Volume 5 Number 12 December 1978
N u c l e i c A c i d s Research
Complex of DNA with chromatin proteins investigated by isopycnic centrifugation in metrizamide
Joanna Rzeszowska-Wolny, J.Filipski, S.Grb'bner and M.Chorazy
Department of Tumor Biology, Institute of Oncology, 44-100 Gliwice, Poland
Received 14 August 1978
ABSTRACT
Complexes of mouse main band DNA with a fraction of non-histone proteins (NHP), having a high affinity for DNA, in
the absenoe ox presence of histones have been investigated
by gradient centrifugation in metrizamide. Two types of
complexes were formed at an input ratio of NHP to DNA
between 1 and 2.5. In metrizamide gradients a majority of
SNA was found In 3 the light complex (at the density of
1.14 - 1.16 g/om ) even at the very high NHP to DNA ratio.
When histones were present in the reaction mixture, most 3of
the DNA was found in the heavy oomplex (1.19 - 1.21 g/om ).
The eleotrophoretio profiles of the proteins recovered from
the heavy and light complexes were different; some fractions
of nonhlstone proteins were present only In the heavy
component.
INTRODUCTION
Nonhlstone proteins (NHP) - DNA Interactions play an
important role in the structure and funotion of ohromatin
(1,2)• Nonhlstone proteins take part in regulation of BNA
transoiiptlon, DNA replication and In various enzymatio
processes in the ohromatin, and maintain the ohromatin and
chromosome struoture (3-8).
We have studied the complexes of DNA with a fraction of
nonhlstone proteins having a high affinity for DNA by
metrizamide gradient oentrifugatlon. The lsopycnlo centrifugation in metrizamide enables separation of oomplexed and
uncomplexed components on the basis of the differences in
buoyant densities between DNA-proteln oomplex, free DNA, and
free proteins (9).
© Information Retrieval Limited 1 Falconberg Court London W1V5FG England
4905
Nucleic Acids Research
MATBBIALS AND METHODS
Isolation of DMA
Moose liver DMA was Isolated following the method of Kay
et al. (10). Labelled SNA was Isolated from mouse kidney
oells cultured in Eagle's medium supplemented with 10 % oalf
serum and containing 12juCi/ml of [/Hj-6-thymidine, specifio
aotivity 20 Ci/mW, (UVWB, CSBB), and 10/ig/ml of chlorocyoline (Polfa, Warsaw, Poland). Main band and satellite DMA
were separated by CSgSO^-Ag* density gradient centrifugatlon
following the prooedure of Corneo et al* (11). In order to
obtain appropriate speoiflo aotivity, labelled DMA was
diluted with nonlabelled UNA. Final activities, measured in
the presence of metrizamlde, were 2 600 opm/ng for satellite
DMA, and 4 500 opm/ng for main band DNA. The moleoular
weight of the DNA was about 9 x 10 daltons.
Isolation of histones and nonhistone proteins
Histones and nonhistone proteins were obtained by ohromatography of reassociated mouse liver chromatin on hydroxyapatite (12). Mouse liver nuolel prepared according to
Vidnell and Tata (13) were extraoted three times with 0.14 M
MaCl, 0.05 M Tris-HCl (pH 7.5). The ohromatln was dlssooiated in 2 M MaCl, 5 M urea, 0.01 M Tris-HCl (pH 8.0) by
homogenizetion in a knife homogenize! and stirred overnight.
Undlssolved fragments were removed by oentrlfugation for
1 hour at 27 000 rpm (Spinoo, rotor 30). A soluble ohromatln
fraction was dialyzed overnight against 13 volumes of distilled water to lower the concentration of MaCl to 0.14 M.
The precipitated ohromatln was obtained by centrifugation
for 30 min at 2 500 x g. Nonhistone proteins from reconstituted ohromatin (about 20 % of all mouse liver nonhistone
proteins) were obtained by ohromatography on hydroxyapatite.
The reconstituted nucleoproteln redlssolved In 0.001 M sodium
phosphate (MaP) buffer (pE 8.0), 2 M NaCl, 5 M urea was sonicated 4 x 10 sec and layered on the top of the 2 x 40 cm
hydroxyapatite oolumn. Histones were eluted with 0.001 M and
NHP with 0.05 M sodium phosphate buffer (pH 8.0), 2 M MaCl,
4906
Nucleic Acids Research
5 II area. The affinity and the speolfiolty of this nonhistone
protein fraotlon fox SNA was determined following the method
of Thomas and Patel (14), and the affinity of binding to
mouse main band UNA and K.coll.DMA was compared.
Isolation of phosphoprotelns
Phosphoprotelns were isolated by the method of Kostraba
et al. (15). Mouse liver nuclei were extracted three times
with 0.1 M NaCl, 0.01 U Tris-HCl (pH 7.4). The pellet was
extraoted three times with 0*2 N HC1, followed by extraction
with a mixtures of chloroform:methanol (1:1, 2:1, v/v) and
extraction with ether* The final pellet was suspended in
0,01 U Tris-HCl (pH 7.4), 0.1 U CHgCOOH, 0.14 U /5-meroaptoethanol and mixed with phenol (1:1, v/v) saturated with the
same buffer. After oentrlfugatlon the phenol phase was
dlalysed for 12 hours against 0.01 U Tris-HCl (pH 7.4),
8.6 M urea and 0.14 U /3-meroaptoethanol, then for 24 hours
against 2 M NaCl, 5 M urea, 0.05 M sodium aoetate buffer
(pH 6.0).
Polyaorvlamlde gel eleotrophoresis
Fractions containing the proteln-DNA complex were
dlalysed against 0.1 % SDS, 0.01 M NaP (pH 7.4), 5.6 mlf
^-meroaptoethanol and subjeoted to eleotrophoresis in 10 %
SOS-polyaorylamide gel as desorlbed by Weber and Osborn (16).
The gels were stained with 0.02 % Coomassie Blue R. After
destalning, the absorption along the gels was measured with
Joyoe-Loebel densltometer. The distribution of |j ij labelled proteins was determined by counting 2 mm gel slices
digested In 1 ml of Soluene 350 (Paokard Instr.Co., Ino.,
USA) in a 3380 Paokard Spectrometer.
Labelling of proteins
1 mC of Na [* i] (Institute for Nuclear Besearch, ^wlerk,
Poland) in 0.1 ml of 0.05 M NaP buffer was added to 4 ml of
protein solution in 0.05 M NaP (pH 8.0), 5 M urea and 2 II
NaCl. Subsequently monoohloramine B in 6 ml of 0,05 M NaP,
5 If urea was mixed with the protein solution. The mixture
was allowed to stand for 2 minutes at room temperature, then
4907
Nucleic Acids Research
the solution of sodium pyxosulphite was added in 5 ml 0.05 M
NaP (pH 8,0), 5 U urea* For eaob mg of protein 4 ng of
•onoehloxaaine and 48 ag of pyrosulphite was used. Labelled
protein solutions were dialysed exhaustively against 2 M
NaCl, 5 If urea, 0.05 If sodium acetate (pH 6.0). The speoiflo
aotivity of histones, measured in the presenoe of metrlzamide
was 3 000 opm/jig. HHP fractions with high affinity for ENA
had a speolfio activity of 2 400 opm/ug, and phosphoprotein
fxaotion 2 100 opm/jog.
Preparation of a BNA-proteln complex
200 ug UNA and appropriate quantities of proteins in
2 U NaCl, 5 U urea, 0.05 U sodium acetate (pH 6.0) were mixed
and adjusted with the same buffer to 1,5 ml, then dialysed
against buffers containing decreasing oonoentration of NaCl:
1, 0.6, 0.4, and 0.15 M NaCl, 5 M urea and 0.05 U sodium
aoetate (pH 6.0). The last dialysis was against 0.15 M NaCl,
0.05 U sodium aoetate (pH 6,0). The recoveries after the
final dialysis were 95 to 99 % for DNA and 60 to 80 % for
proteins.
Pltraoentrifugation in metrizamlde
After final dialysis solutions of DNA-proteln oomplex
were adjusted to 3.75 ml with 0.15 U NaCl, 0.05 M sodium
acetate (pH 6.0), then 1.75 g of metrizamide (Nyegaaxd and
Co.A/S., Oslo, Norway) was added and the samples were centrlfugated for 42 hours at 30 000 xpm, 4°C (Spinco, rotor 40).
Samples were collected by sucking from the bottom, using
a pexistaltio pump. The density of each fraction was determined by measuring the refractive Index using the relation
given by Biokwood and Birnie (17) and correcting the data
for the refraction of the buffer. The recoveries from the
gradient were 80-85 % for DNA and 80-95 % fox proteins.
Determination of radloaotivity
20 jal samples from each fraction of the gradient were
digested with 0.3 ml of Soluene 350 and radloaotivity was
counted In a 3380 Packard Spectrometer (A-B channel 50-250
units, A = 4.48 and EF channel 800 - 1 000 units, A = 4.48).
4908
Nucleic Acids Research
BESULTS AND DISCUSSION
1, Complexes of DNA with nonhlstone proteins
In this work we have studied the oonplexes of nonhistone
proteins with main band DNA, formed under conditions similar
to these used by Riokwood and ifcGlllvray (18), with the
exoeptlon that we have reconstituted the complexes at pH 6.0
and that we have worked with a different nonhistone proteins
fraction. Our fraotion of nonhistone proteins was Isolated
by the method of Patel and Thomas (12). This fraotion had a
high affinity for DNA; 70 % of Its proteins oomplexed with
mouse main band DNA when reconstituted in pH 8*0 following
the method of Thomas and Patel (14). However, Its complexlng
speoifloity was low as it oomplexed to the same extent with
B.ooli DMA.
After the stepwlse dialysis from 2 U NaCl, 5 If urea,
0.05 If sodium acetate (pH 6.0) to 0.15 k NaCl, 0.06 U sodium
acetate (pH 6.0) our NHP fraotion formed two kinds of complexes with the mouse main band DNA. Densities of these
complexes were similar to those obtained by Blokwood and
MoOlllvray (18). Fig. 1 shows the metrlzamlde gradient
oentrlfugatlon profiles of oomplexes obtained at DNA to
protein input ratios of 1:1, 1:1.5, 1:2 and 1:2.5. At all
input ratios two types of oomplexes (light with density of
1*13-1.14 g/om and heavy of 1.20. - 1*22 g/om ) were seen.
With inoreased NHP to DNA ratio, more proteins were found in
the heavy oomplex (from 19 % for DNA:NHP input ratio of 1:1
to 39 % for DNA:NHP ratio of 1:2.5). However, In all the
oases a majority of DNA was found in the light oomponent.
For example at the DNA:NHP input ratio of 1:1 as much as
61 % of DNA was found In the light oomplex, while at DNA:NHP
ratio equal to 1:2.5 the amount of DNA deoreases to 48 %.
It seems that during reoonstltutlon in the first row
protein-protein oomplexing takes plaoe, and subsequently
this oomplex interacts with DNA forming heavy DNA-proteln
oomplex in metrlzamide gradient. Single protein molecules
bound with DNA form the light oomplex.
4909
1145
1-1
50
-
' »
5f 40
30 -
i
1.201
20
o
1.144
Mtf'+NHP*
i
*
t
i
50
1
40
k
i
\
\
50
/i
20
\
1
10
102
4
6
8
10
12
14
16
18
A
/\
1219
-
' \'
20
K
\
\
\
\
80 70 -
60 50 _
40
30 20
10
14
16
-
>
1219
12
MB"+ NHP*
1 : 1.5 .
18
jj
°v
20
2
4
6
A
MEi* + NHP*
1
_
1.134
1
J /
t \
V,
V
• / ^ *
6 10 12 14
16
_
_
IF**
18
20
FRACTION NUMBER
Fig. 1. Metxlzamlde gradient oentxifugation profiles
oomplexes at (A) 1:1, (B) 1:1.5, (C) 1:2 and
(o) [125i] labelled nonhistone proteins, (x)
NHP, nonbistone proteins. Astexies on Fig. 1
labelled.
of mouse main-band DNA-NHP
(O) 1:2.5 input ratios.
[3qDNA. MB, main-band DNA
and Fig. 2 denote compound
Nucleic Acids Research
Blokwood and ifoGillvray (18) have observed considerable
amounts of unoomplexed proteins and WA after the reoonstltutlon In 0.14 11 NaCl. In oar conditions tbe amount of
anooaplexed material Is negligible* It seems tbat this
results from tbe difference In binding affinity between their
HAP2 and oar high affinity NHP fraction. Oar high affinity
NHP fraction forms complexes with DNA at higher ionlo
strength than HAP2 fraotlon. For example, In 0.3 If NaCl HAP2
fraotlon did not bind to IMA at all, whereas with our NHP
fraotlon no unoomplexed material in 0.25 M NaCl was observed
(data not shown). Thus, our proteins have tbe binding
characteristics more similar to the histones than to HAP2
fraotlon.
Fig. 2 shows the metrlzamlde gradient oentrifugatlon
profile of nonhlstone phosphoproteln fraotlon to mouse main
band UNA at the Input ratio of 1:1; the conditions of
reoonstltution were like those in tbe oase of high affinity
NHP fraction.
Phospboproteins and UNA form a heavy complex with a density in metrlzamlde of 1.224 g/om • A small amount of the
proteins bound to the bulk of UNA forms a second light
complex with a density of 1.132 g/cm3. From tbe gradient of
metrlzamide we reoovered 89 % of SNA, and only 60 $ of
phosphoprotelns. Since phosphoproteins are known to aggregate very easily, we suggest that this feature is responsible
for the formation of a oomplex having a density greater than
tbat found fox heavy oomplex of UNA with high affinity
nonhlstone proteins. In some oases (see e.g. Fig. 1C) a heavy
fraotion containing a small amount of DMA associated with
a significant amount of proteins could be observed. It can be
oaused by a contamination of a metrlzamlde gradient with some
material from the bottom of the oentrlfuge tube during the
collection of fractions.
2. Complexes of WA with his tones and nonhlstone proteins
During the reconstltutlon of nucleohlstone some WA molecules bind more hlstone than others (21). This effeot oan be
explained chiefly by the cooperatlvity of oomplex formation,
4911
Nucleic Acids Research
-
5 40
1132
O
30 -
-
/ i
§ 10 C£
O
<c
o
_
\
1.224
1
10-
v
i
^
\
\
•r
A
" " • ^ \ _
c
6
10
^ \
VcyO
l
>
14
x
vJ 18
FRACTION NUMBER
F i g . 2. Metrizamlde gradient oentrifugation profile
of mouse main band DNA-phosphoproteins
complex at 1:1 input rat^Lo (o) [ i 2 5 l ] labelled
phosphoproteins, (x) n~
but preferential binding of some bistones to regions
enriobed in AT or GC could also be of importance (19,20).
When tbe concentration of hiatones in tbe reaction mixture
is sufficiently low, they form fibrillar structures with DNA.
By inoreaslng tbe ooncentratlon of histones, more heavy
ooaplexes with globular structure are obtained (21).
It has been postulated that this heavy complex results
from the binding of excess of bistone to other blstone molecules* The histone-histone interactions are assumed to be
essential for the formation of the globular type of complex.
This kind of interaction may be similar to that existing in
native obromatin where his tones are assembled in the form of
nucleosomes. Thus the study of the Influence of bistones on
the DNA-high affinity NHP complex formation seems to be of
Interest. In this study the ratio of histone to DNA was
oonstant, and the amount of nonhistone proteins was ohanged.
The centrifugation profile of the DNA-hlstone-NHP mixed at
the input ratio of 1:0.8:0.5 shows two (heavy and light) oom4912
Nucleic Acids Research
ponents (Fig. 3 ) . The density of the heavy fraction
3
3
(1.211 g/em ) is very close to the 1.220 g/em density of the
I labelled mouse liver ohromatln, sedlmented in_ the sajie
oondltlons (data not shown). Increasing the input ratio of
HHP to DNA the amount of both DNA and hlstones in the heavy
component lnoreases and the amount of DNA and proteins In the
light complex diminishes considerably (Fig. 3 A,B,C). For
example, at a UNA to NHP ratio of 1:1.5 In the presence of
histones (the DNA to total proteins ratio was 1:2.3, Fig. 3C),
as much as 77 % of ONA is found in the heavy oomponent
(1.208 g/om3) and the rest of- the WA (23 %) Is found in the
light component. The oomplex formed in the absence of hlstones at a WA to NHP ratio of 1:2.5 pontained only 20 % of
a total WA in the heavy component (Fig. 10), In spite of the
faot that the total quantity of proteins was greater than in
the system with histones In the mixture (Fig. 3C). Thus the
presenoe of histones favours the formation of the heavy
component, and it Is possible that the histones which are
known to form strong complexes with the UNA could play a role
of "aggregation centres" for nonhistone proteins during the
reoonstltution process.
The profiles of polyaorylamide gel eleotrophoxesis of
the proteins recovered from the light and heavy components
of ENA-histones-NHP mixed at the ratio of 1:0.8:0.5 show that
the distribution patterns of total|_IJ labelled histones in
these components are similar (Fig. 4 ) . The distribution patterns of total proteins in the heavy and light components
are different (Fig. 5); there are some NHP fractions present
in the heavy oomponent (marked by arrows on Fig. SB) and
absent In the light one.
The oentrifugatlon profile of the oomplex formed in the
mixture of satellite DNA-hlstones-MHP in the ratio 1:0.8:0.5
(Fig. 30) differs from that formed with the main band DNA
(Fig. 3A) under the same conditions. With satellite DNA the
majority of histones and DNA are found In the heavy peak
(Fig. 3D), This profile resembles the one obtained In the
experiment llustrated in Fig. 3C, where main band DNA was
4913
Nucleic Acids Research
o
DMA OR PROTEIN CONTENT
o
o
-b.
O
o
g S
to-jK
"MB
oo -jo
oj
1.1'
>
_z
U3
ODTf
P
8|
1
4914
i
1
I
.
1
1
1
z
Nucleic Acids Research
3O00
SLICE NUMBER
Fig. 4. Bleotrophoretio profiles of the I l] labelled
histones present in (A) light (fractions 8-10)
and (B) heavy (fractions 12-15) component from
the gradient shown on Fig. 3A. NBP were not
labelled In this experiment.
complexed with histones and with a three-fold greater
quantity of NHP.
CONCLUSIONS
A fraction of nonhistone proteins, having high affinity
for BNA, forms two kinds of complexes with UNA. We suggest
that the light complex arises by association of a single protein molecules with UNA. Binding of protein-protein
complexes to UNA brings about the formation of the heavy
oomplex. We suppose that binding of these protein aggregates
Is responsible .tor the comparatively low reoovery of ENA
from the heavy fraction even at the high NHP:IMA ratio.
4915
Nucleic Acids Research
Z
o
If)
w
Z.
<
GEL LENGTH
Fig. 5. Eleotrophoretlo profiles of the total proteins
recovered from (A) light and (B) heavy components of main-band ENA-histone-NHP complexed
at the input ratio 1:0.8:0.5 (the densitograms
from the polyaorylamlde gels shown in Fig. 4.
Histones present in the reaction mixture favour the formation of the heavy complex with ENA. Histones probably
serve as an "aggregation centre" fox nonhistone proteins in
the reoonstitutlon process. Some of the nonhistone proteins
can bind to UNA only in the form of aggregates (protein-pxoteln complexes). They are absent fxom the light complex.
ACKNOWLEDGEMENTS
The authoxs axe indebted to Dr D. Biokwood fox the
helpful discussion. The technical assistance of Mrs K.
Chora^y is acknowledged. Nyegaaxd and Co. A/S (Oslo, Norway)
4916
Nucleic Acids Research
kindly supplied as with a sample of metxizamlde.
This work was partly carried out within the research
projeot 09.7,1. of the Polish Academy of Sciences, and
partly supported by National Cancer Program PB-6/1301.
BBFEBENCES
1. Bradbury, B.M. and Biophysics Group (1975) in:
The Struoture and Function of Chxomatin, pp. 131-148
Elsevler, Excexpta Medloa Ass.Soi.Publ., Amsterdam,
Oxford, Mew York
2. Elgin, S.L.R. and Weintraub, H., (1975) Ann.Bey.
Bloohem. 44, 726-774
3. Johnson, B.M* and Hodden, J.W. (1975) Science 187,
1198-1200
4. Klukas, C.K. (1976) Nature 263, 545-546
5. Kleinsmlth, L.J., Stein, J. and Stein, G. (1975)
in Chromosomal Proteins and Their Bole in the
Regulation of Gene Expression
(Ed. Stein, G.S. and Kleinsmlth, L.J.) 59-67,
Academic Press Inc.
6. Comings, D.E. (1972) in Advances in Human Genetic
(Ed. Harris, H. and Hirsohorn, K.) vol. 3, 267-276
7. Adolph, K.W., Cheng, S.M. and Laemmli, U.K. (1977)
Cell 12, 85-816
8. Paulson, J.B., and Laemmli, U.K. (1977) Cell 12, 817-828
9. Bickwood, D. (1976) In: Biological Separation in
Iodlnated Density Gradient Media, pp. 27-40,
Information Betrival Ltd, London, Washington
10. Kay, E.B.M., Simmons, N.S. and Oounce, A.L. (1952)
J.Am.Chem.Soo. 74, 1724-1726
11. Corneo, G., Ginelli, E., Soave, C. and Bernard!, G.
(1968) Biochemistry 7, 4373-4379
12. Patel, G.L., and Thomas, T.L. (1973) Proc.Nat.Acad.Sol.
70, 2524
13. Widnell, C.C. and Tata, J.B. (1964) Bioohem.J. 92,
313-317
14. Thomas, T.L. and Patel, G.L. (1976) Biochemistry 15,
1481-1489
15. Kostraba, N.C., Montagna, B.A. and Wang, T.y. (1975)
J.Biol.Chem. 250, 1548-1555
16. Weber, K. and Osborn, M. (1969) J.Biol.Chem. 244,
4406-4412
17. Rlokwood, 0. and Birnie, G.D. (1975) FEBS Letters 50,
102-110
18. Rickwood, D. and MacGlllvray, A.J. (1977) Expt.Cell.Res.
104, 287-292
19. Sponar, J. and Sormova, Z. (1972) Eur.J.Biochem. 29,
99-103
20. Clark, R.J. and Felsenfeld, G. (1972) Nature New BJol.
240, 226-229
21. Rickwood, D., Birnie, G.D. and MaoGlllvray, A.J. (1975)
Nucleic Acids Bes. 2, 723-733.
4917
Nucleic Acids Research
4918