ABSTRACT
Upon the onset of mouse myoblast differentiation there is a rapid drop in DNA
methyltransferase activity followed by a genome wide demethylation [Jost and
Jost (1994)
J. Biol. Chem
. 269, 10040-10043]. Here we show by using specific antibodies directed against DNA
methyltransferase that upon differentiation there was a rapid drop in nuclear
DNA methyltransferase whilst the internal control histone H1 remained constant.
The loss of nuclear methyltransferase was not due to a translocation of the
enzyme from the nucleus to the cytoplasm where there was an increase in
creatine phosphokinase protein.
In vitro
run on experiments carried out with growing and differentiating myoblast nuclei
showed no difference in the rate of DNA methyltransferase mRNA synthesis. As
measured by Northern blot hybridization the relative half life of DNA methyltransferase mRNA in growing and differentiating cells in the presence of Actinomycin D was 5 h and 1 h 30 min respectively, whereas in the same cells the half life of histone H4
mRNA was in both cases 80 min. As measured by a combination of pulse chase experiments with labeled leucine and immunoprecipitation, the relative half-life of DNA methyltransferase in growing and differentiating cells was
~18 h and 4 h 30 min respectively.
In cultured G8 mouse myoblast cells there is an initial phase of cell proliferation followed by terminal differentiation which is characterized
by the fusion of myoblast cells to form multinucleated muscle fibres (for reviews see
1
-
4
). Just before the onset of cell differentiation there is a cessation of DNA synthesis and cell division followed by an increase in the activity of muscle specific marker genes such as
creatine phosphokinase, myogenin and MyoD1 (
1
-
4
). We have previously shown that in differentiating mouse myoblasts there is a
transient DNA demethylation which is correlated with a high activity of 5-methyldeoxycytidine excision repair (
5
). Preceeding the excision repair of 5-methylcytosine there is also an abrupt drop in DNA methyltransferase
activity. The activity of DNA methyltransferase shows marked changes as cells
pass through the cell division cycle (
6
-
11
). Activity is low in early G
1
phase, it rises to a peak in S phase and then falls dramatically at mitosis. The rise in activity in late G
1
/early S phase is correlated with an increase in the amount of DNA
methyltransferase mRNA (
6
) where transcriptional and posttranscriptional events may play an essential
role. Here we show that the down regulation of DNA methyltransferase in differentiating mouse myoblasts is occurring essentially at the posttranscriptional and posttranslational levels.
The mouse myoblast G8 cell line was purchased from the American Type Culture Collection. Collagen, rabbit skeletal creatine kinase and aprotinin were obtained from Boehringer. Dulbecco's Modified Eagle's
Medium (DMEM) and fetal calf serum (FCS) were obtained from Gibco. The
Escherichia
coli
expression system was purchased from Novagen. Acrylamide and bisacrylamide were obtained from Serva. Hydrobond
TM
-N, ECL immunoblot detection kits, [[delta]-
32
P]ATP, [[alpha]-
32
P]UTP (3000 Ci/mmol) and L-4,5[
3
H]leucine (162 Ci/mmol) were obtained from Amersham. Dextran sulfate and RNase inhibitors were purchased from Pharmacia. Peroxidase linked goat anti-rabbit IgGs were obtained from BioRad. Dithiothreitol (DTT) and spermidine were obtained from Fluka. All other chemicals were supplied by Merck.
A subclone of a G8 cell line was used for all of the experiments. The cells were
cultured on collagen coated plates in DMEM plus 10% fetal calf serum (FCS), 10%
horse serum and 4.5 g/l glucose (growth medium GM). Differentiation was induced when the cells reached 80-90% confluency by reducing the serum concentration to 2% FCS
(differentiation medium DM).
A
Hin
cII-
Bam
HI fragment of the mouse cDNA encoding part of the N-terminal domain of DNA methyltransferase (amino acid positions 309-847) (
12
) was inserted into the expression vector pET29a(+) and cloned in
E.coli.
The DNA methyltransferase protein was purified on a His.tag affinity column.
Serum from the immunized rabbits was partially purified by ammonium sulfate
precipitation (
13
). Antiserum against skeletal creatine kinase was prepared the same way.
Cells were lysed by vortexing twice for 20 s at 4oC in 10* p.c.v. (packed cell volume) of 10 mM Tris-HCl pH 7.6, 10 mM KCl, 0.5 mM EDTA, 1 mM DTT, 10 [mu]g Aprotinin/ml, 1 mM Na-metabisulfite, 0.5 mM phenylmethylsulfonyl fluoride
(PMSF) and 0.5% NP-40. After centrifugation for 5 min at 2000 r.p.m. at 4oC in a clinical centrifuge, the cytoplasm was separated from the
nuclei. The nuclei were washed twice in the same buffer and lysed in 4* p.c.v. of nuclear lysis buffer containing 25% glycerol, 20 mM Tris-HCl pH 7.6, 450 mM KCl, 0.5 mM EDTA, 1mM DTT, 1mM PMSF, 10 [mu]g/ml Aprotinin, 5 mM spermidine and incubated on ice for 30 min. The chromatin was sedimented for 10 min at 17
000 r.p.m. at 4oC in a microfuge. The supernatants were stored at -80oC.
Proteins (20 [mu]g) were loaded onto a 7.5% SDS-polyacrylamide gel. After electrophoresis, proteins were transferred
onto Immobilon
TM
P membranes (Millipore) with a semi-dry blotter. The membranes were blocked in 10% milk plus 0.1% Tween 20 in
phosphate buffered saline (PBS) either overnight at 4oC or for 1 h at room temperature. Primary antisera were added (dilution
1:1000 in the blocking buffer) and incubated for 2 h at room temperature. Goat
anti-rabbit IgGs conjugated to peroxidase were diluted 1:5000 in the same
buffer and incubated for 1 h at room temperature. Filters were then washed and
developed according to the ECL detection protocol of Amersham.
Total cellular RNA was prepared by the guanidium thiocyanate procedure (
14
). Aliquots of 10 [mu]g of samples were loaded on a 1% formaldehyde agarose gel and
electrophoresis was carried out at 70 V for 2 h. DNA was transferred onto
Hybond
TM
N nylon membranes (Amersham) according to the procedure of Sambrook
et al.
(
15
). The membranes were then UV cross-linked using a Stratagene UV cross linker set for auto-cross linking. Prehybridization was done for 6 h at 65oC in 1 M NaCl, 10% dextran sulfate, 1% SDS and 100 [mu]g/ml of sheared, denatured herring sperm DNA. DNA probes
were labeled by random priming (
15
). Hybridization was carried out for 16 h at 65oC with 1-3 * 10
6
c.p.m./ml of labeled probe. The membranes were washed twice in 2* SSC plus 1% SDS at room temperature and then for a further 20-30 min at 50oC in 0.1* SSC plus 0.1% SDS. Filters were either exposed to X-ray film or processed in a molecular Dymamics
PhosphoImager for quantitation.
G8 cells were collected and washed twice with cold PBS. The cells were lysed in
10 mM Tris-HCl pH 7.6, 10 mM NaCl, 3 mM MgCl
2
and 0.5% NP-40 by vortexing twice for 20 s at 4oC followed by a centrifugation at 2000 r.p.m. for 5 min. The nuclei
were resuspended in glycerol storage buffer containing 50 mM Tris-HCl, pH 8.0, 40% glycerol, 5 mM MgCl
2
and 0.1 mM EDTA. The relative number of nuclei was determined from the OD 260
nm reading of an aliquot lysed in 0.2 M NaOH plus 1% SDS. Each nuclei sample
was diluted to a final concentration of 1 [mu]g/[mu]l DNA with glycerol storage buffer. The diluted nuclei suspension (100 [mu]l) was mixed with an equal volume of 2* reaction buffer containing 10 mM Tris-HCl, pH 8.0, 5 mM MgCl
2
, 0.3 M KCl, 5 mM DTT and 1 mM each of ATP, CTP and GTP in the presence of 40 U
of RNase inhibitor and 100 [mu]Ci of [[alpha]-
32
P]UTP. Incubation was carried out for 30 min at 30oC. Where indicated, nuclei were preincubated for 20 min on ice with 3 [mu]g/ml of [alpha]-amanitin in the absence of the nucleotides triphosphate.
Reaction was started by the addition of all four nucleotides triphosphate and
incubation was continued for 30 min at 30oC. Total RNA was extracted as indicated above. Using a slot blot hybridization system with Hybond
TM
N nylon membranes, 2.5 [mu]g of immobilized DNA was hybridized with 1-2 * 10
6
c.p.m. of labeled RNA. The hybridization buffer contained 200 [mu]g/ml
E.coli
tRNA, 10 mM EDTA pH 8.0, 10 mM PIPES pH 6.5, 5* SSC, 5* Denhardt's solution and 0.1% SDS. Hybridization was performed in
a rotating hybridizer at 45oC for 3 days. The filters were washed twice for 15 min in 5* SSC plus 0.2% SDS at room temperature, followed by two washes in 2* SSC plus 0.2% SDS at room temperature. Membranes slices were
then assembled and exposed to a PhosphoImager screen for quantitation.
pMG is a nearly full length cDNA for murine DNA methyltransferase (EMBL accession no. X14508). The 5' end is an
Eco
RI site (nucleotide position 25) and the 3' end is a
Bgl
II site (nucleotide position 4960). pZAPactI is pBluescript carrying the murine
cDNA for [beta]-actin and pMyoG is pBluescript containing a 800 bp murine cDNA
coding for myogenin which can be spliced out by double digestion with
Xba
I and
Hin
dIII. This fragment was used as a probe in Figure
3
B.
G8 cells were grown until they were 80-90% confluent. Two of the plates were washed once with cold PBS and
frozen in liquid nitrogen and stored at -80oC. The rest of the plates were further incubated in growth medium or
in differentiation medium in the presence of 4 [mu]g/ml Actinomycin D. With this particular cell line a titration curve
indicated that 4 [mu]g/ml of actinomycin D could inhibit RNA synthesis for a maximum of 2-3 h only. Higher concentrations of Actinomycin D proved toxic to the
cells. At different time points two plates from each group were washed with
cold PBS and stored at -80oC as indicated above. Total cellular RNA was extracted as indicated
in Materials and Methods. For each time point 10 [mu]g of RNA was separated on a 1% formaldehyde agarose gel and Northern blot
hybridization was carried out as indicated above using a 600 bp
Bam
HI fragment from the mouse DNA methyltransferase cDNA as a probe. The same
membranes were stripped by washing them twice in a boiling solution of 0.1%
SDS. The membranes were then rehybridized using the histone H4 DNA as a probe.
Radioactive bands were visualized by using a PhosphoImager and the ImageQuant software (Molecular Dynamics).
G8 myoblast cells (80% confluent) were pulse labeled with 50 [mu]Ci/ml of [
3
H]leucine in 5 ml medium, deficient in leucine (high serum) per plate for 20 h (the length of labeling was determined experimentally). Upon
removal of the medium and washing of the plates with PBS, half of the plates
were supplemented with growth medium and the other half with differentiating
medium supplemented with a 1000-fold excess of non-labeled leucine. The cells were incubated at 37oC and collected at the indicated time. They were then washed in
PBS and homogenized with a glass-glass Dounce homogenizer in 0.5 M KCl, 10 mM HEPES pH 7.5, 3 mM MgCl
2
, 0.1 mM EDTA, 10% glycerol, 2 mM benzamidine and 0.1 [mu]g/ml DNase I. After homogenization, the samples were incubated at room
temperature for 20 min and cell debris was sedimented at 17 000
g
for 10 min. Immunoprecipitation was carried out using 0.5-1 * 10
6
d.p.m. of
3
H-labeled protein in 50 [mu]l of buffer containing 0.25 M NaCl, 20 mM HEPES pH 7.5 in the presence
of 5 [mu]l of the ammonium sulfate fractionated antibodies directed against DNA
methyltransferase. Controls received the same amount of pre-immune serum. After incubation at 37oC for 20 min and at room temperature for 40 min each sample was mixed
with a suspension of 50 [mu]l of protein A-Sepharose beads. Incubation was carried out for 1 h at room
temperature and the beads were collected by centrifugation a few seconds at
7000 r.p.m. in a microfuge. The supernatants were removed and the beads were
washed eight times in 1 ml of 2 M NaCl containing 1% Triton X-100, 20 mM HEPES pH 7.5, 5 mM EDTA and 5 mg/ml bovine serum albumin. Bound
proteins were solubilized in 100 [mu]l formic acid and counted for radioactivity.
A difference in the level of DNA methyltransferase protein during myoblast
differentiation could be explained by either a change in the rate of synthesis
of its mRNA and/or a change in its turnover. A possible change in the rate of
synthesis of DNA methyltransferase mRNA was tested by nuclear run on experiments. Experiments were carried out as indicated in Materials and Methods. The upper
panel of Figure
2
A shows that with the DNA methyltransferase pMG probe there was no difference in
the rate of DNA methyltransferase transcription before and during myoblast
differentiation. During the same time, [beta]-actin mRNA showed a slight increase in hybridization signals whereas
the negative control [pBluescript SK(+) without an insert] gave no hybridization. The lower panel of Figure
2
A shows the quantitation of DNA methyltransferase mRNA hybridized to the immobilized pMG probe. Because
of the complexity of the RNA used as a labeled probe, it was necessary to rule
out any nonspecific labeling and nonspecific hybridization. For this purpose
run-on experiments were carried out in the presence and in the absence of [alpha]-amanitin. Run-on experiments were carried out as outlined in
Materials and Methods. The results in Figure
2
B show clearly that [alpha]-amanitin inhibited the synthesis of labeled RNA that hybridized to
the cDNA from pMG and [beta]-actin (in both lanes of + and - [alpha]-amanitin the same amount of labeled RNA was
hybridized). These results indicate that the hybridization was specific and
that the labeled RNA hybridizing to the pMG and [beta]-actin cDNA was transcribed by RNA polymerase II.
A decrease in the level of immunoprecipitable DNA methyltransferase and of its
mRNA in the presence of a constant rate of transcription could possibly reflect
an increase in the turnover of the enzyme and its mRNA. These questions were
addressed for the mRNA by Northern blots combined with the quantitation with a
PhosphoImager and for the protein by pulse chase experiments with labeled leucine combined with immunoprecipitation. Experiments were carried out as outlined in Materials and Methods. Figure
4
(upper panel) shows that upon the onset of myoblast differentiation there was
an ~3-fold increase in the rate of DNA methyltransferase mRNA degradation. In growing and in differentiating cells the apparent half life of the mRNA was 5 and 1.5 h respectively. In the same RNA preparation for growing and
differentiating cells the relative rate of degradation of the histone H4 mRNA
remained constant (Fig.
4
, lower panel).
As measured by pulse chase experiments with tritiated leucine (Fig.
5
) the turnover of the DNA methyltransferase protein increased by ~4-fold during differentiation. The apparent half life of DNA
methyltransferase in growing and in differentiating cells was 18 and 4.5 h
respectively.
Besides the cyclic variation of the DNA methyltransferase activity in nuclei
during the cell cyle (
6
-
10
) there is also a down regulation of the enzyme activity during the
differentiation of erythroleukemia cells (
10
), myoblast (
5
) and teratocarcinoma cells (
17
). A down regulation of the DNA methyltransferase activity during
differentiation could be responsible in part for the genome wide demethylation
occuring in these cells. In the particular case of erythroleukemia cells and
myoblasts where there is a genome wide demethylation in the absence of DNA
replication, the down regulation of DNA methyltransferase could possibly play
an indirect role in the demethylation of DNA. For example, during the last
round of replication it could facilitate the formation of hemimethylated DNA
which was found to be the favoured substrate for 5-methylcytosine DNA glycosylase (
16
). A change in the ratio of DNA methyltransferase to 5-methylcytosine DNA glycosylase would also favor DNA demethylation (
5
). In teratocarcinoma cells, the genome wide demethylation occurs gradually
after several cycles of replication in the presence of a reduced activity of DNA methyltransferase. In this case demethylation of DNA is probably occurring passively during replication. As we have shown in
Figure
2
, the down regulation of DNA methyltransferase is not due to a decrease in the
rate of the synthesis of its mRNA. A similar observation was also made for the
differentiating teratocarcinomas cells (
17
) and in the growth arrested Balb/c 3T3 cells (
6
). The nature of the signals responsible for the increase in the turnover of DNA
methyltransferase and of its mRNA is at present unknown. The signals triggering
the increase in the turnover of DNA methyltransferase mRNA could possibly be
related to those responsible for the posttranscriptional regulation of enzymes
that are expressed preferentially in the S phase of the cell cycle such as
thymidine kinase (
18
), thymidylate synthetase (
19
) and histone H3 (
20
).
We would like to thank Mrs Y. C. Jost for helping us with the tissue culture and
the typing of this manuscript. We would also like to thank Drs Y. Nagamine, S.
Arber and A. Wiederkehr for their gift of [beta]-actin, myogenin and histone H4 DNA. We also thank Prof. T. Bestor
for his generous gift of the DNA methyltransferase cDNA clone. We greatly
appreciate Drs J. Paskowsky, E. Oakeley and S. Schwarz for their critical
reading of this manuscript.
REFERENCES
Return