Purification and characterization of nuclear scaffold proteins which bind to a highly repetitive bent DNA from rat liver

162

Biochimica et Biophysica Acta, 1174(1993) 162-170 © 1993 Elsevier Science Publishers B.V. All rights reserved 0167-4781/93/$06.00

BBAEXP 92516

Purification and characterization of nuclear scaffold proteins which bind to a highly repetitive bent D N A from rat liver Yasuhide Hibino, Kouichi Nakamura, Shuichi Tsukada and Nobuhiko Sugano Cell Biology Division, Faculty of Pharmaceutical Sciences, Toyama Medical & Pharmaceutical University, Toyama (Japan) (Received 3 December 1992) (Revised manuscript received 8 February 1993)

Key words: DNA; Highly repetitive DNA; Bent DNA; DNA-binding protein; Nuclear scaffold; Protein characterization; (Rat liver)

Our previous work (Hibino et al. (1992) Biochem. Biophys. Res. Commun. 184, 853-858) has shown that the binding affinities of a highly repetitive DNA component for rat nuclear scaffold proteins, P123 and P130, depend on the degree of sequence-directed bending of the helix axis. In the present experiment, these proteins have been purified and finally isolated by DNA-Sepharose column chromatography. The p I values of P123 and P130 were 7.2 and 8.1, respectively. The southwestern blotting revealed that a highly repetitive bent DNA (370-bp XmnI fragment) from rat liver binds readily to the isolated proteins under a hypotonic condition (50 mM NaC1) and that the level of the binding affinity for each protein was lowered with increasing NaCI concentration. The sedimentation analysis predicted that direct interaction between the XmnI fragment and P123 or PI30 results in the formation of a complex which consists of two of the fragments and one molecule of the protein, alternatively, one of the fragment and three molecules of the proteins. Distamycin A, an antibiotic which binds specifically to AT-rich DNA, removed the bend in the XmnI fragment and inhibited binding of the fragment to P123 or P130, whereas neither removal of the bend nor binding inhibition was observed with chromomycin A3, an antibiotic specific for GC-rich sites in DNA. These results imply that AT-rich regions in a highly repetitive DNA component cause bending of the helix axis to be recognized by some of nuclear scaffold proteins.

Introduction Highly repetitive D N A sequences were found in various eukaryotic genomes by restriction endonuclease digestion studies [1-2]. These repetitive D N A components have been evidenced to be recognized by a nonhistone nuclear protein [3-8] and to be enriched in nuclear matrix and nuclear scaffold which may be competent for frameworks of metaphase chromosome a n d / o r interphase nucleus [9-11]. Moreover, highly repetitive components from mouse, rat and monkey have been demonstrated to exhibit some properties characteristic of sequence-directed bent D N A [12-14]. Thus, as a preliminary step to investigate the interac-

Correspondence to: N. Sugano, Cell Biology Division, Faculty of Pharmaceutical Sciences, Toyama Medical & Pharmaceutical University, 2630 Sugitani, Toyama City, Toyama 930-01, Japan. Abbreviations: bp, base pairs; PMSF, phenylmethylsulfonylfluoride; POPOP, 2,2'-p-phenylene-bis(5-phenyloxazole); PPO, 2,5-diphenyloxazole; TP, 25 mM Tris-HCl (pH 6.5)/0.1 mM PMSF.

tion between such a repetitive component and a nonhistone nuclear protein, the present studies are concerned with purification and characterization of nuclear scaffold proteins which bind to a highly repetitive bent D N A from rat liver. Materials and Methods

Preparation of nuclei Rat livers were homogenized in 0.34 M s u c r o s e / 6 0 mM KC1/15 mM NaC1/15 m M Tris-HC1 (pH 7.4)/15 mM 2 - m e r c a p t o e t h a n o l / 2 m M E D T A / 1 mM phenylmethylsulfonyl fluoride (PMSF)/0.5 m M E G T A / 0 . 5 mM spermidine/0.15 m M spermine [15] and centrifuged at 4000 x g for 5 min. The pellet was washed twice with the same sucrose buffer and dispersed in 2.2 M sucrose/60 m M KC1/15 m M N a C I / 1 5 mM Tris-HCl (pH 7.4)/15 mM 2-mercaptoethanol/0.5 m M spermidine/0.15 m M spermine/0.1 mM E D T A / 0 . 1 mM E G T A / 0 . 1 m M PMSF. The dispersed material was centrifuged at 105000 x g for 1 h. The pellet was dispersed again in the 2.2 M sucrose buffer and recen-

163 trifuged in the same way. The final pellet was taken as the nuclei.

Preparation of nuclear scaffold fraction The fraction was prepared according to the method of Laemmli and co-workers [16,17]. The nuclei (10 A260) were incubated in 1 ml of 20 mM KC1/3.75 mM Tris-HC1 (pH 7.4)/0.5 mM CUSO4/0.125 mM spermidine/0.05 mM spermine/l% thiodiglycol/Trasylol (100 units/ml, Bayer) at 4°C for 10 min and centrifuged at 4000 x g for 5 min. The pellet was incubated in 100/zl of 20 mM KC1/3.75 mM Tris-HC1 (pH 7.4)/0.125 mM spermidine/0.05 mM spermine/l% thiodiglycol/0.1% digitonin/Trasylol (100 units/ml) at 37°C for 20 min. The incubated material was mixed with 7 ml of 25 mM 3,5-diiodosalicylic acid lithium salt/5 mM Hepes-NaOH (pH 7.4)/2 mM KC1/2 mM EDTA/0.25 mM spermidine/0.1% digitonin/Trasylol (100 units/ml). The mixture was shaken slowly at room temperature for 30 min and centrifuged at 2400 X g for 20 min. The pellet was washed four times with 8 ml of 70 mM NaCI/20 mM KC1/20 mM Tris-HCl (pH 7.4)/10 mM MGC12/0.125 mM spermidine/0.1 mM PMSF/0.05 mM spermine/0.1% digitonin/Trasylol (100 units/ml) and digested with EcoRI (500 units/ml, Takara) at 37°C for 1 h. The digest was further digested with HaelII and HindlII (50 units/ml each, BRL) for 2 h and centrifuged at 2400 x g for 10 min. The pellet was taken as the nuclear scaffold fraction.

Purification and isolation of DNA-binding proteins The scaffold fraction from the nuclei (400 A260) was dissolved in 4 ml of 25 mM Tris-HCl (pH 6.5)/5% glycerol/l% 2-mercaptoethanol/l% SDS and applied onto a column (1.6 x 36 cm) of Sephadex G-200, equilibrated, and eluted with the same buffer. The void fraction was dialyzed against 25 mM Tris-HC1 (pH 6.5)/0.1 mM PMSF (TP) and further fractionated by batchwise operations as follows. The dialyzate was mixed with a gel slurry of CM-Sephadex C-50 (appropriate volume) in TP. The mixture was gently stirred for 30 min and centrifuged. The precipitate was slurried again in TP containing 1.0 M NaC1 and gently stirred for 1 h. The stirred material was centrifuged and the supernatant (1.0 M NaCl-soluble fraction) was dialyzed against TP. The dialyzate was mixed with a gel slurry of DNA-Sepharose 4B (appropriate volume) in TP and treated in the same way. The resulting 1.0 M NaCl-soluble fraction from the gel was dialyzed against 1 mM Tris-HC1 (pH 6.5)/1 mM PMSF. In this connection, DNA-Sepharose 4B was prepared in a conventional way with calf thymus DNA (Worthington) which has an AT content of 57%. To monitor the purification grade, all the resulting fractions were electrophoresed on a slab gel of 0.1% SDS-15% polyacrylamide and stained with 0.1% Coomassie brilliant blue in 50%

methanol/7% acetic acid [18]. Then, the fraction from DNA-Sepharose 4B gel was applied onto a column (0.6 x 10 cm) of the same gel equilibrated with 1 mM Tris-HC1 (pH 6.5)/1 mM PMSF. The elution was performed with the same buffer in a 0-1.0 M NaC1 gradient. To detect protein elution, aliquots of every second column fractions were electrophoresed on a slab gel of 0.1% SDS-7.5% polyacrylamide and stained with Coomassie brilliant blue, as described above. The resulting two protein fractions (column fractions, 4-8 and 12-18) were dialyzed against 1 mM Tris-HC1 (pH 7.0) and lyophilized. The lyophilized materials were taken as DNA-binding proteins and designated P123 and P130, in order of elution.

Determination of pI ualues P123 or P130 was dialyzed against distilled water and lyophilized. The lyophilized material was dissolved in 8.5 M urea/5% 2-mercaptoethanol/5% Pharmalyte (pH 3-10, Pharmacia)/2% Nonidet P-40 (Sigma) and electrophoresed on a cylindrical gel of isoelectric focusing polyacrylamide containing urea and Pharmalyte (pH 3-10), according to the method of O'Farrell [19]. Then, the gel was divided into 43 parts and each of the divided gels was suspended in distilled water at 4°C overnight. The suspension was centrifuged and the pH value of the supernatant was measured. On the other hand, the separate samples of the gel were stained with Coomassie brilliant blue as described above. Each of the stained gels was subjected to a densitometric assay with a TLC scanner (Shimadzu CS-910).

Labeling of restriction fragment A highly repetitive component in the nuclear DNA from rat liver was isolated by HindlII digestion and cloned in pUC9 plasmid [20]. The resulting 370-bp cloned HindlII fragment was subcloned in pUC19 [14]. The clone containing a self-ligated tandem dimer of the 370-bp HindlII fragment was identified through a boiling minilysis procedure [21,22]. The plasmid DNA containing the dimer was isolated in a conventional way and digested with XmnI (NEB) which cleaves only once in the monomer sequence. The digest was extracted with an equal volume of 1 mM Tris-HCl (pH 8.0)/0.1 mM EDTA-saturated phenol/chloroform (1 : 1, v/v) and further with an equal volume of chloroform/isoamyl alcohol (24:1, v/v). The final aqueous phase was subjected to ethanol precipitation. The precipitate was electrophoresed on a slab gel of 1.5% agarose containing ethidium bromide. The 370-bp XmnI fragments were recovered from the gel with a Gene Clean kit (BIO 101) and labeled with [a32p]dCTP (Bresatec) in a T4 DNA polymerase labeling system (BRL), according to the methods described in instruction manuals from the suppliers. The resulting 32p-labeled DNA was expressed as [32p]XmnI frag-

164 ment. The specific activity was adjusted to be 5" 107 cpm//zg DNA.

Binding affinity assay The assay was performed with southwestern blotting according to the method of Miskimins et al. [23]. P123 or P130 (100/xg/ml) was dissolved in 62.5 mM Tris-HC1 (pH 6.8)/10% s u c r o s e / 5 % 2-mercaptoethanol/2% SDS. The solution (10 /xl) was electrophoresed on a slab gel of 0.1% SDS-7.5% polyacrylamide for 3 h and immersed in 48 mM Tris-HC1 (pH 8.3)/39 mM glycine/20% methanol/0.037% SDS for 10 min. The immersed gel was subjected to western blotting on a nitrocellulose membrane in a conventional way. The blotted membrane was immersed in Block Ace (Dainihon Pharmaceutical Co.) for 1 h and incubated in 2 ml of 10 mM Tris-HC1 (pH 7.5)/10 mM MgC12/5% (v/v) Block Ace (binding buffer) containing the [32p]XmnI fragments (3 n g / m l ) and NaC1, at room temperature for 3 h. The concentrations of NaC1 were 50 mM, 0.35, 0.6, 1.0 and 2.0 M. Each of the incubated membranes was rinsed with the binding buffer containing 0.2 M NaCl. The rinsed membrane was air-dried and subjected to an autoradiographic assay. In addition, the [32p]XmnI fragments (3 n g / m l ) were incubated in the binding buffer containing 50 mM NaC1 and distamycin A or chromomycin A 3 (Sigma), at room temperature for 15 min. The concentrations of these antibiotics were 0.3 and 0.6 /zM. On the other hand, the scaffold fraction from the nuclei (2 A260) was electrophoresed and subjected to western blotting on a nitrocellulose membrane, as described above. The blotted membrane was immersed in Block Ace for 1 h and incubated in 2 ml of the binding buffer containing 50 mM NaC1 and the distamycin A-or chromomycin A 3treated [32p]XmnI fragments (3 ng/ml), at room temperature for 3 h. The incubated membrane was rinsed with the same buffer containing 0.2 M NaCl and subjected to an autoradiographic assay.

fraction was counted in a mixture of POPOP, PPO, toluene and Triton X-100.

DNA-bending analysis The analysis was performed according to the method of Wu and Crothers [25]. The cloned plasmid DNA containing a tandem dimer of the 370-bp HindlII fragment was digested with each of DraI (BRL) and XmnI (NEB) which cleave only once in the monomer sequence. The resulting sequence-permuted 370-bp fragments were electrophoretically recovered with a Gene Clean kit as described above. The HindlII and recovered fragments were electrophoresed on a slab gel of 6% polyacrylamide at 4°C or 55°C and the k-factor of each fragment was determined [26]. This factor was defined as the ratio of expected to observed migration distances. In addition, the XmnI fragments (20 ~zg/ml) were incubated in the electrophoretic sample buffer containing 10 mM MgC12 and distamycin A or chromomycin A3, at room temperature for 15 min. The concentrations of distamycin A were 0.1, 0.3, 0.6 and 1.0 /~M. The concentrations of chromomycin A 3

M1234

Sedimentation analysis Each of P123 or P130 ( 1 0 0 / x g / m l ) and P123-P130 mixture (50 /xg e a c h / m l ) was dissolved in 50 mM NaC1/10 mM Tris-HCl (pH 7.4)/1 mM EDTA. The solution (100 ~1) was overlaid on the same buffer in a 10-30% sucrose gradient and centrifuged at 198 000 × g for 24 h. The centrifuged material was fractionated with a density gradient fractionator (ISCO) and subjected to a protein assay according to the method of Lowry et al. [24]. In addition, the [32p]XmnI fragments (2 n g / m l ) were mixed with each of P123 or P130 (100 /xg/ml) and P123-P130 mixture (50/xg e a c h / m l ) in 50 mM or 1.0 M NaC1/10 mM Tris-HC1 (pH 7.4)/1 mM EDTA. The mixture (100 /zl) was centrifuged and fractionated, in the same way. The radioactivity of each

Histones

Fig. 1. Purification of DNA-binding proteins. The nuclear scaffold fraction was filtered through a column of Sephadex G-200. The void fraction was further fractionated by batchwise operations with CMSephadex C-50 and then with DNA-Sepharose 4B. All the resulting fractions were electrophoresed on SDS-polyacrylamide gel and stained with Coomassie brilliant blue. Lane 1, nuclear scaffold fraction; lane 2, void fraction; lane 3, fraction from CM-Sephadex C-50 gel: lane 4, fraction from DNA-Sepharose 4B gel; M, molecular weight marker proteins: from top to bottom, phosphorylase(94000), bovine serum albumin (67000), ovalbumin (43000), carbonic anhydrase (30000), trypsin inhibitor (20100) and lactalbumin (14400).

165

A B 1.0 _ NaCI, M 0

J

M

1

2

FRACTION

2 4 6 8 10 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6

130 000 123 000

Fig. 2. Isolation of DNA-binding proteins. The fraction from DNA-Sepharose 4B gel as shown in Fig. 1 was eluted through a column of the same gel in a 0-1.0 M NaCl gradient. Fraction volume, 1 ml. Aliquots of every second column fractions were electrophoresed on SDS-polyacrylamide gel and stained with Coomassie brilliant blue. (A) Electrophoretic profiles of column fractions. The proteins from column fractions, 4-8 and 12-18, were taken as DNA-binding proteins and designated P123 and P130, respectively. (B) Electrophoretic profiles of P123 and P130. Lane 1, P123: lane 2, P130; M, molecular weight marker proteins; from top to bottom, myosin (200000),/3-galactosidase (116200), bovine serum albumin (67000) and aldolase (42 400).

were 0.1, 0.3, 0.6, 1.0, 10 and 100 /~M. E a c h of the incubated materials was electrophoresed at 4°C or 55°C as described above.

Others All the restriction endonuclease digestions were p e r f o r m e d according to the m e t h o d s described in instruction manuals from the suppliers. T h e D N A content was estimated as 50 tzg/A26 o. T h e protein content was d e t e r m i n e d according to the m e t h o d of Lowry et

of every second column fractions were electrophoresed on SDS-polyacrylamide gel and stained with Coomassie brilliant blue (Fig. 2A). T h e proteins f r o m column fractions, 4 - 8 and 12-18, migrated as single bands to molecular weight positions of 123000 and 130000, respectively, on 0.1% SDS-7.5% polyacrylamide gel (Fig. 2B). Accordingly, these proteins were designated P123 and P130 on the basis of their mobilities. In addition, each of the proteins was electrophoresed on a cylindri-

al. [24].

12

Results

10

•

Purification and isolation of DNA-binding proteins T h e nuclear scaffold fraction was p r e p a r e d from rat liver and subjected to Sephadex G-200 gel filtration. T h e void fraction was further fractionated by batchwise operations with C M - S e p h a d e x gel and then with D N A - S e p h a r o s e 4B gel. All the resulting fractions were electrophoresed on SDS-polyacrylamide ge! and stained with Coomassie brilliant blue (Fig. 1). A very small a m o u n t of histones in the scaffold fraction (lane 1) was not f o u n d in the void fraction (lane 2). The operation with C M - S e p h a d e x gel r e m o v e d almost all the proteins ranging in molecular weight of less than 94000 or higher than about 200000 (lane 3). T h e fraction from D N A - S e p h a r o s e 4B gel exhibited two major protein bands (lane 4). This fraction was further fractionated by the column c h r o m a t o g r a p h y . Aliquots

o 0

5

10

15

MIGRATION, cm

Fig. 3. Determination of pl values. P123 or P130 was electrophoresed on a cylindrical gel of isoelectric focusing polyacrylamide. Then, the gel was divided into 43 parts and subjected to pH assay. On the other hand, the separate samples of the gel were stained with Coomassie brilliant blue and subjected to the densitometric assay, e, pH value; solid curve, densitometric assay of P123 or P130; left peak, P123; right peak, P130.

166 0.12

cal gel of isoelectric focusing polyacrylamide. By the densitomeric and pH assays of the gels, the p l values of P123 and P130 were determined to be 7.2 and 8.1, respectively (Fig. 3).

I

I

I

I

I

I

I

0.10 0.08

Binding affinity assay southwestern blotting revealed that the [32P]XmnI fragment binds readily to P123 or P130 under a hypotonic condition (50 mM NaC1), but not under a hypertonic condition (Fig. 4). In each assay, 1 /zg of the protein was blotted and incubated with 6 ng of the fragment in 2 ml of the buffer. In such a D N A / p r o t e i n ratio, the level of the binding affinity was always higher for P130 than for P123 and was lowered remarkably with increasing NaC1 concentration (Fig. 4C).

|

IIII

0.06

The

0.04 0.02 ,

0.00

I

,

I

20 0.12

'

I

i

IIII

E ¢-

,

I

40 I

,

60 '

I

I

I

,

80 '

I

I

I

,

100 '

120

I

I

0.10

O tO

Sedimentation analysis Each of P123, P130 and P123-P130 mixture was centrifuged in a 10-30% sucrose gradient. All the samples exhibited one peak at the same position in the A 1

2

3

5

1

2

3

130 ¢r

O

03 t'n .<

B 4

0.08 t.d o z 0.06

4

5

ooo

0.04 0.02 I

0.00 0.12

'

I

,

I

,

I

~

40

60

80

I

I

I

I

IIII

I

i

20 I

I

I

,

100

120

I

i

0.10

C

0.08

't

0.06 0.04

C

0.02 0.00

o~ 1 0 0 uJ > I,<-

0

60

D Z Z) O

40

[ 43

20 0 0.05

0.35

12 ~ 0

0.6

40

60

80

100

120

FRACTION NUMBER Fig. 5. Sedimentation analysis of DNA-binding protein. Each of P123, P130 and P123-P130 mixture was centrifuged in a 10-30% sucrose gradient. The centrifuged material was fractionated and subjected to the protein assay. (A) P123; (B) P130; (C) P123-P130 mixture. Vertical bars, positions of S value marker proteins; from left to right, ribonuclease A (1.9 S), chymotrypsinogen A (2.55 S), ovalbumin (3.75 S), bovine serum albumin (4.3 S), aldolase (7.6 S), catalase (11.3 S) and ferritin (17.1 S).

80

d LU rr

20

0.4

1

0

0

0

2

NaCI, M Fig. 4. Binding affinity of [32p]XmnI fragment for P123 or P130. The protein was electrophoresed on SDS-polyacrylamide gel and subjected to western blotting on a nitrocellulose membrane. The blotted membrane was exposed to the [32p]XrnnI fragments in the binding buffer containing NaCI. (A) for PI30; (B) for P123. Lanes 1-5, in 50 mM, 0.35, 0.6, 1.0 and 2.0 M NaCI, respectively; (C) densitometric assay; stippled bar, for P130; hatched bar, for P123.

S value range of 4.3 to 7.6 (Fig. 5). The [32p]XmnI fragment-P123 or P130 mixture was centrifuged in the same way. The radioactivity peak was shown at a position which is heavier than that of ferritin (Fig. 6A or 6B). This experiment was performed with a mixture of 0.2 ng of the fragments and 10/xg of the protein in 100 tzl of the buffer. A mobility shift DNA-binding assay has revealed that such a D N A / p r o t e i n ratio is an

167 1.2

o p t i m u m c o n d i t i o n for t h e analysis ( d a t a not shown here). T h e f r a g m e n t - P 1 2 3 - P 1 3 0 m i x t u r e also b e h a v e d in t h e s a m e m a n n e r (Fig. 6C). H o w e v e r , in 1.0 M NaCI, all t h e r a d i o a c t i v i t y p e a k s s h i f t e d n e a r to t h e p o s i t i o n o f t h e X m n I f r a g m e n t (Fig. 7). T h e m o l e c u l a r weights o f t h e p r o t e i n s a n d t h e f r a g m e n t w e r e e s t i m a t e d on t h e basis o f t h e i r S v a l u e s (Fig. 8). E a c h o f P123, P130 a n d

IIII

I

,

i

i

1.0

i

A

0.8 0.6 0.4

1.4

0,2

1.2

O0 120

1.0 0.8

~

0.6

~

0.4

o

I

0.2

oo

~

:,i

,

,r-

.,

°

i 120

0.6 0.4

: 0 , , ;i

I 4: , 60

8~

1:0

120

C

0.0 0

1.0 0.8 0.6 0.4 0.2 0.0

0.2

0

20

40

60

80

100

120

FRACTION NUMBER Fig. 6. Sedimentation analysis of [32P]XmnI fragment-protein mixture under a hypotonic condition. The fragments were mixed with each of P123, P130 and P123-P130 mixture and centrifuged in a 10-30% sucrose gradient containing 50 mM NaCI. The centrifuged material was fractionated and subjected to the radioactivity assay. (A) Fragment-P123; (B) fragment-P130; (C) fragment-P123-P130, e, mixture; ©, XmnI fragment. Vertical bars, positions of S value marker proteins as shown in Fig. 5.

20

40

60

80

100

120

FRACTION NUMBER Fig. 7. Sedimentation analysis of [32p]XmnI fragment-protein mixture under a hypertonic condition. Each of the fragment-protein mixtures as described in the legend to Fig. 6 was centrifuged in a 10-30% sucrose gradient containing 1.0 M NaCI. The centrifuged material was fractionated and subjected to the radioactivity assay. (A)-(C) as described in the legend to Fig. 6. e, mixture; o, XmnI fragment. Vertical bars, positions of S value marker proteins; from left to right, ribonuclease (A 1.9 S), chymotrypsinogen A (2.55 S), ovalbumin (3.75 S), bovine serum albumin (4.3 S).

P123-P130 m i x t u r e s e d i m e n t e d to a p o s i t i o n o f 6.4 S which c o r r e s p o n d s to a m o l e c u l a r w e i g h t of a b o u t 120000. Such results suggest t h a t P123 a n d / o r P130 m o l e c u l e s d o n o t i n t e r a c t with e a c h o t h e r . T h e X m n I f r a g m e n t a n d t h e f r a g m e n t - P 1 2 3 o r P130 m i x t u r e sedim e n t e d to p o s i t i o n s o f 13 S a n d 21 S, r e s p e c t i v e l y (Fig. 8A). F r o m t h e s e values, t h e f r a g m e n t a n d t h e frag-

168 ment-protein mixture were estimated to have molecular weights of about 290 000 (calcd. 240 000) and about 600 000, respectively (Fig. 8B).

A

1 bp

2

Mabcdefg

Mabcd

ef

ghi

587

DNA-bending analysis 434

The tandem dimer of the 370-bp HindlII fragment was digested with each of DraI and XmnI. The resulting sequence-permuted 370-bp fragments were recovered and then electrophoresed on polyacrylamide gel at 4°C or 55°C (Fig. 9). The mobilities of the fragments were reduced at 4°C (Fig. 9A, lanes a-c). The k-factors of the XmnI, DraI and HindlII fragments were determined to be 1.62, 1.42 and 1.24, respectively. These

267 184

B

1

bp 100

,

,

,

,

,

M ab

c

2 d e f

g

Mabc

de

f

gh

i

587

A

434 267

f 10

3

e

,

I

0

0

20

184

2

I

0

I

40

=

I

60

.

80

I

0

100

120

FRACTION NUMBER 100

I

I

i

•

I

O X

Fig. 9. DNA-bending analysis. A self-ligated tandem dimer of the 370-bp HindIII fragment was digested with each of DraI and X m n l which cleave only once in the monomer sequence. The resulting sequence-permuted 370-bp fragments were recovered and then electrophoresed on polyacrylamide gel at 4°C or 55°C. In addition, the recovered 370-bp X m n I fragments were treated with distamycin A or chromomycin A 3 and electrophoresed in the same way. (A) At 4°C; (B) at 55°C. Lanes a-c in (A) or (B) HindIII, Dral and X m n l fragments, respectively; Lanes d-g in (A-l) or (B-l), with 0.1, 0.3, 0.6 and 1.0 p,M distamycin A, respectively; Lanes d-i in (A-2) or (B-2), with 0.1, 0.3, 0.6, 1.0, 10 and 100/~M chromomycin A 3, respectively.

f "1" (.9

fiq N n-

10

5 b c

0 ..J

o

a i

0

I

5

J

I

10

=

I

15

J

I

20

= 25

S VALUE Fig. 8. Molecular weight estimation by sedimentation analysis. (A) S value determination; I, P123 or P130 and P123-P130 mixture (6.4 S); 2, X m n I fragment (13 S); 3, X m n I fragment-P123 or P130 mixture (21 S). a-g, S value marker proteins; ribonuclease A (1.9 S), chymotrypsinogen A_ (2.55 S), ovalbumin (3.75 S), bovine serum albumin (4.3 S), aldolase (7.6S), catalase (11.3 S) and ferritin (17.1 S). (B) Molecular weight estimation; (1) P123 or P130 and P123-P130 mixture (120 000); (2) X m n I fragment (apparent, 290 000; calcd., 240 000); (3) X m n I fragment-P123 or P130 mixture (600000). a-g, molecular weight marker proteins; ribonuclease A (13700), chymotrypsinogen A (25000), ovalbumin (43000), bovine serum albumin (67000), aldolase (158000), catalase (240000) and ferritin (440000), respectively.

results indicate that sequence-directed bending of the helix axis is the strongest in the XmnI fragment [14,25]. Such anomalously slow gel electrophoretic mobilities were not observed at 55°C (Fig. 9B, lanes a-c). The distamycin A- or chromomycin A3-treated XmnI fragments were electrophoresed on polyacrylamide gel at 4°C. The mobilities were increased with distamycin A, an antibiotic which binds specifically to AT-rich DNA (Fig. 9A-1, lanes e-g), but not with chromomycin A 3 (Fig. 9A-2, lanes d-i). Chromomycin A 3 is well known to be an antibiotic specific for GC-rich sites in DNA [27]. Irrespective of these drugs, the mobilities were not affected at 55°C (Fig. 9B-l, lanes d-g; Fig. 9B-2, lanes d-i). The southwestern blotting revealed that the bindings of the [32p]XmnI fragment to P123 and P130 are inhibited with 0.3 and 0.6 ~M distamycin A, respectively (Fig. 10, lanes b and c), whereas chro-

169

a

b

I

--

c

d

e P130 P123

Fig. 10. Binding affinityof distamycinA- or chromomycinA 3-treated [32p]XmnI fragment for P123 or PI30. The nuclear scaffold fraction was electrophoresed and subjected to western blotting on a nitrocellulose membrane. The blotted membrane was exposed to the antibiotic-treated [32p]XmnI fragments and subjected to the autoradiographic assay. Lane a, without antibiotic; lanes b and c, with 0.3 and 0.6 /zM distamycin A, respectively;lanes d and e, with 0.3 and 0.6/z M chromomycinA 3, respectively.

momycin A 3 was not inhibitory to such bindings (Fig. 10, lanes d and e). Discussion

Our previous work has shown that the binding affinities of a highly repetitive D N A component for rat nuclear scaffold proteins, P123 and P130, depend on the degree of sequence-directed bending of the helix axis [8]. In the present experiments, these proteins have been purified and finally isolated by DNA-Sepharose column chromatography. The southwestern blotting revealed that a highly repetitive bent D N A (370 bp XmnI fragment) from rat-liver nuclei binds readily to the isolated proteins under a hypotonic condition and that the level of the binding affinity for each protein is lowered with increasing NaC1. These results suggest that binding of the fragment to P123 or P130 is held by ionic bonds. Recently, Romig et al. have reported that a constituent of the nuclear matrix and scaffold from H e L a cells, SAF-A, which has an apparent molecular weight of 120 000, binds to AT-rich sites in the human SAR (scaffold attachment region) element and that SAF-A forms large aggregates at low salt concentration [28]. In contrast, P123 a n d / o r P130 molecules did not interact with each other even in 50 mM NaCl, although these proteins were also estimated to have an apparent molecular weight of about 120000. On the other hand, Okazaki and her co-workers have reported that a chromosomal antigen, CENP-B, binds specifically to the 17-bp sequence in a human centromeric a-satellite D N A and suggested that the major complex formed between CENP-B and the D N A con-

tains two molecules of the D N A [29,30]. In our present experiment, the sedimentation analysis predicted that direct interaction between the 370-bp XmnI fragment and P123 or P130 results in the formation of a complex (molecular weight, 600 000) which consists of two of the fragments (molecular weight, 240000 x 2) and one molecule of the protein (molecular weight, 120000), alternatively, one of the fragment and three molecules of the proteins. A mobility shift DNA-binding assay is now supporting the former prediction. The details will be described in our next article. It was reported that a bent D N A fragment from the kinetoplast body of trypanosome parasites is electrophoretically retarded in a polyacrylamide gel and that such an anomalous electrophoretic behavior is due to a stable bent conformation in the fragment which increases difficulty in snaking through polyacrylamide gel pores [31,32]. In addition, it was found that the electrophoretic retardation of the kinetoplast D N A fragment is accentuated at low temperature, whereas at higher than 50°C the fragment shows a normal electrophoretic mobility [26]. Afterward, A runs or A T tracts were demonstrated to cause bending of the D N A helix axis [33-35]. Moreover, Radic et al. have evidenced that a mouse satellite DNA consisting of many A runs is located throughout the centromeric region of the metaphase chromosome and contains a stable curvature which can be alleviated by distamycin A specific for AT-rich D N A [12]. Accordingly, they have inferred that the bend in satellite D N A is recognized by a nonhistone nuclear protein which may be directly involved in centromeric heterochromatin condensation. The 370-bp XmnI fragment has also been sequenced to consist of many A runs or T runs a n d / o r A T tracts (AT content, 62.7%), and has been suggested to have a sequence-directed strong bend at low temperature [8,14]. In the present experiment, we have shown that distamycin A removes the bend in the Xmn I fragment and inhibits binding of the fragment to P123 or P130, whereas neither removal of the bend nor binding inhibition is observed with chromomycin A 3. These results imply that AT-rich regions in a highly repetitive D N A component cause bending of the helix axis to be recognized by some of nuclear scaffold proteins. Thus, interaction between a repetitive bent D N A and nonhistone nuclear protein(s) such as P123 or P130 might play an important role in construction of a higher-order structure essential for mitotic division a n d / o r in centromeric heterochromatin condensation. References 1 Brutlag, D.L. (1980) Annu. Rev. Genet. 14, 121-144.

2 Singer, M.F. (1982) Int. Rev. Cytol. 76, 67-112. 3 Levinger, L. and Varshavsky, A. (1982) Proc. Natl. Acad. Sci. USA 79, 7152-7156.

170 4 Avila, J., Montejo de Garcini, E., Wandosell, F., Villasante, A., Sogo, J.M. and Villanueva, N. (1983) EMBO J. 2, 1229-1234. 5 Strauss, F. and Varshavsky, A. (1984) Cell 37, 889-901. 6 Solomon, M.J., Strauss, F. and Varshavsky, A. (1986) Proc. Natl. Acad. Sci. USA 83, 1276-1280. 7 Harata, M., Ouchi, K., Ohata, S., Kikuchi, A. and Mizuno, S. (1988) J. Biol. Chem. 263, 13952-13961. 8 Hibino, Y., Nakamura, K., Asano, S. and Sugano, N. (1992) Biochem. Biophys. Res. Commun. 184, 853-858. 9 Razin, S.V., Mantieva, V.L. and Georgiev, G.P. (1979) Nucleic Acids Res. 7, 1713-1735. 10 Small, D., Nelkin, B. and Vogelstein, B. (1982) Proc. Natl. Acad. Sci. USA 79, 5911-5915. 11 Asano, S., Hibino, Y., Ikeda, Y., Iwakami, N. and Sugano, N. (1989) Biochem. Int. 19, 871-880. 12 Radic, M.Z., Lundgren, K. and Hamkalo, B.A. (1987) Cell 50, 1101-1108. 13 Martlnez-Balb~s, A., Rodrlguez-Campos, A., Garcla-Ram~rez, M., Sainz, J., Carrera. P., Aymaml, J. and Azorln, F. (1990) Biochemistry 29, 2342-2348. 14 Nakamura, K., Ikeda, Y., Iwakami, N., Hibino, Y. and Sugano, N. (1991) Biochem. Int. 25, 355-362. 15 Hewish, D.R. and Burgoyne, L.A. (1973) Biochem. Biophys. Res. Commun. 52, 504-510. 16 Mirkovitch, J., Mirault, M-E. and Laemmli, U.K. (1984) Cell 39, 223-232. 17 Gasser, S.M. and Laemmli, U.K. (1986) EMBO J. 5, 511-518. 18 Laemmli, U.K. (1970) Nature 227, 680-685. 19 O'Farrell, P.H. (1975) J. Biol. Chem. 250, 4007-4021. 20 Ikeda, Y., Nakamura, K., Iwakami, N., Hibino, Y. and Sugano, N. (1990) Cancer Lett. 55, 201-208.

21 Holmes, D.S. and Quigley, M. (1981) Anal. Biochem. 114, 193197. 22 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual, 2nd ed., pp. 1.29-1.31, CSH Press, USA. 23 Miskimins, W.K., Roberts, M.P., McClelland, A. and Ruddle, F.H. (1985) Proc. Natl. Acad. Sci. USA 82, 6741-6744. 24 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 25 Wu, H-M. and Crothers, D.M. (1984) Nature 308, 509-513. 26 Diekmann, S. and Wang, J.C. (1985) J. Mol. Biol. 186, 1-11. 27 Stankus, A., Goodisman, J. and Dabrowiak, J.C. (1992) Biochemistry 31, 9310-9318. 28 Romig, H., Fackelmayer, F.O., Renz, A., Ramsperger, U. and Richter, A. (1992) EMBO J. 11, 3431-3440. 29 Masumoto, H., Masukata, H., Muro, Y., Nozaki, N. and Okazaki, T. (1989) J. Cell Biol. 109, 1963-1973. 30 Muro, Y., Masumoto, H., Yoda, K., Nozaki, N., Ohashi, M. and Okazaki, T. (1992) J. Cell Biol. 116, 585-596. 31 Marini, J.C., Levene, S.D., Crothers, D.M. and Englund, P.T. (1982) Proc. Natl. Acad. Sci. USA 79, 7664-7668. 32 Marini, J.C., Effron, P.N., Goodman, T.C., Singleton, C.K., Wells, R.D., Wartell, R.M. and Englund, P.T. (1984) J. Biol. Chem. 259, 8974-8979. 33 Koo, H-S., Wu, H-M. and Crothers, D.M. (1986) Nature 320, 501-506. 34 Hagerman, P.J. (1986) Nature 321,449-450. 35 Hagerman, P.J. (1990) Annu. Rev. Biochem. 59, 755-781.