Nucleic Acids Research
Volume 6 Number 10 1979
Cycle specific association of nascent chromatin with nuclear envelope components in Physarum polycephalum
John J.Wille and W.L.Steffens Department of Zoology and Physiology, Louisiana State University, Baton Rouge, LA 70803, USA
Received 28 February 1979 ABSTRACT The action of heparin on isolated nuclei derived from different phases of the mitotic cycle in plasmodia of Physarum ephalum was studied. Heparin addition at two-fold excess over DNA concentration to nuclei in Mg-free low ionic strength buffer (10 mM TRIS-HC1, 10 mM Na2 HPO4, pH=8) releases 60-80% of chromatin from S, G2, and mitotic phase nuclei. The RNA/protein ratio of herparin-solubilized cromatin is constant through S and G2 phases, but rises about two-fold at early prophase coincident with nucleolar breakdown. Purified nuclear envelopes were obtained from heparin-treated nuclei by sedimentation according to Bornens procedures (Nature 244, 28, 1973), and examined by transmission electron microscopy. Residual chromatin is seen at all stages with fine network of DNA fibrils in contact with the envelop. Regardless of time in S 80% of 3H-labeled DNA was released into soluble chromatin with identical 3H/14C ratios. The residual chromatin in nuclear envelopes exhibited a preferential association of early S-DNA in nuclei engaged in early S replication, and late S preferential association in nuclei engaged in late S replication.
INTRODUCTION The polyanion, heparin, produces a variety of effects on isolated nuclei
and chromatin.
It induces nuclear swelling, stimulates chromatin template
efficiency for both endogenous and exogenous RNA polymerase, increases availability of chromatin template for exogenous DNA polymerase, and causes the release of DNA ina highly dispersed state (1).
Heparin-mediated release of
DNA from isolated nuclei has been used to obtain purified nuclear envelopes
(2), and for detecting possible attachment of centrioles to nuclear membranes (3).
The mechanism of heparin-induced release of DNA from chromatin in iso-
lated nuclei has been investigated (4).
It involves the formation of histone-
heparin complexes, which have differing affinities for the five major classes of histones.
At low concentrations, heparin removes histone Hl selectively
C) Information Retrieval Limited 1 Falconberg Court London Wl V 5FG England
3323
Nucleic Acids Research from chromatin, while at increasingly higher heparin concentrations, the core
nucleosome histones (H3-H4, H2A-H2B) are cooperatively released in the form of a heparin-histone complex resistant to mild acid dissociation. The efficiency of DNA release by heparin changes during early interphase in mitotically syn-
chronized Chinese hamster cells (4).
The cell-cycle dependence of DNA release
from nuclei at a given heparin concentration (6) correlates temporally with a
GI-specific phosphorylation event that precedes entry of CHO cells into the Sphase, and which occurs independently of DNA synthesis. The above findings bear strongly on the possible involvement of cell-
cycle specific chromatin modifications, with the putative nuclear membrane events that might trigger the initiation of DNA replication.
To study this
problem, we have investigated the effects of heparin-mediated DNA release on the association of replicating DNA with chromatin-depleted nuclear envelopes from mitotically synchronized macroplasmodial nuclei of the slime mold, Physarum polycephalum.
We report here on the isolation and characterization of
chromatin-depleted nuclear envelopes obtained from early S-phase, late S-phase, G2-phase, and mitotic nuclei obtained from mitotically synchronous plasmodia; and on experiments demonstrating a preferential association of both early S-
phase and late S-phase DNA with the residual nuclear envelope derived from
heparin-treated early S-phase and late S-phase nuclei.
MATERIALS AND METHODS
Synchronous nuclear division:
For experimental use, synchronous macroplas-
modia of Physarum polycephalum, line M3C, were prepared by fusion of micro-
plasmodial fragments as previously described (7).
The first synchronous post-
fusion mitosis (MI) occurred at approximately 6 hours later, with subsequent synchronous nuclear mitoses (M2, M3, etc.) at about 10-12 hour intervals thereafter.
Nuclear staging was done by sampling small pieces of plasmodial tissue
from the edge of macroplasmodia and examining the smear
3324
preparations by phase
Nucleic Acids Research microscope.
The moment of metaphase was chosen as the reference point in
all timing experiments, the S phase beginning within 2 to 3 minutes thereafter.
For labeling of the DNA, at the appropriate time in the experiment,
plasmodia and their underlying millipore filter support were transferred to fresh culture medium containing the radioactive precursor, thymidine (either
14C-thymidine,
0.2 microcuries/ml of culture fluid, or
3H-thymidine,
at 5
microcuries/ml of culture fluid).
Double-isotope labeling proto col: labeled for one full S-phase with
In certain experiments, macroplasmodia were
14C-thymidine
(between cycles M2 and M3) in
order to radioactively label all the nuclear DNA, and then transferred to
medium containing
3H-thymidine
for pulse-labeling of a portion of the subse-
quent S-phase (between M3 and M4).
At the end of the
3H-thymidine
pulse-label-
ing interval, plasmodia were transferred to fresh culture fluid containing unlabeled thymidine (1 mM) to remove excess label and kept on this culture medium
until nuclei were prepared from the prelabeled plasmodia in the following mitoTwo experimental protocols of this basic
tic cycle (between cycles M4 and M5).
design were employed.
In the first design, (designated early S-prelabel), plas-
modia were continuously labeled with
14C-thymidine
for one full S-phase, and
pulse-labeled for the first 60 minutes of the subsequent S-phase, transferred for the remainder of the cycle to "cold" thymidine culture medium for a chase.
Nuclei were prepared on the next cycle at two time-points either after the beginning of the last S-phase, or
phase.
b) 120 minutes into the last S-
In the second design, (designated late S-prelabel) plasmodia were again
continuously labeled with with
a) 30 minutes
3H-thymidine
S-phase; plasmodia
4C-thymidine
for one full S-phase, and pulse-labeled
for 60 minutes when plasmodia were 90 minutes into the second were
again transferred
to
"cold" thymidine medium for the
remainder of that cycle at the end of the pulse interval.
Nuclei were
prepared at the same two collection times (a and b), which corresponded to the time of either the early S-prelabel portion of S-phase, or the late S-
3325
Nucleic Acids Research prelabel of the last mitotic cycle.
Nuclear isolation:
At the appropriate time in the experimental design, nuclei
were isolated by a modification of the procedures of Jockush et al. (8).
Plasmodia were washed once by submerging them in ice-cold distilled water, and homogenized by 10 strokes of a loosely-fitting teflon-glass homogenizer in the following buffer:
10 mM Tris-HCl, pH=8, containing 10 mM MgCl2, 0.25 M sucrose,
and 0.1% Triton-X-100, at 00C. was routinely used.
A ratio of buffer:
plasmodial mass of 15: 1
After homogenization, the unbroken plasmodial tissue and
debris were removed by centrifugation at 5OX g for 10 minutes, and the supernatant containing most of the nuclei filtered through two-layers of cheese-cloth. One volume of the resulting homogenate was layered over one volume of a cushion
of IM sucrose in buffer of the above composition by filling the heavy sucrose
layer from the bottom of the centrifuge tube with a hypodermic syringe. one-step sucrose gradient was spun for 15 minutes at 1,200 X g at 4 C.
sulting pellet contained nuclei free of cytoplasmic contamination.
The The re-
Occasionally,
it was necessary to repeat the one-step sucrose gradient centrifugation step in order to obtain nuclei free of adhering cellular debris.
Purity of nuclear prep-
arations were routinely determined by phase microscopic examination of cell fractions, and by transmission electron microscopic examination of nuclear pellets. The heparin-mediated release of chromatin:
Release of solubilized chromatin
from isolated nuclei and the isolation of chromatin-depleted nuclei was carried out by a modification of the heparin-treatment procedure of Bornens(2).
Nuclear
pellets were resuspended in a Mg-free buffer of the following composition: 10 mM Tris-HCl, 10 mM Na2HPO4, pH=8, and a aliquot of a stock solution of heparin at 5 mg/ml in the same buffer added to give a final concentration of 200
micrograms of heparin or a two-fold excess of heparin over the DNA concentration.
Immediately after addition of heparin, nuclear suspensions were
examined by phase microscopy.
Nuclear swelling and loss of nuclear chromatin
material from intact nuclei were observed to occur within a few minutes. 3326
The
Nucleic Acids Research heparin-treated nuclear suspension also became quite viscous presumably by the release of nuclear DNA, although intact swollen nuclear envelopes were still present.
Kinetic curves for the release of soluble DNA from heparin-treated
nuclei were obtained by UV-spectrophotometric analysis of high speed supernatants of heparin-treated nuclei (37,500 x g to 60 minutes), and were used to
estimate the extent of chromatin release into soluble DNA fraction. Nuclear envelope preparation:
Purified nuclear envelopes were obtained by
diluting the viscous heparin-treated nuclear lysate five-fold with the TrisHCl-Phosphate buffer.
37, 500 X g.
The final suspension was centrifuged for 60 minutes at
The pellet fraction was found to contain intact and broken nu-
clear envelopes, which were used for further biochemical analyses and electron microscopic study.
Deoxyribonuclease treatment of nuclear envelopes:
Washed nuclear envelope
fractions were subjected to DNase I digestion at 50 ug/ml for 15 minutes at 370C. The control untreated nuclear envelope fractions were incubated under the same
conditions with equivalent volumes of Tris-Phosphate buffer added.
At the end
of the enzyme digestion the samples were chilled to 0°C by ice-cold buffer and
filtered through an S + S nitrocellulose filter. out in duplicate.
All digestions were carried
The filters were dried in a hot air oven at
500C
for I hour,
and counted by liquid scintillation spectrometry.
Electron microscopic studies:
Whole plasmodia, isolated nuclei, and nuclear
envelope fractions were obtained as described above and fixed with 2.5% glutar-
aldehyde in 0.1 M sodium cacodylate buffer, pH=7.1. for one hour at
40C,
Fixations were carried out
and the material washed successively with 0.1 M cacodylate
buffer, post-fixed with 1% osmium tetraoxide, and embedded in Araldite plastic, after dehydration steps; and stained both in block and in section with lead
citrate and uranyl acetate.
The blocks were sectioned and examined in a
Hitachi HU II A electron microscope.
3327
Nucleic Acids Research Radioactivity measurements:
The contents of
3H-label
and
14C-label
in intact
purified nuclei, solubilized chromatin, and nuclear envelope fractions were determined by liquid scintillation counting in a Beckman counter.
The samples
were counted in a cocktail containing one part Triton X-100: 2 parts PPO-POPO
in toluene.
3H-cpm 3H-channel).
Corrections were made for channel ratios (0.2%
4C-channel; 21% 14C-cpm spillover into the 3H-cpm and 14C-cpm, and the 3H/ 4C ratios presented
into the of
spillover The values
in the results are
corrected for background counts, and channel ratios.
MATERIALS
Heparin sulfate was obtained from Sigma Chemicals; deoxyribonuclease I from Worthington Biochemicals; and the isotopes
3H-thymidine
(67 Ci/mM) and
14c-
thymidine (54.7 Ci/mM) were obtained from New England Nuclear Corp., Boston, Mass.
RESULTS
The action of heparin on isolated nuclei derived from different phases of the
mitotic cycle in naturally synchronous plasmodia of Physarum polycephalum was studied.
Heparin releases 60-80% of the chromatin DNA from S, G2, and mitotic
phase nuclei.
SDS polyacrylamide gel electrophoresis patterns for solubilized
chromatin yield only the five major histone classes for all phases of the mi-
totic cycle examined.
The RNA: protein ratio of solubilized chromatin was
quantitatively determined.
During the S-phase (M + 1 hour) it is at its
lowest value (normalized to 1.0), remains constant throughout S-phase and most of G2-phase, but increases to 1.55 about 20 minutes before metaphase, coincident with nucleolar migration and breakdown.
Electron microscopic studies on residual nuclear envelopes:
Purified nuclear
envelopes were obtained from heparin-treated S-phase, interphase, late G2-phase
3328
.
Nucleic Acids Research
and mitotic phase nuclei and examined by transmission electron microscopy.
Figure 1 shows a typical nuclear envelope after heparin-treat-
Early S-phase:
ment of early S-phase nuclei (M + 25 minutes).
For comparison, inset (A) shows
the ultrastructural aspect of plasmodial nuclei at the same time in S-phase;
while inset (B) shows the nuclear morphology of early S isolated nuclei just Note the intact inner and outer membranes of the
prior to heparin treatment.
g
,
j
-.M
S
*t
.
.
_-s
#r
*.
-
..
-
i:
.. ,a
.ili
y
-
Pws
.s
'SW ..X.
r
i.
.0-*: .ffi..
,
si
;V%
V:
I
A
C,,+
+
'4
...v@
WS,4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 A1
~~~~ *;f~~~~ ~ ~
X~~~~~~-0
-Md
I
Figure 1. EM micrograph of heparin-extracted early S nucleus. Inset A, in situ plasmodial nucleus at same stage; inset B, isolated nucleus at same stage. C= condensed chromatin, N= nucleolus. Total magnification, 54,100 X; inset A, 10,800 X; inset B, 8,700 X. 3329
Nucleic Acids Research nuclear envelope, nuclear pores, condensed chromatin, interchromatinic areas and nucleolar material in the insets, and the absence of chromatin fibrils and
nucleolar material in the heparin-treated nuclei.
A typical nuclear membrane
is absent, but a structure encloses the nucleus and is similar to the fibrous
lamina of the nuclear cortex described for detergent and salt extracted nuclei
by others (9).
Figure 2 is a micrograph of a typical late S-phase heparin-treatec
Late S-phase:
nucleus.
The inset (top left) shows the corresponding nuclear morphology for
plasmodial late S nuclei (M + 90 minutes into the S-phase).
Note the presence
of a single central reconstructed nucleolus, (N), and the numerous dense chroma-
tin bodies (C) evenly distributed around the nucleoplasm, and a typical double membrane of the nuclear envelope (NE).
In contrast, heparin-treated late S-
phase nuclei have no recognizable nucleolar material, but a network of chromatin-
like fibrils are seen traversing the interior of the nucleus and constituting a
"remnant" nuclear matrix. but many 100
R
The fibrils are predominately 250-300
unit fibrils are also present.
R
in diameter,
Attachment of the chromatin-like
fibrils to the margins of the inner nuclear membrane occur and the fibrils are linked together by ring-like structures, (~ 600 the fibrous matrix.
R)
which are found throughout
The nuclear envelope is clearly present over much of the
nuclear profile, but is interrupted at regions where a fibrous lamina is more apparent. Late G2-phase to early prophase:
early prophase stage nucleus.
Figure 3 is a micrograph of a heparin-treated
The inset (top, left) shows the corresponding
in situ stage in plasmodial nuclei (about 20 minutes prior to metaphase).
The
nucleolus has migrated to the periphery of the nucleus and dense chromatin
bodies are present in the central region of the nucleoplasm.
A microtubular
organizing center (MTOC) of the assembling mitotic spindle is clearly visible in the cup-like depression of the nucleolus (N).
The heparin-treated nuclear
envelopes derived from early prophase nuclei resemble late S, but differ in 3330
;At/B.#;*s>ts&
Nucleic Acids Research
*. shj,-
X X,
**u#Q N
s
s%.
'sf NOx;. ^,
..
..
;F
*'*
e.
:t I
tes g _s\
^
F :@t,,
s _
+Y
¢
P
s
t
*-
*
,'k4
_.r.
%
I
I li
"N
i.
f' I
,,,, f
0-
Figure 2. EM micrograph of heparin-extracted late S phase nucleus. Inset, in situ plasmodial nucleus at same stage. C = condensed chromatin, F = chromatin-like fibrils, M = nuclear membrane, R = ring-like structures. Total magnification, 70,800 X; inset, 11,900 X.
several important respects.
The nuclear membrane is consistently interrupted
by extensive breaks which coincide with the presence of fibrous material re-
sembling heparin-extracted nucleolar material; and
a more
conspicuous fibrillar
matrix is seen attached to the margins of the inner nuclear membrane, or form-
ing a continuous fibrillar network of chromatin fibrils with the fibrous lamina 3331
Nucleic Acids Research
3
,
,N:;t2s,^,^ t
'
.'A
.9
1- .' i,. .4%
.
*
t. Xz>*.t9.
Figure 3. EM micrograph of heparin-extracted early prophase nucleus. Inset, in situ plasmodial nucleus at same stage. MTOC = microtubule-organizing center, F = chromatin-like fibrils, R = ring-like structures. Total magnification, 39,800 X; inset, 16,900 X.
material. Prophase:
Figure 4 is a micrograph of a heparin-treated nuclear envelope de-
rived from prophase nuclei (5 minutes prior to metaphase).
The inset (top,
left) shows the corresponding prophase plasmodial nuclei. Note the condensed chromosome (CC), and absence of nucleolar material. 3332
Typical mitotic spindle
Nucleic Acids Research
4
-..
Figure 4. EM micrograph of heparin-extracted prophase nucleus. Inset, in situ plasmodial nucleus at the same stage. Mt = microtubules, C = chromosomes, F = chromatin-like fibrils, M = nuclear membrane, R = ringlike structures. Total magnification, 44,700 X; inset, 29,370 X.
microtubules (Mt) are also present.
Heparin-treated nuclear envelopes from
prophase nuclei contain a well-developed fibrillar matrix, with chromatin-
like fibrils of approximately 300
X
fibers.
Thinner fibrils are dispersed
between the former and more predominant at prophase than at S-phase.
cleolar material is apparent in the interior of the nuclear envelopes.
No nuThe
3333
Nucleic Acids Research fibrillar network attaches
at many
points to the inner nuclear membrane and
interconnects to foci of more condensed fibrillar material at regions of the
nuclear envelope that are free of typical nuclear double-membranes.
Numerous
ring-like structures are seen both at the fibrous lamina in sections through the nuclear envelope and in the matrix lying in the interior of the nucleus.
Preferential association of replicating DNA with nuclear envelope components: Because plasmodial nuclear DNA is replicated in a fixed order, which is con-
served through successive mitotic S-phases, (10, 11) it is possible to test
whether early S-replicating or late S-replicating DNA molecules show a prefer-
ential association with S phase-specific nuclear membrane components at their
putative times of replication.
Advantage was taken of the technique of heparin-
mediated release of solubilized chromatin from isolated nuclei to obtain purified S-phase-specific chromatin-depleted nuclear envelope preparations, and to detect the existence of an S-phase-specific association of pulse-labeled early S DNA molecules with either early S or late S derived chromatin-depleted nuclear
envelope preparations, or conversely a S-phase-specific association of late S-
replicating DNA molecules with early S or late S-phase derived chromatin-depleted nuclear envelope preparations.
In order to distinguish bulk non-replicating
DNA molecules from phase-specific early S or late S replicating DNA molecules,
the former were labeled for one full S-phase with
14C-thymidine,
and the later
were radioactively-tagged in the next S-phase by pulse-labeling that portion
of the S-phase corresponding to early S (see Methods), or late S (see Methods).
If there is a phase-specific association of early S DNA molecules with early S
nuclear envelope components, and a phase-specific association of late S DNA molecules with late S nuclear envelope components, then these associations
might be less sensitive to removal from nuclear envelopes following heparinmediated release of the solubilized chromatin, and subsequent recovery of the protected complexes in the chromatin-depleted nuclear envelopes. be detected as an elevation of the ratio of 3334
3H-labeled
DNA to
This would
14C-labeled
bulk
Nucleic Acids Research DNA in phase-specific chromatin-depleted nuclear envelope preparations.
I gives the results of two experiment in which the
nuclei
3H/ 14C
Table
ratios of isolated
(A),solubilized chromatin (B),and nuclear envelopes (C) preparations
were determined for both early S-prelabeled experimental design, and late S-
prelabeled design.
Note that regardless of the time in the S-phase when nuclei
were harvested, 80% of the
tin fraction.
3H-labeled
DNA was released into the soluble chroma-
In both the early S-prelabel experiments, no differences were
found between the
3H/ 14C
ratios for either the whole nuclei (Experiment 1, 2.14
versus 2.15; experiment 2, 0.8 versus 0.7), nor in the solubilized chromatin
fraction (experiment 1, 0.83 versus 0.83). preparations, the
3H/ 14C
However, for the nuclear envelope
ratio is consistently higher in the early S nuclear
envelope preparations than in the late S nuclear envelope preparations (experiment 1, 1.0 versus 0.88; experiment 2, 1.06 versus 0.86).
label design the results are less consistent. for an elevated value of the
nation for all fractions.
3H/ 14C
In the late S-pre-
For experiment 1, there is a trend
ratio in the late S phase-specific combi-
Thus, no valid conclusions can be drawn for a prefer-
ential association of late S-replicating DNA with soluble chromatin or nuclear envelope preparations.
However, in experiment 2, no differnece in the H/ 1C
ratios were found between early S and late S whole nuclei (1.79 compared to
1.72), or between early S and late S solubilized chromatin fractions (2.01 compared to 2.07).
Comparison of early S and late S nuclear envelope preparations
reveals an elevated to
3H/ 14C
ratio for late S-phase preparation (0.77 compared
0.96). In other experiments the effects of pancreatic DNase I digestion on the
differential retention of H-labeled DNA in early S versus late S derived
chromatin-depleted nuclear envelopes was studied, and found to abolish the preferential association of early S replicating DNA with phase-specific early S nuclear envelope components
(I,
D of Table
I). Likewise, it abolishes
the
preferential association of late S replicating DNA with phase-specific late S 3335
Nucleic Acids Research
TABLE I PREFERENTIAL ASSOCIATION OF TENPORALLY-CHARACTERISTIC UNITS OF REPLICATION WITH NUCLEAR ENVELOPES IN PHYSARUM
Experimental Design: I.
A.
Early S prelabel
D.
Late S Nuclei
14C-cpm
3H/ 14C
3H-cpm
14C-cpm
3H/ 14C
5949
6920
7392 3239
0.80 2.14
6348 8369
9366 3891
0.70 2.15
1152 5201
1388 2091
0.83 2.49
845 6264
1016 2519
0.83 2.50
Nuclear envelopes Exp. I Exp. 2
1347 1531
1344 1442
1.0 1.1
3098 2043
3519 2389
0.88 0.86
DNase I-treated Nuclear envelopes Exp. 2
989
1395
0.70
1493
2118
0.70
Whole nuclei
Solubilized Chromatin Exp. Exp. 2
C.
Early S Nuclei
3H-cpm Exp. Exp. 2 B.
C-Label and the H/ C Ratio of Radioactive Content of 3H-label, Nuclei and their Fractions After Double-Isotope Labeling their DNA. Stage of Nuclei at Time of Isolation:
Q*-"I*; _'
111-1 -:fiT__ vswvsCL *lUIc UL LbcssaLion;
Late S prelabel A. Whole nuclei Exp. I Exp. 2
II.
Early S Nuclei
3H-cpm
14C-cpm
3H/ 14C
4071 5545
3966 3096
1.10
500 4606
C. Nuclear envelopes Exp. Exp. 2 D. DNase I-treated Nuclear envelopes Exp. 2
Late S Nuclei 3 H-cpm 14 C-cpm
3H/ 14C
1.79
4850 6044
3512
511 2293
1.0 2.01
920 4036
2391
2.74 2.07
7335 1347
6090 1752
1.20 0.77
79 1908
57 2092
1.40 0.96
1190
1665
0.71
1229
1870
0.66
2000
2.43 1.72
B. Solubilized
Chromatin Exp. I Exp. 2
336
derived chromatin-depleted nuclear envelope components (experiment 2, II, D of Table I).
3336
Nucleic Acids Research DISCUSSION
Considerable ultrastructural (12), and biochemical (13) evidence has accumulated for specific chromatin attachment sites to the nuclear envelopes and for the
possible association of newly replicated DNA with nuclear envelope components, (14, 15).
The recent discovery of a nuclear protein framework that persists
after removal of the nuclear membrane, and depletion of chromatin have also re-
vealed an attachment of chromatin-like fibrils to nuclear pore complexes (16,
17). Moreover, Berezney and Coffey (18) have described an association of newly synthesized DNA with a nuclear protein matrix and postulated a role for it in the initiation of DNA synthesis.
Their electron microscopic studies also suggest
that the fibrous matrix framework may be derived from the interchromatinic areas of the nucleus.
Yet, considerable controversy exists regarding the role of nu-
clear membrane components in the control of the spatial and temporal pattern of
eukaryotic DNA replication. Our results suggest that chromatin-depleted nuclei obtained by heparin-
mediated release of histones and chromatin DNA possess a fibrous matrix of chromatin-like fibrils which are interlocked to nuclear pore complexes, and that the structural framework persisting after bulk chromatin release is
enriched in phase-specific replicating DNA molecules.
As pointed out by
Wunderlich (19), the association of nascent replicating DNA with the pore complexes could readily explain the observation that recently initiated DNA molecules are not preferentially associated with the inner nuclear membrane.
In
Physarum, Schel and Wanka (20) have shown that interphase nuclei possess a structural framework interconnected by chromatin fibrils and attached to annular ring-like structures.
Further studies by Schel (21) employing whole mount
and freeze-fracture techniques on isolated nuclei at different phases of the mitotic cycle demonstrate a two-fold increase in the nuclear pore complexes
mainly during the S-phase.
Our results suggest that the early S-nuclear enve-
lope fibrillar matrix is more susceptible to heparin-mediated removal, but by
3337
Nucleic Acids Research late S (M + 60 minutes) the fibrillar network is more resistant, and the frequency of putative pore complexes interconnecting the chromatin-like fibrils increases.
Tentatively, we conclude that the heparin-resistant nuclear envelope
structural framework serves as attachment sites for phase-specific replicating DNA.
It is further speculated that the mode of association may involve the
attachment of initiation sites of the replicating DNA to the pore complexes,
which are seen as ring-like structures persisting after heparin-mediated release of the bulk non-replicating DNA.
Acknowledgments We acknowledge support from an NSF Equipment Grant (PCM 77-08818) to the Department of Zoology and Physiology, Louisiana State University - Baton Rouge.
REFERENCES 1.
2. 3. 4.
5. 6. 7.
8. 9. 10. 11. 12. 13. 14. 15.
16. !7. 18. 19.
3338
Kraemer, R. J. and Coffey, D. S. (1970) Biochim. biophys. acta 224, 568578. Bornens, M. (1977) Meth Cell Biol. 15, 163-175. Bornens, M. (1977) Nature 270, 80-82. Hildebrand, C. E., Gurley, L. R., Tobey, R. A., and Walters, R. A. (1977) Biochim. biophys. acta 477, 295-311. Hildebrand, C. A. and Tobey, R. A. (1975) Biochim. biophys. res. commun. 63, 134-139. Hildebrand, C. E., Tobey, R. A., Gurley, L. R., and Walters, R.A. (1978) Biochim. biophys. acta 517, 486-499. Nygaard, 0. F., Guttes, S. and Rusch, H. (1960) Biochim. biophys. acta 38, 298-306. Jockusch, B. M. and Walker, I. 0. (1974) Eur. J. Biochem 48, 417-425. Davies, H. G. (1967) Nature 214, 208-210. Braun, R., Mittermayer, C. and Rusch, H. (1965) Proc. Nat. Acad. Sci. USA 53, 924-931. Braun, R. and Wili, H. (1969) Biochim. biophys. acta 174, 246-252. Maul, G. G. (1971) J. Cell Biol. 51, 558-563. Yamada, M. and Hanaoka, F. (1973). Nature New Biol. 243, 227-230. Hildebrand, C. E. and Tobey, R. A. (1973) Biochim. biophys. acta 331, 165180. Cabradilla, C. D. and Toliver, A. P. (1975) Biochim. biophys. acta 402, 188-198. Aaronson, R. P. and Blobel, G. (1975) Proc. Nat. Acad. Sci. USA 72, 1007-1011. Scheer, U., Kartenbeck, J., Trendelenburg, M., Stadler, J. and Franke, W. (1976: J. Cell Biol. 69, 1-18. Science 189, 291-293. Berezney, R. and Coffey, D. S. (1975) Wunderlich, F., Berezney, R. and Kleinig, H. (1976) in: Biological membranes. Vol. 3, D. Chapman and D. Wallach, Eds. p 241-333. Acad. Press, New York.
Nucleic Acids Research 20. 21.
Schel, J. H. N. and Wanka, F. (1973) Exp. Cell Res. 82, 315-318. Observation on Interphase Nuclei in Relation to DNA Schel, J. H. N. (1977) Replication. Ph. D. Thesis. 1977. University of Nijmegen, Nijmegen, The Netherlands.
3339
Volume 6 Number 10 1979
Cycle specific association of nascent chromatin with nuclear envelope components in Physarum polycephalum
John J.Wille and W.L.Steffens Department of Zoology and Physiology, Louisiana State University, Baton Rouge, LA 70803, USA
Received 28 February 1979 ABSTRACT The action of heparin on isolated nuclei derived from different phases of the mitotic cycle in plasmodia of Physarum ephalum was studied. Heparin addition at two-fold excess over DNA concentration to nuclei in Mg-free low ionic strength buffer (10 mM TRIS-HC1, 10 mM Na2 HPO4, pH=8) releases 60-80% of chromatin from S, G2, and mitotic phase nuclei. The RNA/protein ratio of herparin-solubilized cromatin is constant through S and G2 phases, but rises about two-fold at early prophase coincident with nucleolar breakdown. Purified nuclear envelopes were obtained from heparin-treated nuclei by sedimentation according to Bornens procedures (Nature 244, 28, 1973), and examined by transmission electron microscopy. Residual chromatin is seen at all stages with fine network of DNA fibrils in contact with the envelop. Regardless of time in S 80% of 3H-labeled DNA was released into soluble chromatin with identical 3H/14C ratios. The residual chromatin in nuclear envelopes exhibited a preferential association of early S-DNA in nuclei engaged in early S replication, and late S preferential association in nuclei engaged in late S replication.
INTRODUCTION The polyanion, heparin, produces a variety of effects on isolated nuclei
and chromatin.
It induces nuclear swelling, stimulates chromatin template
efficiency for both endogenous and exogenous RNA polymerase, increases availability of chromatin template for exogenous DNA polymerase, and causes the release of DNA ina highly dispersed state (1).
Heparin-mediated release of
DNA from isolated nuclei has been used to obtain purified nuclear envelopes
(2), and for detecting possible attachment of centrioles to nuclear membranes (3).
The mechanism of heparin-induced release of DNA from chromatin in iso-
lated nuclei has been investigated (4).
It involves the formation of histone-
heparin complexes, which have differing affinities for the five major classes of histones.
At low concentrations, heparin removes histone Hl selectively
C) Information Retrieval Limited 1 Falconberg Court London Wl V 5FG England
3323
Nucleic Acids Research from chromatin, while at increasingly higher heparin concentrations, the core
nucleosome histones (H3-H4, H2A-H2B) are cooperatively released in the form of a heparin-histone complex resistant to mild acid dissociation. The efficiency of DNA release by heparin changes during early interphase in mitotically syn-
chronized Chinese hamster cells (4).
The cell-cycle dependence of DNA release
from nuclei at a given heparin concentration (6) correlates temporally with a
GI-specific phosphorylation event that precedes entry of CHO cells into the Sphase, and which occurs independently of DNA synthesis. The above findings bear strongly on the possible involvement of cell-
cycle specific chromatin modifications, with the putative nuclear membrane events that might trigger the initiation of DNA replication.
To study this
problem, we have investigated the effects of heparin-mediated DNA release on the association of replicating DNA with chromatin-depleted nuclear envelopes from mitotically synchronized macroplasmodial nuclei of the slime mold, Physarum polycephalum.
We report here on the isolation and characterization of
chromatin-depleted nuclear envelopes obtained from early S-phase, late S-phase, G2-phase, and mitotic nuclei obtained from mitotically synchronous plasmodia; and on experiments demonstrating a preferential association of both early S-
phase and late S-phase DNA with the residual nuclear envelope derived from
heparin-treated early S-phase and late S-phase nuclei.
MATERIALS AND METHODS
Synchronous nuclear division:
For experimental use, synchronous macroplas-
modia of Physarum polycephalum, line M3C, were prepared by fusion of micro-
plasmodial fragments as previously described (7).
The first synchronous post-
fusion mitosis (MI) occurred at approximately 6 hours later, with subsequent synchronous nuclear mitoses (M2, M3, etc.) at about 10-12 hour intervals thereafter.
Nuclear staging was done by sampling small pieces of plasmodial tissue
from the edge of macroplasmodia and examining the smear
3324
preparations by phase
Nucleic Acids Research microscope.
The moment of metaphase was chosen as the reference point in
all timing experiments, the S phase beginning within 2 to 3 minutes thereafter.
For labeling of the DNA, at the appropriate time in the experiment,
plasmodia and their underlying millipore filter support were transferred to fresh culture medium containing the radioactive precursor, thymidine (either
14C-thymidine,
0.2 microcuries/ml of culture fluid, or
3H-thymidine,
at 5
microcuries/ml of culture fluid).
Double-isotope labeling proto col: labeled for one full S-phase with
In certain experiments, macroplasmodia were
14C-thymidine
(between cycles M2 and M3) in
order to radioactively label all the nuclear DNA, and then transferred to
medium containing
3H-thymidine
for pulse-labeling of a portion of the subse-
quent S-phase (between M3 and M4).
At the end of the
3H-thymidine
pulse-label-
ing interval, plasmodia were transferred to fresh culture fluid containing unlabeled thymidine (1 mM) to remove excess label and kept on this culture medium
until nuclei were prepared from the prelabeled plasmodia in the following mitoTwo experimental protocols of this basic
tic cycle (between cycles M4 and M5).
design were employed.
In the first design, (designated early S-prelabel), plas-
modia were continuously labeled with
14C-thymidine
for one full S-phase, and
pulse-labeled for the first 60 minutes of the subsequent S-phase, transferred for the remainder of the cycle to "cold" thymidine culture medium for a chase.
Nuclei were prepared on the next cycle at two time-points either after the beginning of the last S-phase, or
phase.
b) 120 minutes into the last S-
In the second design, (designated late S-prelabel) plasmodia were again
continuously labeled with with
a) 30 minutes
3H-thymidine
S-phase; plasmodia
4C-thymidine
for one full S-phase, and pulse-labeled
for 60 minutes when plasmodia were 90 minutes into the second were
again transferred
to
"cold" thymidine medium for the
remainder of that cycle at the end of the pulse interval.
Nuclei were
prepared at the same two collection times (a and b), which corresponded to the time of either the early S-prelabel portion of S-phase, or the late S-
3325
Nucleic Acids Research prelabel of the last mitotic cycle.
Nuclear isolation:
At the appropriate time in the experimental design, nuclei
were isolated by a modification of the procedures of Jockush et al. (8).
Plasmodia were washed once by submerging them in ice-cold distilled water, and homogenized by 10 strokes of a loosely-fitting teflon-glass homogenizer in the following buffer:
10 mM Tris-HCl, pH=8, containing 10 mM MgCl2, 0.25 M sucrose,
and 0.1% Triton-X-100, at 00C. was routinely used.
A ratio of buffer:
plasmodial mass of 15: 1
After homogenization, the unbroken plasmodial tissue and
debris were removed by centrifugation at 5OX g for 10 minutes, and the supernatant containing most of the nuclei filtered through two-layers of cheese-cloth. One volume of the resulting homogenate was layered over one volume of a cushion
of IM sucrose in buffer of the above composition by filling the heavy sucrose
layer from the bottom of the centrifuge tube with a hypodermic syringe. one-step sucrose gradient was spun for 15 minutes at 1,200 X g at 4 C.
sulting pellet contained nuclei free of cytoplasmic contamination.
The The re-
Occasionally,
it was necessary to repeat the one-step sucrose gradient centrifugation step in order to obtain nuclei free of adhering cellular debris.
Purity of nuclear prep-
arations were routinely determined by phase microscopic examination of cell fractions, and by transmission electron microscopic examination of nuclear pellets. The heparin-mediated release of chromatin:
Release of solubilized chromatin
from isolated nuclei and the isolation of chromatin-depleted nuclei was carried out by a modification of the heparin-treatment procedure of Bornens(2).
Nuclear
pellets were resuspended in a Mg-free buffer of the following composition: 10 mM Tris-HCl, 10 mM Na2HPO4, pH=8, and a aliquot of a stock solution of heparin at 5 mg/ml in the same buffer added to give a final concentration of 200
micrograms of heparin or a two-fold excess of heparin over the DNA concentration.
Immediately after addition of heparin, nuclear suspensions were
examined by phase microscopy.
Nuclear swelling and loss of nuclear chromatin
material from intact nuclei were observed to occur within a few minutes. 3326
The
Nucleic Acids Research heparin-treated nuclear suspension also became quite viscous presumably by the release of nuclear DNA, although intact swollen nuclear envelopes were still present.
Kinetic curves for the release of soluble DNA from heparin-treated
nuclei were obtained by UV-spectrophotometric analysis of high speed supernatants of heparin-treated nuclei (37,500 x g to 60 minutes), and were used to
estimate the extent of chromatin release into soluble DNA fraction. Nuclear envelope preparation:
Purified nuclear envelopes were obtained by
diluting the viscous heparin-treated nuclear lysate five-fold with the TrisHCl-Phosphate buffer.
37, 500 X g.
The final suspension was centrifuged for 60 minutes at
The pellet fraction was found to contain intact and broken nu-
clear envelopes, which were used for further biochemical analyses and electron microscopic study.
Deoxyribonuclease treatment of nuclear envelopes:
Washed nuclear envelope
fractions were subjected to DNase I digestion at 50 ug/ml for 15 minutes at 370C. The control untreated nuclear envelope fractions were incubated under the same
conditions with equivalent volumes of Tris-Phosphate buffer added.
At the end
of the enzyme digestion the samples were chilled to 0°C by ice-cold buffer and
filtered through an S + S nitrocellulose filter. out in duplicate.
All digestions were carried
The filters were dried in a hot air oven at
500C
for I hour,
and counted by liquid scintillation spectrometry.
Electron microscopic studies:
Whole plasmodia, isolated nuclei, and nuclear
envelope fractions were obtained as described above and fixed with 2.5% glutar-
aldehyde in 0.1 M sodium cacodylate buffer, pH=7.1. for one hour at
40C,
Fixations were carried out
and the material washed successively with 0.1 M cacodylate
buffer, post-fixed with 1% osmium tetraoxide, and embedded in Araldite plastic, after dehydration steps; and stained both in block and in section with lead
citrate and uranyl acetate.
The blocks were sectioned and examined in a
Hitachi HU II A electron microscope.
3327
Nucleic Acids Research Radioactivity measurements:
The contents of
3H-label
and
14C-label
in intact
purified nuclei, solubilized chromatin, and nuclear envelope fractions were determined by liquid scintillation counting in a Beckman counter.
The samples
were counted in a cocktail containing one part Triton X-100: 2 parts PPO-POPO
in toluene.
3H-cpm 3H-channel).
Corrections were made for channel ratios (0.2%
4C-channel; 21% 14C-cpm spillover into the 3H-cpm and 14C-cpm, and the 3H/ 4C ratios presented
into the of
spillover The values
in the results are
corrected for background counts, and channel ratios.
MATERIALS
Heparin sulfate was obtained from Sigma Chemicals; deoxyribonuclease I from Worthington Biochemicals; and the isotopes
3H-thymidine
(67 Ci/mM) and
14c-
thymidine (54.7 Ci/mM) were obtained from New England Nuclear Corp., Boston, Mass.
RESULTS
The action of heparin on isolated nuclei derived from different phases of the
mitotic cycle in naturally synchronous plasmodia of Physarum polycephalum was studied.
Heparin releases 60-80% of the chromatin DNA from S, G2, and mitotic
phase nuclei.
SDS polyacrylamide gel electrophoresis patterns for solubilized
chromatin yield only the five major histone classes for all phases of the mi-
totic cycle examined.
The RNA: protein ratio of solubilized chromatin was
quantitatively determined.
During the S-phase (M + 1 hour) it is at its
lowest value (normalized to 1.0), remains constant throughout S-phase and most of G2-phase, but increases to 1.55 about 20 minutes before metaphase, coincident with nucleolar migration and breakdown.
Electron microscopic studies on residual nuclear envelopes:
Purified nuclear
envelopes were obtained from heparin-treated S-phase, interphase, late G2-phase
3328
.
Nucleic Acids Research
and mitotic phase nuclei and examined by transmission electron microscopy.
Figure 1 shows a typical nuclear envelope after heparin-treat-
Early S-phase:
ment of early S-phase nuclei (M + 25 minutes).
For comparison, inset (A) shows
the ultrastructural aspect of plasmodial nuclei at the same time in S-phase;
while inset (B) shows the nuclear morphology of early S isolated nuclei just Note the intact inner and outer membranes of the
prior to heparin treatment.
g
,
j
-.M
S
*t
.
.
_-s
#r
*.
-
..
-
i:
.. ,a
.ili
y
-
Pws
.s
'SW ..X.
r
i.
.0-*: .ffi..
,
si
;V%
V:
I
A
C,,+
+
'4
...v@
WS,4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 A1
~~~~ *;f~~~~ ~ ~
X~~~~~~-0
-Md
I
Figure 1. EM micrograph of heparin-extracted early S nucleus. Inset A, in situ plasmodial nucleus at same stage; inset B, isolated nucleus at same stage. C= condensed chromatin, N= nucleolus. Total magnification, 54,100 X; inset A, 10,800 X; inset B, 8,700 X. 3329
Nucleic Acids Research nuclear envelope, nuclear pores, condensed chromatin, interchromatinic areas and nucleolar material in the insets, and the absence of chromatin fibrils and
nucleolar material in the heparin-treated nuclei.
A typical nuclear membrane
is absent, but a structure encloses the nucleus and is similar to the fibrous
lamina of the nuclear cortex described for detergent and salt extracted nuclei
by others (9).
Figure 2 is a micrograph of a typical late S-phase heparin-treatec
Late S-phase:
nucleus.
The inset (top left) shows the corresponding nuclear morphology for
plasmodial late S nuclei (M + 90 minutes into the S-phase).
Note the presence
of a single central reconstructed nucleolus, (N), and the numerous dense chroma-
tin bodies (C) evenly distributed around the nucleoplasm, and a typical double membrane of the nuclear envelope (NE).
In contrast, heparin-treated late S-
phase nuclei have no recognizable nucleolar material, but a network of chromatin-
like fibrils are seen traversing the interior of the nucleus and constituting a
"remnant" nuclear matrix. but many 100
R
The fibrils are predominately 250-300
unit fibrils are also present.
R
in diameter,
Attachment of the chromatin-like
fibrils to the margins of the inner nuclear membrane occur and the fibrils are linked together by ring-like structures, (~ 600 the fibrous matrix.
R)
which are found throughout
The nuclear envelope is clearly present over much of the
nuclear profile, but is interrupted at regions where a fibrous lamina is more apparent. Late G2-phase to early prophase:
early prophase stage nucleus.
Figure 3 is a micrograph of a heparin-treated
The inset (top, left) shows the corresponding
in situ stage in plasmodial nuclei (about 20 minutes prior to metaphase).
The
nucleolus has migrated to the periphery of the nucleus and dense chromatin
bodies are present in the central region of the nucleoplasm.
A microtubular
organizing center (MTOC) of the assembling mitotic spindle is clearly visible in the cup-like depression of the nucleolus (N).
The heparin-treated nuclear
envelopes derived from early prophase nuclei resemble late S, but differ in 3330
;At/B.#;*s>ts&
Nucleic Acids Research
*. shj,-
X X,
**u#Q N
s
s%.
'sf NOx;. ^,
..
..
;F
*'*
e.
:t I
tes g _s\
^
F :@t,,
s _
+Y
¢
P
s
t
*-
*
,'k4
_.r.
%
I
I li
"N
i.
f' I
,,,, f
0-
Figure 2. EM micrograph of heparin-extracted late S phase nucleus. Inset, in situ plasmodial nucleus at same stage. C = condensed chromatin, F = chromatin-like fibrils, M = nuclear membrane, R = ring-like structures. Total magnification, 70,800 X; inset, 11,900 X.
several important respects.
The nuclear membrane is consistently interrupted
by extensive breaks which coincide with the presence of fibrous material re-
sembling heparin-extracted nucleolar material; and
a more
conspicuous fibrillar
matrix is seen attached to the margins of the inner nuclear membrane, or form-
ing a continuous fibrillar network of chromatin fibrils with the fibrous lamina 3331
Nucleic Acids Research
3
,
,N:;t2s,^,^ t
'
.'A
.9
1- .' i,. .4%
.
*
t. Xz>*.t9.
Figure 3. EM micrograph of heparin-extracted early prophase nucleus. Inset, in situ plasmodial nucleus at same stage. MTOC = microtubule-organizing center, F = chromatin-like fibrils, R = ring-like structures. Total magnification, 39,800 X; inset, 16,900 X.
material. Prophase:
Figure 4 is a micrograph of a heparin-treated nuclear envelope de-
rived from prophase nuclei (5 minutes prior to metaphase).
The inset (top,
left) shows the corresponding prophase plasmodial nuclei. Note the condensed chromosome (CC), and absence of nucleolar material. 3332
Typical mitotic spindle
Nucleic Acids Research
4
-..
Figure 4. EM micrograph of heparin-extracted prophase nucleus. Inset, in situ plasmodial nucleus at the same stage. Mt = microtubules, C = chromosomes, F = chromatin-like fibrils, M = nuclear membrane, R = ringlike structures. Total magnification, 44,700 X; inset, 29,370 X.
microtubules (Mt) are also present.
Heparin-treated nuclear envelopes from
prophase nuclei contain a well-developed fibrillar matrix, with chromatin-
like fibrils of approximately 300
X
fibers.
Thinner fibrils are dispersed
between the former and more predominant at prophase than at S-phase.
cleolar material is apparent in the interior of the nuclear envelopes.
No nuThe
3333
Nucleic Acids Research fibrillar network attaches
at many
points to the inner nuclear membrane and
interconnects to foci of more condensed fibrillar material at regions of the
nuclear envelope that are free of typical nuclear double-membranes.
Numerous
ring-like structures are seen both at the fibrous lamina in sections through the nuclear envelope and in the matrix lying in the interior of the nucleus.
Preferential association of replicating DNA with nuclear envelope components: Because plasmodial nuclear DNA is replicated in a fixed order, which is con-
served through successive mitotic S-phases, (10, 11) it is possible to test
whether early S-replicating or late S-replicating DNA molecules show a prefer-
ential association with S phase-specific nuclear membrane components at their
putative times of replication.
Advantage was taken of the technique of heparin-
mediated release of solubilized chromatin from isolated nuclei to obtain purified S-phase-specific chromatin-depleted nuclear envelope preparations, and to detect the existence of an S-phase-specific association of pulse-labeled early S DNA molecules with either early S or late S derived chromatin-depleted nuclear
envelope preparations, or conversely a S-phase-specific association of late S-
replicating DNA molecules with early S or late S-phase derived chromatin-depleted nuclear envelope preparations.
In order to distinguish bulk non-replicating
DNA molecules from phase-specific early S or late S replicating DNA molecules,
the former were labeled for one full S-phase with
14C-thymidine,
and the later
were radioactively-tagged in the next S-phase by pulse-labeling that portion
of the S-phase corresponding to early S (see Methods), or late S (see Methods).
If there is a phase-specific association of early S DNA molecules with early S
nuclear envelope components, and a phase-specific association of late S DNA molecules with late S nuclear envelope components, then these associations
might be less sensitive to removal from nuclear envelopes following heparinmediated release of the solubilized chromatin, and subsequent recovery of the protected complexes in the chromatin-depleted nuclear envelopes. be detected as an elevation of the ratio of 3334
3H-labeled
DNA to
This would
14C-labeled
bulk
Nucleic Acids Research DNA in phase-specific chromatin-depleted nuclear envelope preparations.
I gives the results of two experiment in which the
nuclei
3H/ 14C
Table
ratios of isolated
(A),solubilized chromatin (B),and nuclear envelopes (C) preparations
were determined for both early S-prelabeled experimental design, and late S-
prelabeled design.
Note that regardless of the time in the S-phase when nuclei
were harvested, 80% of the
tin fraction.
3H-labeled
DNA was released into the soluble chroma-
In both the early S-prelabel experiments, no differences were
found between the
3H/ 14C
ratios for either the whole nuclei (Experiment 1, 2.14
versus 2.15; experiment 2, 0.8 versus 0.7), nor in the solubilized chromatin
fraction (experiment 1, 0.83 versus 0.83). preparations, the
3H/ 14C
However, for the nuclear envelope
ratio is consistently higher in the early S nuclear
envelope preparations than in the late S nuclear envelope preparations (experiment 1, 1.0 versus 0.88; experiment 2, 1.06 versus 0.86).
label design the results are less consistent. for an elevated value of the
nation for all fractions.
3H/ 14C
In the late S-pre-
For experiment 1, there is a trend
ratio in the late S phase-specific combi-
Thus, no valid conclusions can be drawn for a prefer-
ential association of late S-replicating DNA with soluble chromatin or nuclear envelope preparations.
However, in experiment 2, no differnece in the H/ 1C
ratios were found between early S and late S whole nuclei (1.79 compared to
1.72), or between early S and late S solubilized chromatin fractions (2.01 compared to 2.07).
Comparison of early S and late S nuclear envelope preparations
reveals an elevated to
3H/ 14C
ratio for late S-phase preparation (0.77 compared
0.96). In other experiments the effects of pancreatic DNase I digestion on the
differential retention of H-labeled DNA in early S versus late S derived
chromatin-depleted nuclear envelopes was studied, and found to abolish the preferential association of early S replicating DNA with phase-specific early S nuclear envelope components
(I,
D of Table
I). Likewise, it abolishes
the
preferential association of late S replicating DNA with phase-specific late S 3335
Nucleic Acids Research
TABLE I PREFERENTIAL ASSOCIATION OF TENPORALLY-CHARACTERISTIC UNITS OF REPLICATION WITH NUCLEAR ENVELOPES IN PHYSARUM
Experimental Design: I.
A.
Early S prelabel
D.
Late S Nuclei
14C-cpm
3H/ 14C
3H-cpm
14C-cpm
3H/ 14C
5949
6920
7392 3239
0.80 2.14
6348 8369
9366 3891
0.70 2.15
1152 5201
1388 2091
0.83 2.49
845 6264
1016 2519
0.83 2.50
Nuclear envelopes Exp. I Exp. 2
1347 1531
1344 1442
1.0 1.1
3098 2043
3519 2389
0.88 0.86
DNase I-treated Nuclear envelopes Exp. 2
989
1395
0.70
1493
2118
0.70
Whole nuclei
Solubilized Chromatin Exp. Exp. 2
C.
Early S Nuclei
3H-cpm Exp. Exp. 2 B.
C-Label and the H/ C Ratio of Radioactive Content of 3H-label, Nuclei and their Fractions After Double-Isotope Labeling their DNA. Stage of Nuclei at Time of Isolation:
Q*-"I*; _'
111-1 -:fiT__ vswvsCL *lUIc UL LbcssaLion;
Late S prelabel A. Whole nuclei Exp. I Exp. 2
II.
Early S Nuclei
3H-cpm
14C-cpm
3H/ 14C
4071 5545
3966 3096
1.10
500 4606
C. Nuclear envelopes Exp. Exp. 2 D. DNase I-treated Nuclear envelopes Exp. 2
Late S Nuclei 3 H-cpm 14 C-cpm
3H/ 14C
1.79
4850 6044
3512
511 2293
1.0 2.01
920 4036
2391
2.74 2.07
7335 1347
6090 1752
1.20 0.77
79 1908
57 2092
1.40 0.96
1190
1665
0.71
1229
1870
0.66
2000
2.43 1.72
B. Solubilized
Chromatin Exp. I Exp. 2
336
derived chromatin-depleted nuclear envelope components (experiment 2, II, D of Table I).
3336
Nucleic Acids Research DISCUSSION
Considerable ultrastructural (12), and biochemical (13) evidence has accumulated for specific chromatin attachment sites to the nuclear envelopes and for the
possible association of newly replicated DNA with nuclear envelope components, (14, 15).
The recent discovery of a nuclear protein framework that persists
after removal of the nuclear membrane, and depletion of chromatin have also re-
vealed an attachment of chromatin-like fibrils to nuclear pore complexes (16,
17). Moreover, Berezney and Coffey (18) have described an association of newly synthesized DNA with a nuclear protein matrix and postulated a role for it in the initiation of DNA synthesis.
Their electron microscopic studies also suggest
that the fibrous matrix framework may be derived from the interchromatinic areas of the nucleus.
Yet, considerable controversy exists regarding the role of nu-
clear membrane components in the control of the spatial and temporal pattern of
eukaryotic DNA replication. Our results suggest that chromatin-depleted nuclei obtained by heparin-
mediated release of histones and chromatin DNA possess a fibrous matrix of chromatin-like fibrils which are interlocked to nuclear pore complexes, and that the structural framework persisting after bulk chromatin release is
enriched in phase-specific replicating DNA molecules.
As pointed out by
Wunderlich (19), the association of nascent replicating DNA with the pore complexes could readily explain the observation that recently initiated DNA molecules are not preferentially associated with the inner nuclear membrane.
In
Physarum, Schel and Wanka (20) have shown that interphase nuclei possess a structural framework interconnected by chromatin fibrils and attached to annular ring-like structures.
Further studies by Schel (21) employing whole mount
and freeze-fracture techniques on isolated nuclei at different phases of the mitotic cycle demonstrate a two-fold increase in the nuclear pore complexes
mainly during the S-phase.
Our results suggest that the early S-nuclear enve-
lope fibrillar matrix is more susceptible to heparin-mediated removal, but by
3337
Nucleic Acids Research late S (M + 60 minutes) the fibrillar network is more resistant, and the frequency of putative pore complexes interconnecting the chromatin-like fibrils increases.
Tentatively, we conclude that the heparin-resistant nuclear envelope
structural framework serves as attachment sites for phase-specific replicating DNA.
It is further speculated that the mode of association may involve the
attachment of initiation sites of the replicating DNA to the pore complexes,
which are seen as ring-like structures persisting after heparin-mediated release of the bulk non-replicating DNA.
Acknowledgments We acknowledge support from an NSF Equipment Grant (PCM 77-08818) to the Department of Zoology and Physiology, Louisiana State University - Baton Rouge.
REFERENCES 1.
2. 3. 4.
5. 6. 7.
8. 9. 10. 11. 12. 13. 14. 15.
16. !7. 18. 19.
3338
Kraemer, R. J. and Coffey, D. S. (1970) Biochim. biophys. acta 224, 568578. Bornens, M. (1977) Meth Cell Biol. 15, 163-175. Bornens, M. (1977) Nature 270, 80-82. Hildebrand, C. E., Gurley, L. R., Tobey, R. A., and Walters, R. A. (1977) Biochim. biophys. acta 477, 295-311. Hildebrand, C. A. and Tobey, R. A. (1975) Biochim. biophys. res. commun. 63, 134-139. Hildebrand, C. E., Tobey, R. A., Gurley, L. R., and Walters, R.A. (1978) Biochim. biophys. acta 517, 486-499. Nygaard, 0. F., Guttes, S. and Rusch, H. (1960) Biochim. biophys. acta 38, 298-306. Jockusch, B. M. and Walker, I. 0. (1974) Eur. J. Biochem 48, 417-425. Davies, H. G. (1967) Nature 214, 208-210. Braun, R., Mittermayer, C. and Rusch, H. (1965) Proc. Nat. Acad. Sci. USA 53, 924-931. Braun, R. and Wili, H. (1969) Biochim. biophys. acta 174, 246-252. Maul, G. G. (1971) J. Cell Biol. 51, 558-563. Yamada, M. and Hanaoka, F. (1973). Nature New Biol. 243, 227-230. Hildebrand, C. E. and Tobey, R. A. (1973) Biochim. biophys. acta 331, 165180. Cabradilla, C. D. and Toliver, A. P. (1975) Biochim. biophys. acta 402, 188-198. Aaronson, R. P. and Blobel, G. (1975) Proc. Nat. Acad. Sci. USA 72, 1007-1011. Scheer, U., Kartenbeck, J., Trendelenburg, M., Stadler, J. and Franke, W. (1976: J. Cell Biol. 69, 1-18. Science 189, 291-293. Berezney, R. and Coffey, D. S. (1975) Wunderlich, F., Berezney, R. and Kleinig, H. (1976) in: Biological membranes. Vol. 3, D. Chapman and D. Wallach, Eds. p 241-333. Acad. Press, New York.
Nucleic Acids Research 20. 21.
Schel, J. H. N. and Wanka, F. (1973) Exp. Cell Res. 82, 315-318. Observation on Interphase Nuclei in Relation to DNA Schel, J. H. N. (1977) Replication. Ph. D. Thesis. 1977. University of Nijmegen, Nijmegen, The Netherlands.
3339
