of the MT gene by heavy metals [8,10–12] and

oxidative stress [13,14] The MREs consist of 12-base pair

sequences that contain a seven-nucleotide core sequence (TGCRCNC) surrounded by less well-conserved flanking nucleotides [15] MTs are small (6–7 kDa), cysteine-rich, metal-binding proteins that function in the maintenance of metal homeostasis and detoxification by forming strong complexes with several types of metal ions [16,17] Given the existence of MREs in the promoter of the WD gene, it seems plausible that metal ions function in the transcriptional regulation of the WD gene via the adjacent MREs The Ku protein is associated with a DNA-dependent protein kinase [18] and is involved in double-stranded DNA break repair, V(D)J recombination, and telomere mainten-ance [19–21] Sequence-specific DNA binding of Ku family proteins has been reported also for the genes encoding small nuclear RNA, the T-cell receptor, the transferrin receptor, collagenase, and heat shock proteins (HSPs) [22–28] Recently, it was shown that overexpression of the Ku-80 subunit suppresses the MT-I gene, which chelates heavy metals in fibroblast cells [29]

In this study, we used mutational analysis to show that MREa ()434 to )438) is the key transcriptional regulatory element of the WD gene We then used electrophoretic mobility shift assays (EMSAs) to detect an MREa-binding activity in HepG2 cell nuclear extracts We purified and characterized the MREa-binding activity, which consisted

of two polypeptides of approximately 70 and 82 kDa N-terminal and internal amino acid sequencing and immuno analyses revealed that MREa-binding proteins are either

Correspondence to O.-J Yoo, Department of Biological Sciences,

Korea Advanced Institute of Science and Technology, 373-1

Kusong-dong, Yusong-gu, Taejon 305-701, Korea Fax: +82 42 869 8160,

Tel.: +82 42 869 2626, E-mail: ojyoo@mail.kaist.ac.kr

Abbreviations: WD, Wilson Disease; MRE, metal response element;

MT, metallothionein; RNAi, RNA interference; IR, inverted repeats;

EMSA, electrophoretic mobility shift assay; ZAP, zinc activated

proteins.

(Received 2 November 2001, revised 5 February 2002,

accepted 4 March 2002)

identical or closely related to the human Ku proteins We

also performed RNA interference (RNAi) assays and assays

using a truncated Ku-80 mutant to study the regulation of

WD gene expression by the Ku proteins Finally, we assessed

the activity of the WD promoter in the Ku-80-deficient Xrs5

and Ku-80 transfected Xrs5 cell line Our results suggest that

the Ku proteins that bind specifically to MREa play an

important role as a basal regulator of WD gene

transcrip-tion

M A T E R I A L S A N D M E T H O D S

Cell culture

The human hepatoma cell line HepG2 was obtained from

KRIBB and grown in Dulbecco’s modified Eagle’s medium

(pH 7.4) containing 10% fetal bovine serum and 1·

antibiotic–antimycotic solution (Life Technologies, Inc.)

The hamster ovary cell line CHO-K1 and the

Ku-80-deficient cell line Xrs-5 were obtained from the American

Type Culture Collection and maintained Ham’s F12K

medium and alpha minimum essential medium,

respect-ively, that containing 10% fetal bovine serum and 1·

antibiotic–antimycotic solution

Human WD promoter-reporter gene constructs

A 1.6-kb construction of the WD promoter, named

pPWD-Luc, was described previously [7] Trinucleotide mutations in

each of the four MREs of the WD gene promoter were

constructed using the PCR-based quick change site-directed

mutagenesis method (Stratagene) The following

oligonucle-otides were used in the MRE mutagenesis, with the

trinucleotide mutations indicated in lowercase letters

(underlined bases denote the MRE core sequence): MREa

mut, 5¢-GGGCGCCaatGCCCCCGTTCC-3¢; MREe mut,

5¢-GGCCATTGGCTGGCCTTaatGCACAGCGGATCG

ATTTTC-3¢; MREc mut, 5¢-CCAGTACAGTGTCGG

AGCattCCAGCGCGAGGTGGCCG-3¢; MREd mut,

5¢-CGGGAGGACGGCGGCGCattACTTTGAATCAT

CCGTG-3¢ Mutations in the MREs were identified and

confirmed by automated fluorescent DNA sequencing

(Perkin-Elmer 377)

Transfection and luciferase assays

DNA transfections were performed according to the

procedure provided by Life Technologies HepG2 cells

were cultured at 40–60% confluence in 6-well dishes and

transfected with 1 lg of DNA mixture containing various

Ku cDNAs, WD promoter report construct (pPWD-Luc)

[7], and pSVb-gal which served as an internal control for

transfection efficiency The transfected cells were cultured

for a further 24 or 48 h, and the expression levels of

luciferase and b-galactosidase were determined Luciferase

activity was analysed with the Promega luciferase assay kit

The cells were harvested in reporter lysis buffer, and the

lysate was spun in a microcentrifuge for 15 s

Chemilumi-nescence was measured with a luminometer (Berthold), and

the b-galcatosidase activities were determined as described

[30] When determination of the exact number of transfected

cells was required, transfected cells were distinguished from

nontransfected cell by visualization using X-Gal [31]

Transfections using the above constructs were repeated three or more times, and the average result is presented Preparation of HepG2 nuclear extracts

Nuclear extracts were prepared as described previously [32], with slight modifications HepG2 cells (9.4· 106cells per dish) were washed with ice-cold NaCl/Piand scraped off the dish into 5 mL ice-cold NaCl/Pi Cell pellets were resus-pended in 5 vol hypotonic buffer (10 mM Hepes pH 7.9, 1.5 mMMgCl2, 10 mMKCl, 0.2 mM phenylmethanesulfo-nyl fluoride, 0.5 mMdithiothreitol), and the cell suspensions were homogenized with a Dounce homogenizer (20 strokes, type B pestle) Nuclear proteins were suspended in 1 vol low salt buffer (20 mMHepes pH 7.9, 25% glycerol, 1.5 mM MgCl2, 20 mM KCl, 0.2 mM EDTA, 0.2 mM phenyl-methanesulfonyl fluoride, 0.5 mM dithiothreitol), followed

by the addition of 4MKCl to a final concentration of 0.3M Nuclear suspensions were stirred for 30 min on ice and dialysed against dialysis buffer (20 mMHepes pH 7.9, 20% glycerol, 100 mM KCl, 0.2 mM EDTA, 0.2 mM phenyl-methanesulfonyl fluoride, 0.5 mMdithiothreitol) for 4 h at

4C Nuclear extracts were frozen in aliquots at)70 C EMSAs

The double-stranded MREa probes were end-labelled with [c-32ATP] (Amersham) and T4 polynucleotide kinase HepG2 cell nuclear proteins (10 lg) were incubated in reaction buffer containing 17 mM Hepes pH 7.9, 32 mM Tris/HCl pH 7.8, 13% glycerol, 25 mM KCl, 0.8 mM dithiothreitol, 4 lg poly(dI-dC), and 2–4 fmol 32P end-labelled, double-stranded MRE oligonucleotides (50 000 c.p.m per reaction) in a total volume of 20 lL, and the reaction mixture was incubated for 30 min at 25C For competition experiments, a molar excess of unlabelled MRE oligonucleotides were added to the binding reaction and incubated for 15 min prior to the addition of the labelled MRE DNA probe as specified Protein–DNA complexes were separated electrophoretically in a 4% polyacrylamide gel (acrylamide : bisacrylamide, 30 : 0.8 in 0.5· Tris/borate/EDTA) After electrophoresis, the gel was dried, and the32P-labelled protein–DNA complexes were detected by autoradiography For supershift EMSAs, nuclear extracts from HepG2 cells were incubated with a mixture of Ku-70/-80 monoclonal antibodies (mAb; 0.2 lg

or 1 lg; Clone 162, Neomarkers Co.) for 5 min, followed by incubation with 32P-labelled MRE oligonucleotides for

30 min at 25C

The oligonucleotide sequences used as the 32P-labelled probes and competitors were: MREa, 5¢-GGGCGCC TGCGCCCCCGTTCC-3¢ ()441 to )421); MREa mut1, 5¢-GGGCGCCAATGCCCCCGTTCC-3¢; MREa mut2 :

denote the functional core of the MREs and the Sp1 binding element, and the mutated bases are indicated by italic type)

Western blot analysis The MREa-binding proteins purified from HepG2 cells by the avidin–biotin method and lysates (30 lg) of cells

transfected with the Ku-80 inverted-repeat (IR) gene were

separated by SDS/PAGE on a 10% polyacrylamide gel

Proteins were transferred to a nitrocellulose filter using

standard procedures The membranes were then incubated

with mAbs to the Ku-70 (Clone N3H10, Neomarkers Co.)

and Ku-80 (Clone 111, Neomarkers Co.) proteins and with

polyclonal antibodies to the WD protein (transfection

experiments only) The antigen–antibody complexes were

detected using a second antibody conjugated to horseradish

peroxidase and the ECL detection system (Amersham)

South-Western blotting

The South-Western blotting assay was carried out as

described by Dikstein [33] HepG2 nuclear extracts (30 lg)

were resolved by SDS/PAGE on a 10% polyacrylamide gel

and electroblotted onto a nitrocellulose membrane After

transfer, the filter was incubated in blocking solution (5%

nonfat milk, 50 mMTris/HCl pH 7.5, 100 mMNaCl, 1 mM

EDTA, and 1 mMdithiothreitol) for 1 h at room

tempera-ture and then washed twice with TNE-100 buffer (10 mM

Tris/HCl pH 7.5, 100 mM NaCl, 1 mM EDTA, 1 mM

dithiothreitol) The filter-bound proteins were then

dena-tured by incubation in denaturing buffer (7Mguanidine/

HCl, 50 mMTris/HCl pH 8.0, 50 mMdithiothreitol, 2 mM

EDTA, and 0.25% nonfat milk) for 1 h at room

temper-ature and rentemper-atured by incubation in 50 mM Tris/HCl

pH 7.5, 100 mMNaCl, 2 mMdithiothreitol, 2 mMEDTA,

0.1% NP-40, 0.25% nonfat milk for 20 h at 4C To

measure the MREa-binding activity of the filter-bound

proteins, the filter was preincubated with TNE-100

con-taining 10 lgÆmL)1poly(dI-dC) for 1 h at room

tempera-ture, and then 32P end-labelled, double-stranded MREa

oligonucleotide was added (2· 106 c.p.m.ÆmL)1) to the

incubation mixture After 1 h of incubation, the filter was

washed three times with TNE-100 and subjected to

autoradiography

Purification of Ku proteins

The Ku proteins that bound to MREa were purified from

HepG2 nuclear extracts using the avidin–biotin method of

Otsuka et al [34] with slight modifications An MREa

probe was composed of a chemically synthesized

oligo-nucleotide with a biotin head at the 5¢-end HepG2 nuclear

extracts were incubated with avidin–agarose beads (Sigma)

at 4C for 30 min to eliminate materials in the extracts that

bind nonspecifically to the beads After removing the beads

by filtration, 200 lg of the nuclear extract proteins were

incubated in reaction buffer containing 18 mMHepes/KOH

pH 7.9, 10% glycerol, 40 mM KCl, 2 mM MgCl2, 10 mM

dithiothreitol, 125 lgÆmL)1 poly(dI-dC), and 50 pmol of

the biotinylated MREa probe In addition, 50 pmol

MREaMut1 oligonucleotides were added to the binding

reaction as a substitute for poly(dI-dC) for the competition

assay The binding reaction was carried out for 30 min at

25C After elinimating by centrifugation denatured

pro-teins generated during the binding reaction, the mixture was

combined with 20 lL avidin–agarose beads and incubated

at room temperature for 30 min with shaking The beads

were washed successively with 0.5 mL 0.1MKCl buffer III

(20 mM Hepes/KOH pH 7.9, 1 mM dithiothreitol, 20%

glycerol, 0.01% NP-40) and 0.5 mL 0.2 KCl buffer III,

and then the proteins were extracted from the beads by incubation in 0.5MKCl buffer III (0.5 mL) for 30 min at

4C The extracts were concentrated to 10% of the original volume of the nuclear extract by trichloroacetic acid precipitation The concentrated sample was analysed by SDS/PAGE (10% polyacrylamide)

Amino acid sequencing of purified MREa-binding

Ku proteins The Ku proteins described in the previous section were purified from 700 lg HepG2 cells, which yielded 0.5–1 lg

of Ku protein After incubation of the purified Ku proteins

in SDS sample buffer for 15 min at 37C, the proteins were separated by SDS/PAGE and electrotransferred onto poly(vinylidene difluoride) membrane (Sequi-blot; Bio-rad) Internal peptide sequencing of the 70-kDa protein was performed according to the Current Protocol method [35] Concentrated Ku proteins were digested with 50 lL 70% formic acid at 37C for 48 h The peptide fragments generated by the formic acid treatment were lyophilized completely and separated by electrophoresis The Ku protein fragments were visualized in the gel by Coomassie blue staining, the bands were sliced out of the gel and subjected to automated amino acid sequencing

Construction of the IR Ku-80 gene and the truncated Ku-80 (DKu80)

For the double-stranded RNAi assay, the Ku-80 IR vector was constructed according to the method of Tavernarakis

et al [36] with slight modification The Ku-80 cDNA (2.2 kb) was amplified by PCR from a Ku-80 clone (a generous gift from Y Sung-Han, Medical College of Georgia, Augusta) The amplified Ku-80 cDNA was digested with EcoRI at a site downstream of the coding region, ligated to generate the IR, digested with BamHI at a site upstream of the coding region, and then inserted into the expression vector pcDNA 3.1(+) (Fig 7A) The IR orientation of Ku-80 was identified by automated fluores-cent DNA sequencing (Perkin-Elmer 377) We also con-structed a truncated Ku-80 gene, in which the region spanning amino acids 351–416 was deleted; this was used for transfections into HepG2 cells as described above Competitive RT-PCR

Competitive RT-PCR assay was performed to determine the amount of transcript of WD gene in HepG2 cells transfected with Ku 80 IR After the Ku 80 IR vector (1 lg) was transfected as described previously, total RNA was isolated from time-dependent HepG2 cells using the RNeasy kit (Qiagen) Reverse transcription was performed

at 65C for 5 min and 37 C for 60 min followed by 5 min denaturation at 95C in a total volume of 50 lL of reaction mixture using the RT-PCR kit (Stratagene) PCR amplifi-cation was performed using LA Taq (Takara, Japan) in a total volume of 50 lL containing 55 pg of competitor At this concentration, the band intensity of the amplified product from the competitor was same as that of the product from the endogenous WD gene Amplification conditions were: 95C for 3 min followed by 18 cycles of

95C for 45 s, 55 C for 45 s, 72 C for 45 s, and 72 C for

5 min (Perkin-Elmer PCR system 9700) After competitive

RT-PCR, a 10-lL aliquot was electrophoresed in a 1%

agarose gel and the bands were visualized by staining with

ethidium bromide The intensities of the amplified

frag-ments were quantified by densitometric analysis As an

internal control 791 base pairs of a-actin was amplified

using forward and reverse primers The primers used for

competitive RT-PCR and predicted product sizes are given

in Table 1

R E S U L T S

Effect of MREa mutations on WD gene promoter

activity in HepG2 cells

The relative positions of the MREs within the promoter

region of the WD gene are indicated in Fig 1 The four

MREs are located in the proximal region of the WD gene

promoter between)434 and +114, with MREa and MREe

in the forward orientation, and MREc and MREd in

reverse orientation

The effects of the mutiple MREs on promoter activity of

the MT gene have been shown to be dependent on the

position of the individual MRE [37] For this reason, we

assessed the functional contributions of MREa, MREe,

MREc, and MREd on WD gene promoter activity Trinucleotide mutations were generated in the core sequence

of each MRE (Fig 1) and these WD gene promoter variants were fused to a luciferase reporter gene HepG2 cells were transfected with these reporter genes and luciferase activity in the cell extracts was measured after

48 h The mutaton within MREa resulted in a marked decrease in promoter activity (0.5% of wild-type); however, mutations in the other MREs showed no significant changes

in luciferase activity compared to the native WD promoter These results show that MREa has the greatest effect on

WD gene promoter activity Several studies have reported that MREa in the promoter of the MT gene possesses transcriptional regulatory activity and that a variety of nuclear transcription factors interact specifically with MREa [11,33,38,39] These findings suggest that transcrip-tional activator proteins that regulate expression of the WD gene may bind to MREa

Identification of proteins that interact specifically with MREa

We next performed EMSAs to determine whether MREa-binding proteins are present in HepG2 cell nuclear extracts For the EMSAs we used a32P-labelled synthetic

oligonu-Table 1 Oligonucleotide primers and conditions used for competitive RT-PCR analysis.

Target

gene

Forward primer Reverse primer

PCR products (base pairs) Annealing

temperature (C) Target Competitor

5¢-CCGGTCAGCCAGCTGCTG

5¢-TACATGGTGGTGCCGCCAGA

Fig 1 WD gene promoter fragments used in the transfection analyses and the effects of MRE mutations on WD gene promoter activity (A) Schematic representation of the WD promoter ( )1265 to +335) is shown with the position of the MREs indicated (arrow direction conveys MRE orientation) The sequences of the MRE mutants used in this study are shown (B) HepG2 cells were transfected with wild-type and MRE mutant

WD gene promoters fused to a luciferase reporter gene in the pGL2 reporter construct [38] The results were normalized using b-galactosidase activity Results are mean ± SD pGL indicates the luciferase activity in cells that had been transfected with the pGL as a negative control pGL promoter (pGLpro.) that contains the SV40 promoter is used as a positive control Luciferase activity for each construct was normalized to b-gal activity, and the relative increases were calculated as the ratio of normalized activity in MREa/mutant transfected cells to that in pGLpro transfected cells Data represent the mean ± SD of three independent transfection assays for each construct N, Nucleotides; R, purine; +1, transcription start site.

cleotide that contained a copy of MREa (Fig 2A) We also

performed competition experiments with various

oligonu-cleotide competitors to determine whether the MREa–

protein complex displays sequence specificity Binding of the

HepG2 nuclear proteins to MREa was inhibited by

the addition of a molar excess of unlabelled MREa to the

binding reaction (Fig 2B, lanes 3, 4, 5) It has been shown

that the MREs overlap with a potential binding site for the

transcription factor Sp1 [40] However, addition of a molar

excess of oligonucleotide that contained the Sp1 binding site

did not affect formation of the MREa–protein complex

(Fig 2B, lanes 6, 7) We also constructed two mutant

MREa oligonucleotides to analyse the effect of MREa

sequence on formation of the MREa–protein complex

Mutant competitors were introduced either outside (MREa

mut2) or within (MREa mut1) the MREa consensus

sequence Neither MREaMut1 nor MREaMut2 affected

MREa–protein complex formation (Fig 2B, lanes 8, 9) In

addition to MREa-binding complexes, several other bands

were detected, possibly formed by nonspecific protein–

DNA complexes, as they were not competed out by any of

the competitors used in Fig 2B These results suggest that

both the proximal and the consensus sequences of MREa

function in sequence-specific interaction with the HepG2

nuclear proteins

Characterization and purification of MREa-binding

factors

To determine the molecular mass of the MREa-binding

proteins, we performed South-Western analysis A 70-kDa

polypeptide was detected only when a32P-labelled MREa

oligonucleotide was used as a probe (Fig 3A) We also

observed another band of 82 kDa when the MREa-protein

band from the EMSA was excised from the native gel and

resolved by SDS/PAGE (unpublished data) We purified

the MREa-binding proteins using the avidin–biotin affinity

method [34] HepG2 nuclear proteins were incubated with a

biotinylated MREa oligonucleotide and trapped by avidin–

agarose beads The proteins were extracted from the avidin–

agarose beads by washing with a buffer containing 0.5M

KCl (see Materials and methods for details) Two major

polypeptides of 70 and 82 kDa were detected by SDS/

PAGE (Fig 3B, lane 3) Also, there was no difference in the quantity of purified MREa-binding proteins whether the MREa mut1 oligonucleotide was present in the reaction or not (Fig 3C) This result confirmed that these proteins interacted with MREa in sequence-dependent manner, as shown in Fig 2B (lanes 6, 7)

To perform N-terminal amino-acid sequencing on the two proteins, the purified protein bands were blotted onto a polyvinylidene difluoride membrane, and the individual proteins were subjected to N-terminal sequencing The N-terminus of the 70-kDa protein was blocked by modi-fication Therefore, the protein was digested with 70% formic acid, which cleaves Asp–Pro (D–P) peptide bonds (located at amino-acid positions 342 and 343 in the 70-kDa protein), and the resulting peptide fragments were se-quenced The sequences of the peptides from the 82 kDa and 70-kDa proteins were found to be almost (90%) identical to peptide sequences from the Ku-80 and Ku-70 subunits of the human Ku autoantigen (Fig 4) To confirm that the MREa-binding protein is homologous with the Ku proteins, immuonoblot analyses and supershift EMSAs were performed using mAbs to Ku-70 and Ku-80 The purified 70- and 82-kDa MREa-binding proteins reacted with antibodies to Ku-70 (Clone N3H10) and Ku-80 (Clone 111), respectively (Fig 5A) EMSAs performed with HepG2 nuclear extract in the presence or absence of a mixture of Ku-70/-80 mAbs (Clone 162) revealed that the MREa–protein complex was supershifted when the anti-bodies were added to the binding reaction (Fig 5B) These results indicate the two MREa-binding proteins are closely related or identical to the Ku proteins

The effects of RNAi of the Ku-80 subunit and the truncated Ku-80 mutant on WD gene expression

To examine the effects of the Ku protein on the regulation

of WD gene expression, we measured the amount of WD protein expressed in HepG2 cells where the Ku-70 or/and Ku-80 subunits were transiently overexpressed We observed no significant difference in the amount of WD protein expressed in wild-type cells and cells that overex-pressed the Ku proteins (unpublished data), consistent with previous studies showing that WD proteins are expressed

Fig 2 Detection of nuclear proteins that bind

specifically to MREa (A)32P-labelled

double-stranded oligonucleotide (Oligo) containing

MREa was incubated without (lane 1) and

with (lane 2) HepG2 Nuclear extracts (10 lg).

(B) Competition experiments were performed

in the absence (C, lane 2) or presence of

varying concentrations of unlabelled MREa

(lanes 3–5), Sp1 (lanes 6–7), MREaMut1 (lane

8), and MREaMut2 (lane 9) Lane 1, oligo

only The solid arrowhead indicates the

spe-cific MREa–protein complex and is competed

away with unlabelled MREa The open

arrowhead denotes a nonspecific band.

Fig 3 Characterization and purification of the MREa-binding protein (A) South-Western analysis of the MREa-binding protein Nuclear extracts (30 lg) from Cos cells (lane1) and HepG2 cells (lane 2) were separated by SDS/PAGE (10% polyacrylamide) followed by electrotransfer to a nitrocellulose membrane Blots were incubated with a32P end-labelled MREa probe as described in Materials and methods and subjected to autoradiography A 70-kDa polypeptide is indicated by solid arrowhead (B) SDS/PAGE analysis of MREa-binding proteins purified by the biotin-avidin method (see Materials and methods) Lane 1, proteins eluted by 0.1 M KCl; lane 2, proteins eluted by 0.2 M KCl; lane 3, proteins eluted by 0.5 M KCl; lane 4, avidin–agarose beads (B) After elution by 0.5 M KCl (C) Confirmation of sequence-specific binding of MREa to the two proteins (see Materials and methods) Purified proteins without (lane 1) and with (lane 2) 50 pmol of mutant competitor (MREa mut 1) The positions of molecular size marker (SM) are indicated at the left of each panel The purified protein bands are indicated by solid arrowheads.

Fig 4 Microsequencing analysis of the MREa-binding proteins.

(A) The N-terminal sequence of the 82-kDa band was identical to

amino-acid positions 6–16 of the Ku-80 protein (B) The 70-kDa band

did not yield an N-terminal sequence, probably due to modification.

Internal sequencing of a peptide generated by formic acid digestion of

the 70 kDa protein was identical to an internal sequence of the Ku-70

protein X, Residues not properly identified by the sequencing

proce-dure; query, sequenced amino acids.

Fig 5 Confirmation that MREa-binding proteins are related to the Ku proteins (A) Immunoblot analysis of proteins eluted from the avidin– agarose beads as described in Fig 3B Both the 70- and 82-kDa pro-teins interacted with mAb N3H10B Ku-70) and mAb 111 (anti-Ku-80) The positions of molecular size markers (SM) are indicated at the left The Ku-70 and Ku-80 proteins are indicated by solid arrow-heads (B) Supershift EMSA of the MREa-Ku complexes with mAb

162 (anti-Ku-70/80 dimer) The 32 P end-labelled MREa probe used in Fig 2 was incubated with nuclear extracts from HepG2 for 5 min and then incubated for an additional 30 min either in the absence (lane 1)

or presence (lane 2, 3) of varying concentrations of mAb 162 Bands shifted by the antibodies are indicated by solid arrowheads.

constitutively [41,42] To interfere transiently with

expres-sion of the Ku-80 protein in vivo, we performed RNAi

assays (Fig 6A) Double-stranded RNAi is an effective

method for disrupting expression of specific genes in higher

organisms, especially Caenohabditis elegans [43–45]

Recently, it was reported that RNAi caused by in vivo

expression of IR versions of specific genes has advantages

over RNAi by directly introduced double-stranded RNA

[36] We constructed a Ku-80 IR vector that was able to

synthesize hairpin double-stranded RNA and transfected

the vector into HepG2 cells Immunoblot analysis showed

that expression levels of the Ku-80 and WD proteins were

significantly reduced from 0.5 to 12 h after transfection

(Fig 6B, lanes 2, 3, 4, 5), and then recovered to wild-type

levels after 24 h as cellular division proceeded (Fig 6B,

lanes 6, 7, 8) Ku-80 gene expression recovered before WD

gene expression, suggesting that the Ku-80 protein is a

constitutive transcriptional regulator of WD gene

expres-sion On the other hand, there was no significant change in

the concentration of Ku-70 protein, although it seemed that

there was a slight reduction from 0.5 to 12 h after

transfection, especially at 0.5 h In addition, the effect of

Ku-80 on the WD gene expression was examined by

competitive RT-PCR In consistency with the immunoblot

analysis result, the mRNA level of WD gene was

signifi-cantly diminished from 0.5 to 12 h after transfection and then completely restored to wild-type level by 40 h after transfection For further investigation on the role of Ku-80 protein in the basal transcription of WD disease gene, a truncated Ku-80 (DKu-80) cDNA, in which an internal region harbouring a part of the dimerization domain was deleted, was constructed as shown in Fig 7 The DKu-80 cDNA and/or Ku-70 were cotransfected with pPWD-Luc into HepG2 cells

There was no significant change in WD promoter activity

in Ku-70-transfected cells In contrast, the promoter activity was remarkably reduced to a 50% of normal activity in DKu-80-transfected and Ku70/DKu-80-transfected cells (Fig 7) It is implied that the lack of the dimerization domain in the DKu-80 cDNA decreased the WD promoter activity in a dominant-negative manner These results suggest that Ku-80 protein is an important factor in the regulation of WD promoter activity

Comparison of WD promoter activity in Ku-deficient Xrs-5 cells to the activity in CHO cells

To verify that Ku-80 is an essential regulatory factor in WD gene expression, pPWD-Luc was transfected into both a Ku-80 normal-cell line, CHO-K1, and a Ku-80-mutant cell

Fig 6 Construction of a Ku-80 IR gene for RNAi assays and the expression pattern of the WD and Ku-80 proteins after transfection (A) Ku-80 cDNA was amplified using two primers that introduced EcoRI and BamHI sites at the ends of the Ku-80 gene The EcoRI site was used to generate the IR, and the BamHI site was used to ligate the Ku-80 IR gene to pcDNA3.1(+) to yield pc80IR Expression of the Ku-80 IR gene was driven by the cytomegalovirus (CMV) promoter (B) Immunoblot analysis showing that the WD and Ku proteins were expressed in transiently transfected HepG2 cells HepG2 cells were transfected with 0.5 lg of pc80IR and 0.5 lg of the pRSVb-gal reporter gene After transfection, cells were incubated for 0.5, 4, 8, 12, 24, 30, 38 h and harvested in 200 lL lysis buffer Whole-cell lysates (30 lg) were resolved by SDS/PAGE (10% polyacrylamide), and immunoblotting was performed with the antibodies described in Fig 5A The control (C) was a whole cell lysate of untransfected HepG2 cells The WD protein is indicated by the open arrowhead, and the Ku-70 and -80 proteins are indicated by closed arrowheads In addition, the asterisks represent reappearance of the WD proteins after 24 h (upper panel) The levels of actin protein are not significantly different at all time points (lower panel) (C) WD mRNA expression as quantified by competitive RT-PCR in HepG2 cells transfected Ku-80 IR Results of each values were expressed as the following ratio: intensity of WD cDNA/intensity of competitor and then divided by intensity

of amplified actin cDNA PC, the result of competitive RT-PCR using mRNA isolated from nontransfected HepG2 cells as a positive control.

line, Xrs5 The activity of the WD gene promoter was

reduced by 50% in Xrs5 cells 24 h after transfection,

compared to WD gene promoter activity in CHO-K1 cells

(Fig 8A) To examine whether the WD promoter activity was recovered after expression of Ku-80 protein in Xrs5 and CHO cells, the Ku-80 cDNA and the pPWD-Luc were cotransfected After 48 h, the promoter activity in Xrs5 cells transfected with Ku-80 was recovered to about 27% higher than that observed in nontransfected Xrs-5 cells In contrast, there were no significant changes after transfection

of the Ku-80 in CHO cells (Fig 8B) These results clearly showed that the Ku-80 protein plays a key role in constitutive expression of the WD gene

D I S C U S S I O N

It has been proposed that the WD protein acts as a copper-specific pump that mediates the export of copper from the cytosol, similar to the P-type ATPase [14] On the basis of the fact that MREs mediate the transcriptional response of the MT gene to heav y metals [10,12], the presence of MREs

in the promoter of the WD gene suggests that MREs and their cognate binding proteins function in the regulation of

WD gene expression Because the T1 and C3 nucleotides within the MRE consensus heptamers are major determi-nants in the sequence-specific binding of zinc activated protein (ZAP) [39], and because single point mutations may not obliterate MRE function [46], we mutated the first three nucleotides within the core sequence of MREs and tested the effect of these mutations on WD gene transcription We found that of the four MREs, MREa was the most significant element in the transcriptional regulation of WD gene Previous studies report that MREa is also the most crucial MRE for transcription of the MT gene and that there is a relationship between the distance from each MRE

to the TATA box and their influence on promoter activity

of the MT gene, the proximal element MREa exhibiting the strongest effect In addition, it is known that function of distal MREs is dependent on MREa in proximal promoter region [46] In contrast with the above findings, MREa, the

Fig 7 Construction of DKu-80 protein and pattern of promoter activity

in HepG2 cells after transfection of DKu-80 The full-length wild-type

Ku-80 protein and the truncated Ku-80 (DKu-80) are represented

schematically, and the location of a deleted region is indicated The

dimerization domain is indicated by solid box, and the region involved

in DNA binding activity is by diagonally cross-hatched box (upper

panel) Changes of WD promoter activity were assayed in 48 h after

cotransfection of the pGL, Ku-70, DKu-80, and Ku-70/DKu-80

cDNA into HepG2 cells (lower panel) PC (positive control), promoter

activity in HepG2 cells transfected with pBluescript SK DNA

con-taining the same amount of Ku cDNA.

Fig 8 Difference of WD promoter activity between Xrs-5 and CHO cells (A) A decrease in WD promoter activity in Xrs5 cells CHO-K1 and Xrs5 cells were transiently transfected with a WD promoter-luciferase reporter gene (pPWD) [38] Twenty-four hours after transfection, the cells were assayed for luciferase activity (B) Restoration of promoter activity after transfection of Ku-80 gene in Xrs5 cells, but not in CHO cells Ku-80 cDNA and pPWD were transiently expressed in Xrs5 (upper panel) and CHO cells (lower panel) Data represent means ± SD of three inde-pendent transfection assays for each construct pGL indicates the luciferase activity in cells that had been transfected with the pGL as a negative control NC, Negative control transfected with the same amount of pBluescript SK instead of the Ku-80 cDNA.

most distal cis-element from transcription start, played the

most important role in WD gene promoter activity Further

investigation is needed to elucidate the exact reason why the

most distal MREa from transcription start site is important

for activity, especially in a TATA-less promoter such as the

WD gene

Several proteins that bind specifically to MREs, MTF-1,

ZAP, and ZRF (zinc regulatory factor), have been identified

in mouse and human cells [38,40,47] EMSAs performed

with unlabelled MREa or the Sp1 binding site as

compet-itors revealed that binding of the Ku protein to the MREa of

the WD gene is sequence specific Excess Sp1 oligonucleotide

reduced slightly the amount of the MREa–Ku protein

complex (Fig 2B, lane 7), possibly due to the ability of the

Ku protein to bind DNA in a sequence-independent manner

[48,49] In the MREa–protein complex detected by EMSA

(Fig 2A), both the 82- and 70-kDa proteins were visualized

by silver staining (unpublished data) In contrast, only the

70 kDa subunit was observed by South-Western blot

analysis (Fig 3A) This is consistent with previous studies

showing that the Ku-70 protein interacts with DNA without

requiring Ku-80 [50]

Using the avidin–biotin system, we purified two

MREa-binding proteins from HepG2 cell nuclear extracts It was

clearly shown that the two proteins bound to MREa in a

sequence-specific manner, by using competitors during the

purification process (Fig 3C) as well as EMSA (Fig 2B) to

exclude the possibility of sequence-independent binding of

Ku protein to double-stranded DNA ends The purified

proteins were of the same molecular sizes as those shown by

silver staining of an SDS/polyacrylamide gel after EMSAs

N-terminal sequencing of the 70-kDa protein was inhibited

by modification, consistent with previous reports showing

that the amino terminus of the human Ku-70 protein is

blocked and therefore inaccessible to Edman degradation

[51,52]

EMSA supershift assays (Fig 5B) and immunoblot

analyses (Fig 5A) using mAbs specific to human Ku-70

and Ku-80 revealed that the MREa-binding proteins

described herein and the Ku proteins are immunologically

related The Ku protein complex is known to contain

equimolar amounts of the 70- and 80-kDa polypeptides,

which form heterodimers [50,53,54] Various transcription

factors such as CHBF, CTCBF, TREF, and PSE1, each

of which recognize specific promoters elements, are known

to be identical or related to the Ku protein [22,24,25,55]

The Ku proteins have been shown to inhibit the

expression of stress-responsive proteins For example,

overexpression of Ku-70 and Ku-80 or Ku-70 alone

specifically inhibits HSP70 expression [56], whereas

over-expression of Ku-80 alone suppresses only MT-I

expres-sion [29]

We examined the effect of the Ku proteins on the

transcriptional regulation of the WD gene First, we

overexpressed Ku protein in HepG2 cells, and the results

indicated that overexpression of Ku protein did not alter

the concentration of WD protein in the cell (unpublished

data) We also inhibited expression of the Ku-80 protein

using the RNAi IR gene method In vivo RNAi with an IR

gene successfully inactivates a specific gene in established

cell lines as well as in nematodes [44,45] We found that

transfection of a Ku-80 IR gene into cells inhibited the

expression of the Ku-80 protein and the WD protein

(Fig 6B and C) In addition, the concurrent decrease in the level of Ku-70 from 0.5 to 12 h after transfection of Ku-80 IR gene is consistent with previous report that the stability of Ku-70 is compromised by the absence of Ku-80 [57]

This is the first report to show that Ku-80 expression is suppressed effectively through cell-line transfection In several studies, it has been reported that cells expressing truncated Ku-80 protein exhibit increased sensitivity to radiation and diminished DNA repair [58,59], although there are still some arguments in the exact locations of domains in Ku-80 [60–63] Based on the facts that amino acids 371–510 of Ku-80 mediate dimerization with Ku-70 protein, and that amino acids 179–510 of Ku-80 are involved in Ku-80-dependent DNA binding [64], we con-structed a Ku-80 deletion mutant (DKu-80) (Fig 7) Expression of the DKu-80 cDNA in HepG2 cells resulted

in decreased WD promoter activity, suggesting that DKu-80 protein inhibits the formation of endogenous Ku protein complex and its binding to the WD promoter by competing with native Ku-80 for Ku-70 protein We then verified that Ku-80 is required for transcriptional regulation of the WD gene in the Ku-80-deficient Xrs5 cell line (Fig 8) These findings suggest that Ku-80 binds to MREa and may be an essential component of the transcription machinery of the

WD gene However, it is not known if or how Ku-70 functions in the regulation of WD gene expression The reduction of luciferase activity in Xrs5 compared to CHO-K1 was smaller than the reduction of luciferase activity caused by the MREa mutation as shown in Fig 1B

It is possible that a low level of Ku-80 transcript present in the Xrs5 cell line [65] recovers the luciferase activity of the

WD promoter to some degree, while a mutation within MREa is able to confer a completely negative effect on WD gene promoter activity

It remains to be elucidated why WD promoter activity in Ku-80-expressing Xrs5 cells was not restored completely to the activity observed in CHO cells It could be that human Ku-80, whose amino acid sequence is 21% diverged from Chinese hamster Ku-80, rescues less efficiently in hamster species Also, it is consistent with previous report that CHO mutant cells transfected with Syrian hamster Ku-80 exhibit reduced X-ray resistance and V(D)J recombination com-pared with wild-type CHO cells [57] To rule out this possibility, we transfected the Chinese hamster Ku-80 clone into Xrs-5 cells However, there was no difference in the recovery of the promoter activity between human and Chinese hamster Ku-80 proteins The exact reason for this lack of difference is not clearly understood and further characterization is required

The Ku proteins have been shown to be components of the mammalian DNA-depedent protein kinase, which regulates other DNA-binding factors such as Sp1 and p53 through phosphorylation [66], and directly modulates RNA polymerase I-mediated transcription [25,54] It will be of interest to determine how this kinase influences WD gene expression and to decipher the mechanism by which it interacts functionally with constitutive transcriptional fac-tors like Sp1 and Ku It will also be of interest to examine whether additional proteins bind to the other three MREs in the WD promoter, and if so, to determine the precise mechanisms by which such proteins modulate WD gene expression

A C K N O W L E D G E M E N T S

We thank S.-H Yoo for providing Ku-70 and -80 cDNAs and K.-D.

Park for critically reading the manuscript We also thank our colleagues

in Dr Yoo’s laboratory for useful discussions This work was supported

by the Molecular Medicine Research Group Program grant

(98-MM-01-01-A-01) from the Ministry of Science and Technology through the

BioMedical Research Center at KAIST.

R E F E R E N C E S

1 Bull, P.C., Thomas, G.R., Rommens, J.M., Forbes, J.R & Cox,

D.W (1993) The Wilson disease gene is a putative copper

trans-porting P-type ATPase similar to the Menkes gene Nature Genet.

5, 327–337.

2 Kaler, S.G., Gallo, L.K., Proud, V.K., Percy, A.K., Mark, Y.,

Segal, N.A., Goldstein, D.S., Holmes, C.S & Gahl, W.A (1994)

Occipital horn syndrome and a mild Menkes phenotype associated

with splice site mutations at the MNK locus Nat Genet 8, 195–

202.

3 Petrukhin, K., Lutsenko, S., Chernov, I., Ross, B.M., Kaplan,

J.H & Gilliam, T.C (1994) Characterization of the Wilson disease

gene encoding a P-type copper transporting ATPase: genomic

organization, alternative splicing & structure/function predictions.

Hum Mol Genet 3, 1647–1656.

4 Vulpe, C., Levinson, B., Whitney, S., Packman, S & Gitschier, J.

(1993) Isolation of a candidate gene for Menkes disease and

evi-dence that it encodes a copper-transporting ATPase Nat Genet 3,

7–13.

5 Yamaguchi, Y., Heiny, M.E & Gitlin, J.D (1993) Isolation and

characterization of a human liver cDNA as a candidate gene for

Wilson disease Biochem Biophys Res Commun 197, 271–277.

6 Lutsenko, S & Kaplan, J.H (1995) Organization of P-type

ATPases: significance of structural diversity Biochemistry 34,

15607–15613.

7 Oh, W.J., Kim, E.K., Park, K.D., Hahn, S.H & Yoo, O.J (1999)

Cloning and characterization of the promoter region of the Wilson

disease gene Biochem Biophys Res Comm 259, 206–211.

8 Stuart, G.W., Searle, P.F & Palmiter, R.D (1985)

Identifica-tion of multiple metal regulatory elements in mouse

metallothio-nein-I promoter by assaying synthetic sequences Nature 317,

828–831.

9 Stuart, G.W., Searle, P.F., Chen, H.Y., Brinster, R.L &

Palmi-ter, R.D (1984) A 12-base-pair DNA motif that is repeated

several times in metallothionein gene promoters confers metal

regulation to a heterologous gene Proc Natl Acad Sci USA 81,

7318–7322.

10 Carter, A.D., Felber, B.K., Walling, M.J., Jubier, M.F., Schmidt,

C.J & Hamer, D.H (1984) Duplicated heavy metal control

sequences of the mouse metallothionein-I gene Proc Natl Acad.

Sci USA 81, 7392–7396.

11 Czupryn, M., Brown, W.E & Vallee, B.L (1992) Zinc rapidly

induces a metal response element-binding factor Proc Natl Acad.

Sci USA 89, 10395–10399.

12 Karin, M., Haslinger, A., Holtgreve, H., Richards, R.I.,

Krau-ter, P., Westphal, H.M & Beato, M (1984) Characterization of

DNA sequences through which cadmium and glucocorticoid

hormones induce human metallothionein-IIA gene Nature 308,

513–519.

13 Dalton, T.P., Li, Q., Bittel, D., Liang, L & Andrews, G.K (1996)

Oxidative stress activates metal-responsive transcription factor-1

binding activity Occupancy in vivo of metal response elements in

the metallothionein-I gene promoter J Biol Chem 271, 26233–

26241.

14 Scriver, C.R., Beaudet, A.L., Sly, W.S & Valle, D (1995) The

Metabolic and Molecular Bases of Inherited Disease, Vol II,

pp 2211–2236 McGraw-Hill Inc., New York.

15 Imbert, J., Culotta, V., Furst, P., Gedamu, L & Hamer, D (1990) Advances in Inorganic Biochemistry, Vol 8, pp 138–164 Elsevier Publishers, New York.

16 Kagi, J.H (1991) Overview of metallothionein Methods Enzymol.

205, 613–626.

17 Vallee, B.L (1979) Metallothionein: historical review and per-spectives Experientia 34(Suppl), 19–39.

18 Dv ir, A., Peterson, S.R., Knuth, M.W., Lu, H & Dynan, W.S (1992) Ku autoantigen is the regulatory component of a template-associated protein kinase that phosphorylates RNA polymerase II Proc Natl Acad Sci USA 89, 11920–11924.

19 Boulton, S.J & Jackson, S.P (1998) Components of the Ku-dependent non-homologous end-joining pathway are involved

in telomeric length maintenance and telomeric silencing EMBO J.

17, 1819–1828.

20 Chu, G (1997) Double strand break repair J Biol Chem 272, 24097–24100.

21 Weaver, D.T (1996) Regulation and repair of double-strand DNA breaks Crit Rev Eukaryot Gene Expr 6, 345–375.

22 Genersch, E., Eckerskorn, C., Lottspeich, F., Herzog, C., Kuhn,

K & Poschl, E (1995) Purification of the sequence-specific tran-scription factor CTCBF, involved in the control of human collagen IV genes: subunits with homology to Ku antigen EMBO

J 14, 791–800.

23 Giffin, W., Torrance, H., Rodda, D.J., Prefontaine, G.G., Pope, L & Hache, R.J.G (1996) Sequence-specific DNA binding by Ku autoantigen and its effects on transcription Nature

380, 265–268.

24 Kim, D., Ouyang, H., Yang, S.H., Nussenzweig, A., Burgman, P.

& Li, G.C (1995) A constitutive heat shock element-binding factor is immunologically identical to the Ku autoantigen J Biol Chem 270, 15277–15284.

25 Knuth, M.W., Gunderson, S.I., Thompson, N.E., Strasheim, L.A.

& Burgess, R.R (1990) Purification and characterization of proximal sequence element-binding protein 1, a transcription activating protein related to Ku and TREF that binds the prox-imal sequence element of the human U1 promoter J Biol Chem.

265, 17911–17920.

26 Li, G.C., Yang, S.-H., Kim, D., Nussenzweig, A., Ouyang, H., Wei, J., Burgman, P & Li, L (1995) Suppression of heat-induced hsp70 expression by the 70-kDa subunit of the human Ku auto-antigen Proc Natl Acad Sci USA 92, 4512–4516.

27 Messier, H., Fuller, T., Mangal, S., Brickner, H., Igarashi, S., Gaikwad, J., Fotedar, R & Fotedar, A (1993) p70 lupus auto-antigen binds the enhancer of the T-cell receptor beta-chain gene Proc Natl Acad Sci USA 90, 2685–2689.

28 Roberts, M.R., Han, Y., Fienberg, A., Humihan, L & Ruddle, F.H (1994) A DNA-binding activity, TRAC, specific for the TRA element of the transferrin receptor gene copurifies with the Ku autoantigen Proc Natl Acad Sci USA 91, 6354–6358.

29 Ghoshal, K., Li, Z & Jacob, S.T (1998) Overexpression of the large subunit of the protein Ku suppresses metallothionein-I induction by heavy metals P roc Natl Acad Sci USA 95, 10390– 10395.

30 Sambrook, J., Fritsch, E.F & Maniatis, T (1989) Molecular Cloning: a Laboratory Manual, 2nd edn Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.

31 Lin, S., Yang, S & Hopkins, N (1994) LacZ expression in germline transgenic zebrafish can be detected in living embryos Dev Biol 161, 77–83.

32 Dignam, J.D., Lebovitz, R.M & Roeder, R.G (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei Nucleic Acids Res 11, 1475– 1489.

33 Dikstein, R., Heffetz, D., Ben-Neriah, Y & Shaul, Y (1992) c-abl has a sequence-specific enhancer binding activity Cell 69, 751– 757.