Characterization of an upstream regulatory sequence and its binding protein in the mouse apolipoprotein E gene

BB.

ELSEVIER

Biochimica et Biophysica Acta 1262 (1995) 124-132

Biochi~ic~a et BiophysicaA~.ta

Characterization of an upstream regulatory sequence and its binding protein in the mouse apolipoprotein E gene Young-Ki Paik a,*, Catherine A. Reardon b,1, John M. Taylor b, Byung-Kwon Choi a a Department of Biochemistry and Bioproducts Research Center, Yonsei University, 134 Shinchon-dong Sudaemoon-ku, Seoul, 120-749, South Korea b Gladstone Institute of Cardiovascular Diseases, University of California, San Francisco, P.O. Box 419100, San Francisco, CA 94141-9100, USA Received 9 December 1994; revised 13 February 1995; accepted 13 February 1995

Abstract

The mouse apolipoprotein (apo) E gene from strain C57BL/6 was isolated from a genomic DNA library and its complete nucleotide sequence, together with 1.3 kilobase of 5' flanking DNA and 300 base pairs of the 3' flanking DNA, was determined. Regulatory sequences in the proximal 5' flanking region of the gene were identified. Using a chloramphenicol acetyltransferase transient assay system, positive and negative cis-acting sequences were mapped within 380 base pairs of the 5' flanking region of the mouse apoE gene. Two nuclear protein binding sites were identified within this region by DNase I footprinting. We have characterized one of these regions, termed mouse apoE regulatory sequence (MARS-2), which spans nucleotides -151 to -133. Gel mobility shift assays using oligonucleotides of the MARS-2 sequence having specific deletions or substitutions as probes or competitors showed that the essential sequence of MARS-2 required for nuclear protein binding consists of 16 nucleotides encompassing -151 to -136. When nuclear extracts from different cells were examined, L cells and mouse liver nuclear protein contained the highest levels of binding protein for the MARS-2 probe. This protein, termed MARS-2 binding protein, was purified from mouse liver nuclear extracts to homogeneity using gel filtration and MARS-2 oligonucleotide-specific column chromatographic procedures. The Mr = 66000 binding protein showed a gel mobility shift band that was identical to that of crude nuclear extracts. Keywords: Apolipoprotein E gene; Nuclear factor; Transcriptional regulatory element; Gel mobility shift assay; Protein binding domain; (Mouse)

1. Introduction

Apolipoprotein (apo) E is a component of various lipoprotein classes in mammals and it is a major determinant of plasma cholesterol levels [1]. It serves as a ligand for the low density lipoprotein (LDL) receptor, the LDL receptor related proteins (LRP) [2], and the recently described very low density lipoprotein (VLDL) receptor [3]. In addition to its role in plasma lipoprotein metabolism, apoE may also be involved in the pathogenesis of Alzheimer's diseases [4,5]. It has been identified in extra-

Abbreviations: Apo, apolipoprotein; MARS-2BP, mouse apolipoprotein E regulatory element 2 binding protein; CHO, Chinese hamster ovary; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; CAT, chloramphenicol acetyltransferase; RSV, Rous sarcoma virus; SDS-PAGE, sodium dodecylsulfate-polyacrylamide gel electrophoresis; DTT, dithiothreitol; PMSF, phenylmethanesulfonyl fluoride; VLDL, very low density lipoprotein. * Corresponding author. Fax: + 82 2 3629897. Present address: Department of Pathology, University of Chicago, IL 60637, USA. 0167-4781/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSD10167-4781(95)00048-8

cellular amyloid plaques, as well as intracellular neurofibrillary tangles in the brains of Alzheimer's patients [4,6]. The primary site of apoE synthesis is in the liver [7]. However, various extrahepatic tissues, such as the brain, adrenal, spleen, ovary, testis, and kidney, also contain significant amount of apoE m R N A [8]. In addition, mouse peritoneal macrophages [9] and human monocyte-derived macrophages [10] also synthesize apoE. ApoE secretion and m R N A levels in mouse peritoneal macrophages have been shown to be regulated by intracellular cholesterol [11-13], and the apoE gene may be controlled by cyclic AMP in the adrenal and ovary glands [14-16]. Early studies of the human apoE gene showed that its expression is modulated by multiple regulatory elements in the proximal promoter region [17-19]. Recent studies with transgenic mice showed that nearly all regulatory elements that direct the tissue specific expression of the human apoE gene are located downstream of the gene [20-22], and that they require an interaction with specific elements in the proximal promoter. Although the nucleotide sequences of the mouse apoE

Y.-K. Paiket al. / Biochimica et BiophysicaActa 1262 (1995) 124-132 gene has been reported previously for the Balb/c strain [23], studies on the regulatory sequences and their binding proteins have not been conducted. In this paper, we describe our studies on the isolation of the mouse apoE gene from the C57BL/6 strain, as well as the mapping and characterization of a cis-acting regulatory element and its associated binding protein.

2. Materials and methods

2.1. Isolation and sequencing of the mouse apoE gene A mouse (strain C57BL/6) genomic DNA library in lambda bacteriophage was a generous gift of Dr. Mark Davis at Stanford University. A 710 base pair (bp) fragment (XhoI-BgllI) of a mouse apoE cDNA clone [24] was labeled by random-priming and used to screen the library by plaque lift hybridization [25]. Three positive plaques were isolated and characterized. The sequence of 4.1 kilobase (kb) of genomic DNA was determined by the dideoxy chain termination method [26] following subcloning into M13 vectors. The M13 universal primer or mouse apoE gene specific primers (17 to 20 nucleotides) were used to prime DNA synthesis. The complete sequence of both strands covered by overlapping fragments was determined [26].

125

2.4. Synthesis of oligonucleotides The oligonucleotides were synthesized on an Applied Biosystems 380B synthesizer in the trityl-on mode, and they were purified by the Oligonucleotide Purification Cartridge system (Applied Biosystems, Foster City, CA). Sequences of synthetic oligonucleotides that were used for gel mobility shift assay are listed below. MARS-1

5'-gatcTGGC~A~A-3'

(-184 to -170)

MARS-2 (-151 t o - 1 3 3 )

3'-ACCGCCCCCTCCCCTctag-5'

5'-gaJcTCCAAACTCCA(L'IL-Tr~-3' 3

'

-

~

t

a

g

-

5

'

Other oligonucleotides including variant forms of MARS-2 are listed in Fig. 6A. 2.5. Isolation and preparation of nuclear extracts

For analysis of the promoter activity of the 5' flanking region, the vector pLS1 was used [17]. Fragments of the mouse apoE gene containing progressively smaller regions of the 5' flanking DNA, the first exon and 147 bp of the first intron were subcloned immediately upstream of the CAT gene. BgllI linkers were added to the 5' end of the DNA inserts and SacI linkers to the 3' end to facilitate subcloning into the pLS 1 vector. The precise end points of these deletion mutants were determined by DNA sequencing [26,27]. These constructs (15 /zg) were co-transfected with a vector containing the /3-galactosidase gene (10 /xg) under the control of the Rous sarcoma virus promoter [28] into cultured cells via the calcium phosphate DNA precipitation method [29]. Transfected cells were harvested 38 to 42 h after transfection and CAT activity and /3-galactosidase activity was measured as described previously [17].

Nuclear extracts were prepared from cultured HepG 2, CHO, L, P388D, and HeLa cells essentially as described [30]. For preparation of the mouse liver nuclear extract, male ICR mice (20-25 g body weight) were maintained on standard chow. Mice were killed by decapitation at the midpoint of the dark period of their light cycle, and the livers were excised and processed for nuclear protein preparation as described [31]. Nuclei were suspended at a concentration of ( 2 - 3 ) . 108/ml in a buffer A (20 mM Hepes, pH 7.9, 0.42 M NaC1, 1.5 mM MgC12, 0.2 mM EDTA, 0.5 mM dithiothreitol, 25% glycerol, and 0.5 mM phenylmethylsulfonyl fluoride). The mixture was stirred gently for 30 min at 4° C. The supernatant fluid was dialyzed against 250 volumes of buffer B (20 mM Hepes, pH 7.9, 0.1 M KC1, 0.2 mM EDTA, 0.5 mM dithiothreitol, 20% glycerol, 0.5 mM phenylmethylsulfonyl fluoride) for 2 h with one change of buffer. The dialysates then were centrifuged at 12000 rpm for 10 min at 4°C, and the supernatant fluid was stored in aliquots at - 8 0 ° C. The protein concentration of the extract was determined by the method of Bradford [32] using bovine serum albumin as a standard.

2.3. Tissue cultures

2.6. Gel mobility shift assay of nuclear extracts

Cells were grown in media supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 g / m l streptomycin. HepG 2 cells were maintained in modified Eagle's medium, CHO cells in F-12, and Dulbecco's modified Eagle's medium (50:50). HeLa, P388D, and L cells were grown in Dulbecco's modified Eagle's medium.

Gel mobility shift assays were performed essentially as described [33]. The standard reaction mixture contained 0.5 ng of 32p 5' end-labeled DNA, 10 /xg of nuclear extract, and 2 /zg of poly(dI-dC), poly(dI-dC) incubated for 30 min at 30 ° C in buffer C (10 mM Hepes at pH 7.9, 50 mM KCI, 0.1 mM EDTA, 0.25 mM dithiothreitol, 10%

2.2. Construction of CAT plasmids and analysis of CAT activity

126

Y.-K. Paik et al. / Biochimica et Biophysica Acta 1262 (1995) 124-132

glycerol, and 0.25 mM phenylmethylsulfonyl fluoride). In oligonucleotide competition experiments, a 120-fold molar excess of the unlabeled competitor oligonucleotides were preincubated with the reaction mixture for 10 min prior to the addition of 5' end-labeled D N A probe. The DNA-protein complex was examined by 4.5% polyacrylamide gel electrophoresis in 0.5 × TBE buffer (44.5 mM Tris, pH 8.3, 44.5 mM boric acid, and 0.5 mM EDTA).

volumes of buffer C containing 50 mM KC1 and loaded then to an affinity column prepared with the footprint sequence (MARS-2, nucleotides - 151 to - 133) that was prepared as described [34], except that CNBr-activated Sepharose 4B was used. Liver cell nuclear extract, or Sephacryl S-300 fractions ( ~ 10 mg of protein), was loaded onto the affinity column. To elute the bound proteins, nuclear extract buffer C with 0.8 M KC1 or with a step gradient of 0.1-0.8 M KCI, was used. Sodium dodecylsulfate-polyacrylamide gel electrophoresis was carried out according to the method of Laemmli [35].

2.7. Purification o f nuclear proteins

In a typical purification procedure, about 100 g of mouse liver tissues were used for preparation of the nuclear extracts (yield was usually 150 to 200 mg). In one experiment, 200 mg of the nuclear extracts were fractionated by Sephacryl S-300 superfine (Pharmacia LKB) gel filtration column chromatography essentially as described [19], except that the column was washed with nuclear extract buffer A. The activity of each fraction was examined by gel mobility shift assay. Active fractions were pooled and applied to a heparin-agarose column (total bed volume, 5 ml in Bio-Rad econo column) that had been equilibrated with buffer C containing 100 mM KC1 and eluted with buffer C containing 1 M KC1. The active fractions collected from this step was dialyzed against 200

2.8. Preparation o f R N A and p r i m e r extension analysis

Total cellular RNA was isolated from L cells transfected with pMAE3-SVCAT or p R S V C A T plasmid (Fig. 2) by the guanidine-HC1/CsC12 method [36] and primer extension analysis was performed essentially as described [37]. A 20-mer single-stranded oligonucleotide primer corresponding to nucleotides 15 to 34 of the coding sequence of the CAT gene was end-labeled using T4 polynucleotide kinase and [3/-32 P]ATP. 20 ng of the 32P-labeled primer was hybridized to 20 ~ g of total cellular RNA and extended as described [37]. The extension products were analyzed on 6% polyacrylamide, 8 M urea gels.

A

Cb-TBL6

Exon 1

Exon 2

42 rq

66 I---7 757

BALB/C

Exon 3

i

169

Exon 4

833

i

540

42

66

[--]

[--]

759

f 375

169 I

833

I

540

I

l

377

B -1004

-774

-542 -539

¢:~-7BL6

5"-GATCT

...... A C C C T

......T C T C ,

BALB/c

5'-GATCT

...... A C - C T

......T - T G - C

-226

-176

-163

GC

-263 ......G A C C C

......G C ~

....

......G A - C C

......G C - - G

....

-47

+i ,

C5-7BL6 BALB/c

-257 -256

CC-AG

......G C - G G

......G C - G G

......C G - G C

,

......G C T C A - 3 "

Fig. 1. Comparison of the mouse apoE gene sequence obtained from C57BL/6 and BALB/c strains. (A) Schematic outline of the apoE gene. Relative positions of the exons are shown by open boxes above the line. Lengths of sequence are shown for exons (above the boxes) and for introns (below the lines). (B) Sequence differences in the 5' flanking region between the C57BL/6 and BALB/c genes. Arrows indicate the positions in which different nucleotide sequences were found between the two sequences. The transcription initiation site is marked ( • ) with + 1.

127

Y.-K. Paik et al. / Biochimica et BiophysicaActa 1262 (1995) 124-132 2.9. DNase I footprinting assays

The footprinting reactions were performed as previously described [17]. Various ~Lmounts of the nuclear extracts were preincubated with 1 /zg of poly(dI-dC) in the reaction mixture for 20 min at room temperature, prior to the addition of 5 ng of 5' end-labeled (by T4 polynucleotide kinase) DNA fragment. The mixture was incubated for 40 min on ice, digested with DNAse I, and analyzed on 8% polyacrylamide, 8 M urea sequencing gels as described [191. 2.10. Materials

[ a - 32P]dCTP and [y-32 P]ATP were obtained from Amersham (Arlington Heights, IL). Nucleotide triphosphates, poly(dI-dC), poly(dI-dC), Sephacryl S-300, and CNBr-activated Sepharose 4B were purchased from Pharmacia (Piscataway, N J). RNAsin, reverse transcriptase of avian myeloblastosis virus (AMV), and c~-amanitin were from Promega (Madison, WI), Life Sciences (St. Petersburg, FL), and Boehringer-Mannheim (Indianapolis, IN) respectively. Restriction endonucleases, T4 DNA ligase, T4 polynucleotide kinase, and the Klenow fragment of DNA polymerase I were purchased from New England Biolabs (Boston, MA). Protein assay dye was obtained from Bio-Rad (Richmond, CA). Various chemicals and solvents were purchased from Sigma or Merck. []4C]Chloramphenicol (50-57 m C i / m m o l ) was purchased from Du Pont-New England Nuclear. Acetyl-coenzyme A (lithium salt) and O-nitrophenyl-/3-o-galactopyranoside were purchased from Pharmacia and Sigma, respectively. Cell culture media and nutrients were obtained from Gibco. Unless stated otherwise, procedures involving recombinant DNA, enzymes and reagents were used under the conditions recommended by the suppliers and in accordance with the guidelines of the National Institutes of Health. All other reagents were of the: best grade available.

3. Results and discussion 3.1. Isolation o f the mouse apoE gene and its nucleotide sequence analysis

The nucleotide sequence of the entire C 5 7 B L / 6 mouse apoE gene coding sequence, 1.3 kb of 5' flanking DNA and 300 bp of 3' flanking DNA, were sequenced (data not shown, submitted to EMBL/GenBank). The gene contains four exons and three introns, with the lengths shown in Fig. 1A. The nucleotide sequence is 82% identical to the previously reported nucleotide sequence of the mouse apoE gene obtained from B A L B / c strain [23]. However, in this latter sequence, the lengths of the first and third introns of the apoE gene were determined to be 759 bp and 377 bp, respectively, while we de~Lerrnined that they were 757 and

pMAEI

5'.Flanking Region

Relative Activity

Fine FIrll

e~o,~,,~

-589 se~.

(%) CHO

L

(n =3)

-589 pMAE2 pMAE3

pMAE4 pMAE5 pMAE6

~

100

-380

100

105, 9.5 94,8.2 -221

145*12.3 119, 7.6

~

"sl~/ll~lm~l~r~ "5~lm~mmmrc~ .

3

9

~

39*

2.7

4 7 t 3.9

44,

3.5

35 • 3.6

29* 2.1 ~,*

2.5

Fig. 2. Promoteractivityof the mouse apoE-CATgene recombinants.The numbers to each side of the solid bars indicate the nucleotide positions in base pairs relative to the transcription start site. The hatched box represents the first exon. The relative CAT activities of the different deletion constructs transfected into CHO and L cells are expressedas a percentage of that of pMAEI (-589 to + 189). The CAT specific activity of each construct was normalized to the cotransfected 13-galactosidaseactivity [14]. The values in the columns on the right represent the means+ S.D. (n= 3). 375 bp in length in the corresponding C 5 7 B L / 6 gene. There are also several differences in the sequence of the 5' flanking DNA (Fig. 1B). The coding parts of the C 5 7 B L / 6 gene (in exons two, three, and four) predict an amino acid sequence that is identical to the previously published sequence derived from the B A L B / c apoE gene (data not shown). 3.2. Promoter activity o f the mouse apoE gene

To identify regulatory regions in the mouse apoE gene promoter, a nested series of fragments containing various amounts of 5' flanking region, together with the complex first exon and 147 bp of the first intron (nucleotide + 189), were generated by restriction endonuclease digestion. They were ligated to the CAT gene and analyzed for promoter activity using a transient assay system as described previously [17]. These deletion constructs were transfected into CHO and L cells and their CAT activity measured (Fig. 2). The promoter activity of the longest fragment, beginning at - 5 8 9 , was set at 100% for results from both CHO and L cells. Deletion of the 209 nucleotides between - 5 8 9 and - 3 8 0 caused almost no change in the apoE promoter-directed CAT activity. Deletion of the region between nucleotides - 3 8 0 and - 2 2 1 resulted in stimulation of the CAT activity, suggesting the presence of a negative element(s) in this region. However, deletion of the nucleotides between - 2 2 1 and - 8 6 caused 2 to 3-fold reduction in the promoter activity relative to the - 2 2 1 region. This result suggests the presence of positive cisacting sequences in this region. Little change in CAT activity was observed when the nucleotides between - 8 6 and - 39 were deleted. Therefore, we have focused on the region spanning nucleotides - 2 2 1 to - 8 6 in the 5'-flanking region of the mouse apoE gene.

128

Y.-K. Paik et al. / Biochimica et Biophysica Acta 1262 (1995) 124-132

09

the C A T gene constructs were e x a m i n e d by p r i m e r extension analysis. To facilitate detection of the primary transcripts p r o d u c e d by the m o u s e a p o E / C A T recombinant, the S V 4 0 - e n h a n c e r sequence was ligated into the p M A E 3 C A T vector 600 bp upstream f r o m the a p o E gene p r o m o t e r [38,17]. This S V 4 0 enhancer-containing plasmid, designated as p M A E 3 - S V C A T , was transfected to L cells and total R N A prepared. To control for the transfection efficiency, L cells were also transfected with p R S V - C A T [28].

m

242-45 -59

GC-box

201-

-133

160-

MARS-2

-151 110 -170 gO-

~ii~!iiiiiiiiiii ¸'~'¸ iii!i!ii~!!il ¸~¸'~¸~'

1

MARS-1

-184

2

Fig. 3. Primer extension analysis of the CAT gene transcripts. Total cellular RNA from L ceils transfected with pMAE3-SVCAT and pRSVCAT was primed with an oligomer corresponding to nucleotides + 15- + 34 of the protein coding sequence of the CAT gene. The positions of the primer extension products of RNA initiating at the transcription start sites from the apoE promoter (pMAE3-SVCAT, 260 bp, lane 1) and the rous sarcoma virus promoter (pRSV-CAT, 110 bp, lane 2) are shown by the arrows,

3.3. Primer extension analysis o f the recombinant apoEC A T constructs T o verify that the products o f the m o u s e a p o E / C A T r e c o m b i n a n t g e n e reflected correctly initiated m R N A transcripts, and that C A T activity m e a s u r e d in this system did indeed reflect transcriptional regulation o f the marker gene, total cellular R N A from L cells that were transfected with

1

2345

Fig. 4. DNase I footprint analysis of the proximal 5' flanking sequence of the mouse apoE gene fragments by L cell nuclear proteins. A mouse apoE DNA probe (356 bp: -369 to - 1 4 ) was 5' end-labeled at nucleotide -369. Lanes 1 and 2, G and G + A reactions of Maxam and Gilbert sequencing reactions; lane 3, no nuclear extract; lane 4, 72 /zg of nuclear extract; lane 5, 36 /zg of nuclear extract. Sequences protected from DNase I digestion are indicated by the brackets, with the nucleotide positions at the boundaries indicated on the right.

Y.-K. Paik et al. / Biochimica et Biophysica Acta 1262 (1995) 124-132 As shown in Fig. 3, primer extension of the transfected cell R N A yielded bands of the expected size; 260 nucleotides from p M A E 3 - S V C A T (189 and 71 nucleotides from the mouse apoE and C A T gene, respectively) and 110 nucleotides from pRSV-CAT.

3.4. DNase 1 footprinting analysis of the apoE promoter region To examine whether nuclear proteins interact with this region, D N A s e I footprinting assays were carried out using a D N A probe corresponding to nucleotides - 3 6 9 to - 14 and L cell nuclear extracts. Fig. 4 shows that there is a protected region located at nucleotide - 1 8 4 to - 1 7 0 (designated as mouse apoE regulatory sequence, MARS-1), and another protected region of nucleotides - 151 to - 133 (MARS-2). Each protected region is within the region containing transcriptionally active sequences. In MARS-2, we observed that 11 nucleotides were identical in sequence

129

to the 3' region of the PET (positive element for transcription, see Ref. [19]) found at position - 161 to - 141 of the human apoE gene promoter [19]. The PET enhancer was found to be the most important regulatory sequence in this promoter, required for mediating the expression of several tissue-specific control elements [21]. Because of its potential importance for mouse apoE gene expression, we directed our efforts to characterizing the M A R S - 2 sequence in our subsequent studies.

3.5. Binding pattern of MARS-2 binding proteins To determine if nuclear protein binding to M A R S - 2 is present in different cultured cells, gel mobility shift assays were carried out using the nuclear extracts prepared from HepG2, HeLa, CHO, L, and P388D cells. Fig. 5A shows nuclear protein binding activity in each extract, and that L cell nuclear extracts have the highest binding activity among the cultured cells examined (lane 4). Two gel

B

C

@

*~'~*~ G*~ x"

Amount of Poly(dl-dC). Poly(dl-dC)

1

2

3

4

5

6

7

8

9

10

-b a--

b--

o

1234:5 Fig. 5. Gel mobility shift assay of MARS-2 by nuclear extracts from various cell sources. (A) Cultured cell nuclear extracts. The 5' end-lal~led MARS-2 oligonucleotide was incubated with t0 /~g of nuclear extracts of cultured HepG 2 (lane t), HeLa (lane 2), GHO (lane 3), L (lane 4), and p388D (lane 5)

cells. Bands a and b are the retained oligonucleotide due to nuclear protein binding. (B) Intact mammalian tissues. Lanes 1-2 are gel mobility shift bands from incubation of 10 /xg of nuclear extracts from human placental tissues and mouse liver respectively. (C) Effects of poly(dI-dC) • poly(dI-dC) on binding of mouse liver nuclear exl:ractsto the MARS-2 probe. Lanes 1- 10 are 0, 0.1,0.2, 0.3, 0.4 0.5, 1.0, 1.5, 2.0, and 2.5 ~g of poly(dl-dC) • poly(dl-dC), respectively, added to the reaction mixture.

130

Y.-K. Paik et al./Biochimica et Biophysica Acta 1262 (1995) 124-132

retention bands were observed in each case, with variable amounts of the lower band detected in each extract. Mouse liver had abundant amounts of binding protein, whereas it was not detected in human placenta extracts (Fig. 5B). The

upper band in the mouse liver nuclear extracts appeared to represent higher protein binding affinity for the sequence as indicated by gel mobility shift using mouse liver nuclear extracts in the presence of poly (dI-dC) • poly (dI-dC) (Fig.

A -151

-130

Wt

5'-

TCCAAACTCCACCTC-I-I-I'CCTC -3'

M1

5' - TC . . . . CTCCACCTC-I-FICCTC -3'

M2

5'-

TCCAAA .--CACCTCTTTCCTC -3'

M3

5'-

TCCAAACTC...CTCFFICCTC -3'

M4

5 ' - TCCAAACTCCAC..- TTICCTC -3'

M5

5' - TCCAAACTCCACCTC . . . . CTC -3'

M6

5' - TCCAAACTCCACCTC-Iqq-C • • • -3'

M7

5' - T C C A A A . . . C A C " " TFI-CCTC -3'

M8

5' - TCCAAACTC,~CCTCIqq'CCTC -3'

M9

5' - TCCAAACTCCA~CTCTVrCCTC-3'

B

C Cornl~titors VV( M 2 M 4 M 6 M1 M 3 M 5 M 7 M 8 M 9

M2 M4 M6 M1 M3 M5 M7 M8 M9 Wt

--

!i

a

u

bC-

1 2

3 4

5

6

7

8

9

D

Fig. 6. Gel mobility shift analysis of the MARS-2 sequence. (A) Oligonucleotides used in both competition and protein binding analysis. The nucleotides between positions - 151 and - 130 on the 5' strand are shown. Letter 'Wt' represents the wild type MARS-2 sequence, M1-7 are MARS-2 deletion mutants with the dot (.) indicating the nucleotides deleted, and M8-9 are MARS-2 substitution mutants with ( • ) indicating the C to A substituted nucleotides. (B) Gel mobility analysis of MARS-2 and the variant MARS-2 oligonucleotides. 5' end-labeled Wt oligonucleotide (lane 1) or variant oligonucleotides (lanes 2-10) were incubated with 10 /zg of mouse liver nuclear extract. Letters (a, b, and c) at left side indicate the oligonucleotide retained at different positions. Different extract preparations yielded variable amounts of band c, which appeared to represent non-specific protein binding. (C) Gel mobility shift of Wt MARS-2 with oligonucleotide competitors. 5' end-labeled MARS-2 oligonucleotide was incubated with 10/zg of liver nuclear extract and oligonucleotide competitors in 120-fold molar excess (lanes 1-13; lane 14, no competitor). The identical results were obtained with L cell nuclear extracts (data not shown).

Y.-K. Paik et al. /Biochimica et Biophysica Acta 1262 (1995) 124-132 5C). It was noticed that there appears to be essentially no difference between the L cell nuclear extracts and mouse liver nuclear extracts for binding to the M A R S - 2 probe (Fig. 5 A - C ) . W e also fou:ad subsequently that the M A R S - 2

131

binding protein (MARS-2BP) produced at least two bands in all the nuclear extracts examined, including the mouse liver.

3.6. Determination o f the protein binding domain o f the MARS-2

A 94 67

D

43

30

-

20.5-

1

B

1

am

b-

2

2

3

3

4

4

5

To characterize the importance of individual nucleotides and the boundary of the M A R S - 2 sequence that is essential for nuclear protein binding, a series of double-stranded oligonucleotides that contained various portions of M A R S 2 were synthesized (Fig. 6A). As shown, M A R S - 2 has three 5'-CTC-3' motifs, and it has a dyad symmetric sequence. To test their ability to bind specific nuclear protein(s), these oligonucleotides were used as probes or competitors in gel mobility shift assays using mouse liver nuclear extracts. As seen in the Fig. 6B, wild type (wt) probe and the M6 probe, which deleted the 3'-termini CTC sequence, were equally effective in the binding of the nuclear protein. The M5 probe, in which nucleotides - 133 to - 1 3 6 were deleted, yielded band a but not band b. None of the other deletion mutant oligonucleotides (M 1-4 and M7) produced either band a or band b. Furthermore, C ~ A substitutions at nucleotides - 143 or - 141 (M8 or M9, respectively) resulted in a lack of band a and band b binding activity. When the gel mobility pattern was examined using the wild type oligonucleotide probe in the presence of a 120-fold molar excess of the mutant oligonucleotides, essentially the same results were obtained (Fig. 6C). Only oligonucleotides M5, M6, and wild type probe were effective competitors for band a and band b protein binding. These results suggest that the required sequences for M A R S - 2 binding protein resided within residues - 151 to - 136, and that specific nuclear proteins bind to this sequence. The identical results were obtained with the L cell nuclear extracts (data not shown).

3. 7. Purification and characterization o f MARS-2BP The M A R S - 2 B P was purified from mouse liver nuclear extracts using chromatographic procedures (Fig. 7A). Crude

Fig. 7. Silver staining and gel mobility shift assay of the purified MARS-2BP. (A) Sodium dodecylsulfate-polyacrylamide(8%) gel electrophoresis was performed on mouse liver nuclear protein fractions at different stages of purification, followed by silver staining. Lane 1, crude nuclear extracts treated with 0.1% SDS; lane 2, 3 /zg of protein from pooled Sephacryl S-300 fractions with MARS-2-binding activity; lane 3, 300 ng of protein eluted from heparin-agarose column; lane 4, 500 ng of protein from the first affinity column passage; lane 5, 30 ng of protein eluted from the second MARS-2-oligonucleotide affinity column. The numbers on the left represent the molecular sizes (X 10-3) of protein standards. (B) Gel mobility shift analysis of MARS-2-binding protein. 5' end-labeled MARS-2 oligonucleotide was analyzed by the gel mobility shift assay under standard conditions. Lane 1, no nuclear extract; lane 2, 10 /.~g of crude nuclear extract; lane 3, 100 ng of binding protein-enriched fraction from Sephacryl S-300 fractions; lane 4, ~ 300 ng of the purified protein from a MARS-2 oligonucleotide-specificaffinity column.

132

Y. -K. P aik et al. /Biochimica et Biophysica Acta 1262 (1995) 124-132

nuclear extracts (about 200 mg) were applied to a Sephacryl S-300 gel filtration c o l u m n and the eluted fractions were e x a m i n e d for M A R S - 2 b i n d i n g activity by gel mobility shift assay. The active fractions (about 10 rag) were pooled and applied to a M A R S - 2 specific oligonucleotide affinity column. Fractions were collected by eluting with stepwise increments of KCI. After two consecutive passages over the affinity column, the final protein fractions were collected and analyzed for M A R S - 2 b i n d i n g activity and molecular size. The analysis of the purified M A R S - 2 B P by gel electrophoresis under denatured conditions showed a M r = 66 000 (Fig. 7A). The purified protein (about 5 0 / x g ) yielded a gel mobility shift b a n d that was identical to the band produced by crude nuclear extracts (Fig. 7B). In conclusion, we have isolated the mouse apoE gene and characterized one of its promoter sequences, M A R S - 2 . The nuclear protein that binds to M A R S - 2 was purified to homogeneity. The purified protein should be useful for in vitro r e c o n s t i t u t i o n to e l u c i d a t e the r e g u l a t o r y mechanism(s) controlling the expression of the mouse apoE gene. Further studies also will be aimed at characterizing the M A R S - 3 footprint sequence and determining the roles of both elements in apoE promoter activity.

Acknowledgements This work was supported in part by Korea Research Foundation, Genetic E n g i n e e r i n g Research F u n d ( 1 9 9 2 1994) (to Y.K.P.), and by grant HL37063 from the American National Heart, Lung, and Blood Institute (to J.M.T).

References [1] Mahley, R.W. (1988) Science 240, 622-630. [2] Kowal, R.C., Herz, J., Weisgraber, K.H., Mahley, R.W., Brown, M.S. and Goldstein, J.L. (1990) J. Biol. Chem. 265, 10771-10779. [3] Takahaski, S., Kawarabayasi, Y., Nakai, T., Sakai, J. and Yamamoto, T. (1992) Proc. Natl. Acad. Sci. USA 89, 9252-9256. [4] Corder, E.H., Saunders, A.M., Strittmatter, W.J., Schmechel, D., Gaskell, P.C., Small, G.W., Roses, A.D., Haines, J.L. and PericakVance, M. (1993) Science 261,921-923. [5] Strittmatter, W.J., Saunders, A.M., Schmechel, D., Pericak-Vance, M., Enghild, J. (1993) Proc. Natl. Acad. Sci. USA 90, 1977-1981. [6] Weisgraber, K.H., Roses, A.D. and Strittmatter, W.J. (1994) Curr. Opin. Lipidol. 5, 110-116. [7] Kraft, H.G., Menzel, H.J., Hoppichler, F., Vogel, W., Uterman, G. (1989) J. Clin. Invest. 83, 137-142.

[8] Elshourbagy, N.A., Liao, W.S., Mahley, R.W. and Taylor, J.M. (1985) Proc. Natl. Acad. Sci. USA 82, 203-207. [9] Basu, S.K., Ho, Y.K., Brown, M.S., Bilheimer, D.W., Anderson, R.G.W. and Goldstein, J.L. (1982) J. Biol. Chem. 257, 9788-9795. [10] Werb, Z. and Chin, J.R. (1983) J. Exp. Med. 158, 1272-1293. [11] Mazzone, T., Gump, H., Diller, P. and Getz, G.S. (1987) J. Biol. Chem. 262, 11657-11662. [12] Mazzone, T., Basheeruddin, K. and Poulos, C. (1989) J. Lipid. Res. 30, 1055-1064. [13] Mazzone, T. and Reardon, C. (1994), J. Lipid. Res., 35, 1345-1353. [14] Wyne, K.L., Schreiber, J.R., Larsen, A.L. and Getz, G.S. (1989) J. Biol. Chem. 264, 981-989. [15] Reyland, M.E., Gwyne, J.T., Forgez, P., Prack, M., Williams, D.L. (1991) Proc. Natl. Acad. Sci. USA 88, 2375-2379. [16] Nicosia, M., Moger, W.H., Dyer, C.A., Prack, M.M., Williams, D.L. (1992) Mol. Endocrinol. 6, 978-988. [17] Paik, Y.-K., Chang, D.J., Reardon, C.A., Walker, D.W., Taxman, E. and Taylor, J.M. (1988) J. Biol. Chem. 263, 13340-13349. [18] Smith, J.D., Melian, A., Left, T. and Breslow, J.L. (1988) J. Biol. Chem.. 263, 8300-8308. [19] Chang, D.J., Paik, Y.-K., Leren, T.P., Walker, D.W., Howett, G.J. and Taylor, J.M. (1990) J. Biol. Chem. 265, 9496-9504. [20] Simonet, W.S., Bucay, N., Pitas, R.E., Lauer, S.J. and Taylor, J.M. (1991) J. Biol. Chem. 266, 8651-8654. [21] Simonet, W.S., Bucay, N., Lauer, S.J. and Taylor, J.M. (1993) J. Biol. Chem. 268, 8221-8229. [22] Shachter, N.S., Zhu, Y., Walsh, A., Breslow, J.L. and Smith, J.D. (1993) J. Lipid Res. 34, 1699-1707. [23] Horiuchi, K., Tajima, S., Menju, M. and Yamamoto, A.J. (1989) J. Biochem. 106, 98-103. [24] Reue, K.L., Quon, D.H., O'Donnell, K.A., Dizikes, G.J., Fared, G.C. and Lusis, A.J. (1984) J. Biol. Chem. 259, 2100-2107. [25] Shambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Edn., Cold Spring Harbor Laboratory Press, Cold Spring Harbor. [26] Sanger, F., Nicklen, S. and Coulson, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467. [27] Maxam, A.M. and Gilbert, W. (1980) Methods Enzymol. 65, 499560. [28] Gorman, C.M., Merlino, G.T., Willingham, M.C., Pastan, I. and Howard, B.H. (1982) Proc. Natl. Acad. Sci. USA 79, 6777-6781. [29] Graham, F.L. and Van der Eb, A. (1973) J. Virol. 52, 456-467. [30] Dignam, J.D., Lebovitz, R.M. and Roeder, R.G. (1983) Nucleic Acids Res. 11, 1475-1489. [31] Shapiro, D.J., Sharp, P.A., Wahli, W.W. and Keller, M.J. (1988) DNA 7, 47-55. [32] Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. [33] Carthew, R.W., Chodosh, L.A. and Sharp, P.A. (1985) Cell 43, 439-448. [34] Kadonaga, J.T. and Tjian, R. (1986) Proc. Natl. Acad. Sci. USA 83, 5889-5893. [35] Laemmli, U.K. (1970) Nature 227, 680-685. [36] Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Biochemistry 18, 5294-5299. [37] Edlund, T., Walker, M.D., Barr, P.J. and Rutter, W.J. (1985) Science 230, 912-916. [38] Gorman, C.M., Moffat, L.F. and Howard, B.H. (1982b) Mol. Cell. Biol. 2, 1044-1051.