|
|
|
|||||||
|
|||||||||
(Received for publication, September 6,
1994; and in revised form, October 26, 1994) From the
Poly(ADP-ribose) polymerase (PADPRP) is biologically significant
in the rejoining of DNA strand breaks. Post confluent cultures of
3T3-L1 preadipocytes showed marked increases in PADPRP protein and
activity when the cells were induced to differentiate into adipocytes.
When this increase in PADPRP expression was prevented in stably
transfected 3T3-L1 cells by induction of PADPRP antisense RNA
synthesis, the cells did not differentiate nor undergo the two or three
rounds of DNA replication that are required for initiation of the
differentiation process. 3T3-Ll cells expressing PADPRP antisense RNA
under differentiation conditions were easily detached from plates and
in some cases eventually died. When newly expressed PADPRP protein and
DNA synthesis was assessed in cells at zero time or at 24 h after
induction of differentiation by incorporation of bromodeoxyuridine or
[ Poly(ADP-ribose) polymerase (PADPRP), ( As mentioned
above, past approaches to determining the role of PADPRP in the
differentiation process depended heavily on the use of chemical
inhibitors (for example, nicotinamide, thymidine, benzamide, and
coumadin) (Yamagoe et al., 1991) of poly(ADP-ribose)
synthesis. However, the use of these inhibitors has limitations because
of their potential effects on other biological processes (Milam and
Cleaver, 1984). To alleviate these problems, we have previously
explored the feasibility of establishing stably transfected mammalian
cells that express PADPRP antisense RNA (Ding et al., 1992).
Despite the fact that the half-life of PADPRP in cells is relatively
long (48-72 h), the induction of PADPRP antisense RNA in HeLa
cells decreases both PADPRP activity and protein in the nucleus (Ding et al., 1992). Additionally, chromatin of PADPRP-depleted
cells has an altered structure, as assayed by deoxyribonuclease I
susceptibility. Interestingly, given the proposed role for PADPRP in
the repair of DNA strand breaks, cells depleted of PADPRP cannot
initiate DNA rejoining after strand breaks, which, as mentioned above,
is a process thought to have an important role in differentiation
(Farzaneh et al., 1982). The 3T3-L1 preadipocyte cell line
represents a useful tool for studying mechanisms of cellular
differentiation (Mackall et al., 1976; Reed and Lane, 1980;
Pekala et al., 1981). When appropriately induced by a defined
hormonal treatment (insulin, dexamethasone, and
methylisobutylxanthine), 3T3-L1 preadipocytes differentiate in culture
into cells that possess morphological and biochemical characteristics
of adipocytes (Pekala et al., 1981). A marked increase in the
concentrations of lipogenic and lipolytic enzymes as well as of other
adipocyte-specific proteins accompanies acquisition of the adipocyte
phenotype (Ntambi et al., 1988). Depending upon the assay
used, both decreases as well as increases in PADPRP activity have been
reported as one of the earliest events during differentiation in these
cells (Pekala and Moss, 1983; Janssen and Hilz, 1989). Accordingly,
with the antisense expression conditions for HeLa cells as a guide, we
have analyzed how altering the concentration of PADPRP, by antisense
RNA expression, may affect the onset and progression of cellular
morphological changes in 3T3-L1 preadipocytes during differentiation.
Additionally, a putative role for PADPRP in apoptosis in this system is
discussed.
Figure 1:
Structure and restriction sites of the
pMAM-As plasmids containing murine PADPRP cDNA in antisense orientation
downstream of the MMTV LTR. The pMAM-As expression vector contained the
glucocorticoid-inducible MMTV promoter ligated to a 1.1-kb murine
PADPRP 5` cDNA fragment, in reverse orientation, that encoded the
DNA-binding domain and a portion of the automodification domain of the
enzyme. A transcription start site within the MMTV LTR was located 260
base pairs upstream of the cloning site, and the SV40 early splicing
and polyadenylation regions were located 1.0 kb downstream of the
PADPRP sequence. The expected transcript size was
Southern
hybridization of genomic DNA and Northern analysis of PADPRP mRNA were
performed with
Total
RNA from 1
Figure 2:
Dexamethasone-induced synthesis of
PADPRP antisense RNA in 3T3-L1 cells stably transfected with plasmid
pMAM-As. A, control (nontransfected) 3T3-L1 cells and 3T3-L1
cells stably transfected with pMAM-As (lines As 5 and As
6) were incubated with dexamethasone (1 µM) for the
indicated time periods. Total RNA was isolated, and 10 µg were
subjected to Northern analysis as described under ``Materials and
Methods.'' B, Northern blots were hybridized with a
The
influence of dexamethasone-induced PADPRP antisense RNA on the
concentration of endogenous PADPRP mRNA in cell lines As 5 and As 6 was
determined by Northern analysis with a murine PADPRP cDNA probe (data
not shown). In control 3T3-L1 cells, treatment with dexamethasone for
up to 72 h had no effect on the steady-state concentration of PADPRP
mRNA. In contrast, treatment of As 5 and As 6 with dexamethasone at
various time points reduced the concentration of PADPRP mRNA by
80-95%.
Figure 3:
Effect of PADPRP antisense RNA synthesis
on cellular PADPRP concentration. Control (nontransfected) 3T3-L1 cells (A) and As 5 and As 6 cells (B) were incubated with
dexamethasone (1 µM). At the indicated times, cells were
washed with PBS, and their protein concentration was determined. Equal
amounts of total cellular protein (50 µg) were subjected to
electrophoresis on an SDS-polyacrylamide gel, and the separated
proteins were then transferred to a nitrocellulose filter. The filter
was incubated with rabbit antiserum to porcine thymus PADPRP at a
dilution of 1:2000. No Dex, cells incubated in the absence of
dexamethasone. The positions of molecular mass standards (STD)
are indicated.
In control cells, incubation with
dexamethasone for 72 h did not affect PADPRP concentration (Fig. 3A). Furthermore, the amounts of PADPRP in
uninduced As 5 and As 6 cells were approximately the same as that in
control cells. In contrast, the concentration of PADPRP was markedly
reduced in As 5 and As 6 cells after induction of PADPRP antisense RNA
synthesis. The PADPRP concentration in these cells decreased by
Figure 4:
PADPRP activity and concentration in
control 3T3-L1 preadipocytes during differentiation. Two days after
achieving confluency, 3T3-L1 cells were induced to differentiate with
dexamethasone, methylisobutylxanthine, and insulin as described under
``Materials and Methods.'' At the indicated times, cells were
assayed for PADPRP activity by the sonication method, as described
under ``Materials and Methods,'' with 10-20 µg of
protein for each determination. Values are the means of duplicate
determinations. Inset, immunoblot analysis of PADPRP
concentration. Equal amounts (50 µg) of total cellular protein were
subjected to immunoblot analysis as described in Fig. 3. The arrow indicates the 113-kDa band corresponding to the
full-length murine PADPRP.
The
catalytic activity of PADPRP is influenced by the accessibility of
acceptor molecules, and is also linearly dependent on the number of DNA
strand breaks and effected by several other properties of chromatin
(Butt et al., 1978, 1979; Benjamin and Gill, 1980; Berger et al., 1987; Jacobson and Jacobson, 1989). Because of these
potential complications, we also examined the effect of differentiation
on PADPRP by immunoblot analysis, which had not previously been
examined. A marked increase in immunoreactive PADPRP was observed
during the first 2 days of differentiation (Fig. 4, inset). Variability in the number of immunoreactive bands
observed may be due to the fact that PADPRP has been reported to
degrade in cells into peptides of discrete sizes (Hotlund et
al., 1983; Surowy and Berger, 1985), which are also detectable
with the porcine anti-PADPRP antibody. PADPRP activity decreased to
below control values by 4 days, as the cells terminally differentiated
into adipocytes. At the fully differentiated state (day 8),
immunoreactive PADPRP was virtually undetectable. These results are
consistent with previous studies showing low PADPRP activity in
terminally differentiated cells (Cherney et al., 1985).
The effect of PADPRP antisense RNA was subsequently examined
during differentiation (Fig. 5). The marked increases in PADPRP
activity and protein observed during the first 2 days of
differentiation of control 3T3-L1 cells (Fig. 4) were not
apparent in antisense cells; PADPRP activity and protein remained
constant (Fig. 5). A slightly higher immunological reaction of
the truncated PADPRP was noted at day 2, although this was not
associated with increased PADPRP at this time point. Furthermore, in
contrast to control 3T3-L1 cells (both nontransfected and transfected
with pMAM-neo only), As 5 and As 6 cells did not accumulate cytoplasmic
triglyceride as visualized by phase-contrast microscopy and oil red O
staining (Fig. 6). In both control cell lines, triglycerides
were observed from day 3 of differentiation and were visible in almost
90% of the cells by day 7. As 5 and As 6 cells also did not exhibit any
of the morphological changes observed for control cells after induction
of differentiation.
Figure 5:
PADPRP activity and concentration in
3T3-L1 As 5 cells after incubation with inducers of differentiation.
Two days after achieving confluency, 3T3-L1 cells were exposed to
inducers of differentiation. At the indicated times, cells were assayed
for PADPRP activity by the sonication method. Values are the means of
duplicate determinations. Inset, equal amounts (20 µg) of
total cellular protein were subjected to immunoblot analysis of PADPRP
as described in Fig. 4. The arrow indicates the 113-kDa
band corresponding to the full-length murine
PADPRP.
Figure 6:
Cytoplasmic triglyceride accumulation in
control and pMAM-As-transfected 3T3-L1 cells. Seven days after exposure
to inducers of differentiation, cells were stained with Oil Red O dye
and examined by phase-contrast microscopy. A control
(nontransfected) 3T3-L1 cells. B, mock-transfected 3T3-L1
cells (cells with transfected with pMAM-neo vector devoid of
PADPRP cDNA. C, As 5 cells; D, As 6
cells.
These antisense cells did not undergo the normal
two or three rounds of cell division during the initial 2 days after
exposure to inducers of differentiation (Fig. 7). Furthermore,
the cells became easily detached from plates after 5 days of exposure
to differentiation inducers and eventually died, as revealed by their
failure to exclude trypan blue. In some instances, the antisense cells
progressed through one round of cell division (Fig. 8, inset); however, this process resulted in even more extensive
cytotoxicity. These results are consistent with a potential role of
PADPRP in apoptosis (Nosseri et al., 1994; Kaufmann et
al., 1994) and antisense 3T3-L1 cells might prove useful to
further study this important biological process. Although the role of
PADPRP in cells that undergo apoptosis is still unclear, it has been
observed that inhibitors of PADPRP may either enhance or reduce the
extent of apoptosis (Rice et al., 1992, Nosseri et
al., 1994). Confluent antisense cells, treated with dexamethasone
alone, did not undergo cell death, even after 2 weeks of exposure to
this agent. Thus, the cytotoxicity observed in the antisense cell
lines, exposed to differentiation inducers, was probably not
attributable to antisense expression per se. The expression of
several other antisense transcripts in 3T3-L1 cells, such as those to
actin or ferritin heavy chain, has been shown not to affect
differentiation or to have adverse effects on growth (Wenz et
al., 1992).
Figure 7:
Proliferation of control and
pMAM-As-transfected 3T3-L1 cells after incubation with inducers of
differentiation. Two days after achieving confluency, control
(nontransfected) 3T3-L1 cells and antisense cells were incubated with
inducers of differentiation as described under ``Materials and
Methods.'' Cells were collected at the indicated times and counted
with a hemocytometer and trypan blue dye and with a Coulter
counter.
Figure 8:
Time course of DNA replication in 3T3-L1
control and antisense cells during differentiation, as measured by the
rate of [
Figure 9:
Determination of BrdU incorporation and
expression of PADPRP during the first 24 h of differentiation in intact
cells by immunocytochemical methods. Two days after achieving
confluency, control (non-transfected) 3T3-L1 cells (A) and
antisense cells (B) were incubated with inducers of
differentiation for 24 h. Selected nuclei from A were enlarged (C). Cells were collected at zero time (prior to adding
differentiation inducers) and 24 h after induction of differentiation
as described under ``Materials and Methods.'' For BrdU
incorporation, the cells were incubated with 10 µM BrdU
for 1 h and immunodetected by a mouse monoclonal anti-BrdU, followed by
rhodamine-labeled anti-mouse IgG. To detect PADPRP immunologically,
fixed cells were incubated with rabbit anti-PADPRP, followed by
fluoresceinlabeled anti-rabbit IgG, as described under ``Materials
and Methods.''
Selected nuclei from Fig. 9A were enlarged (Fig. 9C). Both
BrdU and PADPRP appear to be concentrated in distinct intranuclear
granular foci, although it is not clear if they represent similar
substructures of the nuclei. As evident in Fig. 9B,
there was negligible BrdU and PADPRP immunocytochemical staining in
nuclei of antisense 3T3-L1 cells after 24 h of induction of
differentiation.
In the experiment in Fig. 10, both control and antisense
cells were examined for interaction of PADPRP and DNA pol
Figure 10:
PADPRP and DNA polymerase
With control cells at zero time (Fig. 10, lane 1),
PADPRP was not observed to co-immunoprecipite with DNA pol Although chemical inhibitors have proven useful in
investigating potential functions of PADPRP (Caplan et al.,
1979; Morioka et al., 1979; Brac and Ebisuzaki, 1985) they are
not without potential complications (Milam and Cleaver, 1984). Thus, we
have applied a more specific approach, that of antisense RNA synthesis,
to determine the effect of inhibition of poly(ADP-ribosyl)ation on the
differentiation of 3T3-L1 preadipocytes. Both PADPRP protein and
activity show a transient, marked increase on exposure of cells to
inducers of differentiation (Fig. 5). The increase in PADPRP
expression was prevented in this study by antisense RNA synthesis.
Under these conditions, both cellular proliferation and differentiation
did not occur ( Fig. 6and Fig. 7). The synthesis of
antisense RNA corresponding to Various models, all of which involve chromatin restructuring, have
been proposed to explain the necessity for DNA replication prior to
differentiation (Villarreal, 1991). Implicit in many of these models is
the repositioning or alteration of nucleosomes, which may result in the
activation or inhibition of specific genes and may involve various
nuclear protein modifications, including poly(ADP-ribosyl)ation. In
this regard, poly(ADP-ribosyl)ation may aid in either relaxation (de
Murcia et al., 1986) or condensation of chromosomal proteins
around the replicating regions of chromatin (Butt et al.,
1980). A role for poly(ADP-ribosyl)ation in the regulation of
chromatin structure is supported by various previous observations. In
HeLa cells synthesizing PADPRP antisense RNA, we observed a
hypersensitivity of chromatin to deoxyribonuclease 1 (Ding et
al., 1992). Chromatin regions corresponding to replicating regions
of DNA shows transient hypersensitivity to nuclease digestion (Smerdon,
1989), which may be related to localized changes in nuclear
organization of the 300-Å fiber of chromatin. Additional
experimental systems have indicated that poly(ADP-ribosyl)ation of
nucleosomal proteins may influence the structure of chromatin and
thereby facilitate DNA replication and other nuclear processes (Jump et al., 1980; Althaus et al., 1985; Chatterjee et
al., 1989; Jacobson and Jacobson, 1989). Recently, Mathis and
Althaus(1986) showed a significant influence of poly(ADP-ribosyl)ation
and chromatin structure on DNA repair: in the presence of chemical
inhibitors of PADPRP, DNA adducts were inaccessible for repair and
tended to accumulate in nonnucleosomal regions of chromatin. The
initial stages in the rejoining of DNA strand breaks were shown to be
inhibited in HeLa cells depleted of PADPRP as a result of PADPRP
antisense RNA synthesis (Ding et al., 1992). Thus, the low
viability of 3T3-L1 cells induced both to differentiate and to
synthesize PADPRP antisense RNA may result from a toxic accumulation of
Okazaki fragments generated during the early stages of differentiation.
In this regard, Lonn and Lonn(1988) showed that the accumulation of
unligated Okazaki fragments is increased significantly in human cells
incubated with the PADPRP inhibitor, 3-aminobenzamide. PADPRP
contains two zinc fingers, which are required for the binding of the
enzyme to both single-strand and double-strand DNA breaks in a
sequence-independent manner (de Murcia et al., 1983; Cherney et al., 1987). Furthermore, the catalytic activity of PADPRP
is dependent on DNA strand breaks. The replication forks associated
with differentiation-related DNA synthesis contain numerous free ends
that could activate the poly(ADP-ribosyl)ation of nucleosomal proteins,
which, in turn, might alter the structure of these regions. In this
regard, Satoh and Lindahl(1992) have proposed a model, on the basis of in vitro data, in which non-poly(ADP-ribosyl)ated PADPRP binds
to ends of DNA during DNA replication or repair. Subsequently, the
large negative charge resulting from extensive automodification of
bound PADPRP would promote removal of the enzyme from the DNA ends to
allow DNA ligation. This proposed mechanism is in agreement with our
previous data showing that the addition of NAD (the substrate of
PADPRP) to chromatin in vitro increases the accessibility of
either endogenous or exogenously added DNA polymerase to DNA primers
(Roberts et al., 1974). In further support of the putative
association of poly(ADP-ribosyl)ation with DNA synthesis is the
observation that the replicating regions of chromatin are selectively
retained relative to nonreplicating regions, when either polyoma or
SV40 minichromosomes are subjected to anti-poly(ADP-ribose) antibody
affinity chromatography (Baksi et al., 1987). We recently
tested the above model with bacterially expressed mutants of PADPRP
with deletions in the three major functional domains of the enzyme
(Smulson et al., 1994). Deletion mutants with an intact
amino-terminal DNA binding domain inhibit DNA repair, whereas a mutant
with a deletion in the DNA-binding domain does not inhibit the in
vitro assay. However, whether the deletion is in the NAD-binding,
active site domain, or the automodification domain, the inhibition of
repair exerted by these mutant proteins is not alleviated by NAD. In
the current work, we have provided data showing the
co-immunoprecipitation of PADPRP and DNA pol Reddy and Pardee(1980)
provided interesting data earlier and Reddy and Fager(1993), more
recently, which identify a nuclear complex of ``DNA
precursor-synthesizing enzymes juxtaposed with the replication
apparatus comprising DNA polymerase, other enzymes and structural
proteins.'' By immunofluorescent imaging analysis, Li et
al.(1993) further showed that DNA pol
Volume 270,
Number 1,
Issue of January 6, 1995 pp. 119-127
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
H]thymidine into DNA, significant incorporation
was shown to occur in control cells after 24 h, but not in antisense
cells. Furthermore, during the first 24 h, the co-immunoprecipitation
of PADPRP and DNA polymerase 
was observed in control cells,
whereas no such complex formation was noted in the induced antisense
cells, nor in uninduced control cells.
)a
chromatinassociated enzyme, has an absolute requirement for DNA for
activity and is proportionally activated by the presence of DNA strand
breaks (Smulson et al., 1977; Juarez-Salinas et al.,
1979; Benjamin and Gill, 1980; Berger et al., 1980). With
nuclear NAD as a substrate, PADPRP catalyzes the poly(ADP-ribosyl)ation
of a subclass of nuclear proteins, including histones, high mobility
group chromosomal proteins, topoisomerases (Kasid et al.,
1989), protein ICP4 of herpesvirus (Blaho et al., 1992), and
simian virus 40 (SV40) large T antigen (Baksi et al., 1987).
PADPRP and the nuclear protein modification it catalyzes are thought to
participate, together with other enzymes, in DNA repair and replication
as well as in other cellular processes in which cleavage and rejoining
of DNA segments may be required. These processes include cellular
differentiation and transformation, sister chromatid exchange, and gene
rearrangements and transpositions (Farzaneh et al., 1982;
Ferro and Olivera, 1984; Ueda and Hayaishi, 1985). Support for a role
of poly(ADP-ribosyl)ation in the differentiation process has been
provided by a number of studies, some of which examined the effects of
chemical inhibitors of PADPRP. Accordingly, DNA strand breakage, which
activates PADPRP as well as the enzyme NADase, has been implicated in
the terminal differentiation process of HL-60 cells (Farzaneh et
al., 1987). In this regard, in HL-60 cells induced to
differentiate with retinoic acid, the concentration of PADPRP mRNA
increased from an initially low value and remained elevated for
12-24 h, after which the concentration declined steadily (Bhatia et al., 1990b). A similar pattern was observed during dimethyl
sulfoxide-induced differentiation of HL-60 cells, with the exception
that the PADPRP mRNA concentration increased later in the
differentiation process (Bhatia et al., 1990b). Suzuki et
al.(1989) observed a reduction in PADPRP mRNA concentration after
exposure of HL-60 cells to retinoic acid for 3 days, after which time
45% of the cells showed the differentiated granulocytic phenotype.
Earlier studies suggested that an increase in PADPRP activity precedes
the differentiation process (Caplan and Rosenberg, 1975). In contrast,
decreased PADPRP activity has been correlated with the appearance of
specific gene products (Rastle and Swetly, 1978).
Vector and Probes
A 1.1-kb XhoI
fragment of murine PADPRP cDNA, including 150 base pairs of
5`-untranslated sequence and the region encoding the DNA-binding domain
and the amino-terminal portion of the automodification domain of
PADPRP, was subcloned in the antisense orientation into the expression
vector pMAM-neo (Clontech) under control of the mouse mammary
tumor virus (MMTV) long terminal repeat (LTR). The antisense
orientation of the insert with respect to the promoter was confirmed by
restriction enzyme mapping of the plasmid, which was designated
pMAM-As, with BssHII, StuI, and BssHII
together with SacI (see Fig. 1).
2.4 kb. The
entire expression plasmid comprised 9.4 kb.
P-labeled murine PADPRP cDNA that had been
prepared by random primer extension (>10
cpm/µg of
DNA). Antisense PADPRP RNA was detected by Northern analysis with a
single-stranded RNA probe that was generated from the plasmid, pGEM-3Z
(Promega). The plasmid was manipulated to contain the 1.1-kb EcoRI 5` fragment of murine PADPRP cDNA downstream of T7
promoter. The plasmid was linearized, and a P-labeled RNA
probe complementary to PADPRP antisense RNA was synthesized with T7 RNA
polymerase.
Cell Culture and Transfection
3T3-L1 preadipocytes
(Green and Kehinde, 1974, 1975) were grown in Eagle's medium
supplemented with 10% fetal bovine serum, penicillin (100 units/ml),
and streptomycin (100 mg/ml streptomycin) (all from Life Technologies,
Inc.). For transfection, 5 
10
cells in 100-mm
dishes were incubated overnight with 20 µg of calcium
phosphate-precipitated pMAM-As DNA (Stratagene). Transfectants were
selected after 48 h by culturing in medium containing G418 (0.4 mg/ml)
(Life Technologies, Inc.) until colonies appeared. Fifty clones were
selected and grown under continued G418 selection for further study.DNA and RNA Analysis
Genomic DNA from 5
10
to 6 10 cells of the selected clones
was extracted as described earlier (Ding et al., 1992). Twenty
micrograms of DNA were digested for Southern analysis with BamHI and SacI. As expected, DNA fragments of 1.0 kb
(corresponding to the MMTV LTR and murine PADPRP cDNA immediately
downstream of the promoter) and 1.5 kb (corresponding to the 3` of the
cDNA and a portion of the vector portion) hybridized to the mouse cDNA
probe. Digestion with XhoI generated a hybridizing DNA
fragment of 1.1 kb, corresponding to the entire cDNA insert. 10 cells from each antisense cell line
was isolated by guanidinium-phenol extraction (Chomczynski and Sacchi,
1987) and was analyzed by Northern analysis with the P-labeled RNA probes complementary to PADPRP antisense RNA
or with
P-labeled murine PADPRP cDNA.
Immunoblot Analysis
SDS-polyacrylamide gel
electrophoresis and transfer of separated proteins to nitrocellulose
were performed by standard procedures. Protein immobilized on
nitrocellulose was stained with 0.1% Ponceau S to confirm equal
transfer. Polyclonal rabbit antibodies to porcine thymus PADPRP, which
have been shown to bind to murine PADPRP (Ludwig et al.,
1988), were provided by Dr. Helmuth Hilz (Hamburg University). The
immune complexes formed by these antibodies and PADPRP were detected
with alkaline phosphataseconjugated antibodies to rabbit immunoglobulin
G.Growth Curve
Cells (5 10) were
seeded on 100-mm dishes and, 1 day later, incubated in the presence or
absence of dexamethasone (1 µM). Samples were collected at
various times, and the cell number and viability determined by cell
counting with a Coulter counter and with a hemocytometer in the
presence of trypan blue.Induction of Antisense RNA Synthesis
Culture
dishes (75 cm
) were inoculated with 1 10 cells. The following day, cells were incubated in DMEM
supplemented with 10% FBS and dexamethasone (1 µM). Cells
were collected for analysis at various times.Induction of Differentiation of 3T3-L1
Preadipocytes
Culture dishes (75 cm) were inoculated
with 1 10 control 3T3-L1 cells or stably
transfected 3T3-L1 cells, and the cells were grown to confluence in
DMEM with 10% FBS. After reaching confluency, the cells were maintained
for an additional 2 days. To initiate differentiation, we supplemented
the DMEM with methylisobutylxanthine (0.5 mM), dexamethasone
(1 µM), insulin (1.7 µM), and 10% FBS. After
48 h, the medium was replaced with DMEM supplemented with insulin (1.7
µM) and 10% FBS. After a further 48 h, the medium was
replaced with DMEM supplemented with only 10% FBS.Oil Red O Staining for Triglyceride Droplets
Cells
were washed twice with phosphate-buffered saline (PBS) and then fixed
with 3.7% formaldehyde in PBS for 10 min. The cells were washed twice
with deionized water, incubated with Oil Red O dye (0.3%) for 1 h at
room temperature, rinsed with water, and then observed by
phase-contrast light microscopy.PADPRP Activity Assay
Cells (10
)
were harvested by scraping, washed three times with ice-cold PBS, and
sonicated three times with 10-s bursts in order to break cells and
introduce an excess of DNA strand breaks (required for PADPRP
activity). The initial velocity of [P]NAD
incorporation into acid-insoluble acceptors was measured at 25 °C
for 1 min, according to the method described by (Cherney et
al., 1985).
Immunocytochemical Methods
Cells at zero time or
24 h after induction of differentiation were pulsed with 10 µM bromodeoxyuridine (BrdU) for 1 h, washed twice with ice-cold PBS,
fixed with cold 95% ethanol for 30 min at room temperature, and
subsequently denatured with 0.07 M NaOH in 70% ethanol for 2
min. After washing twice with PBS, the cells were incubated for 4 h in
a humid chamber with both mouse monoclonal anti-BrdU (Becton Dickinson)
and anti-PADPRP, diluted 1:10 and 1:1000, respectively, with PBS
containing 12% bovine serum albumin, followed by a final incubation in
the dark with a mixture of rhodamine-labeled anti-mouse (Tago, 1:10)
and goat fluorescein-labeled anti-rabbit (1:40). Between antibody
incubations, the cells were washed twice with PBS and once with water.
The plates were then cut and mounted with Vectashield, and observed
under a Zeiss immunofluorescence microscope.DNA Replication Assay by
[
Cells at
indicated time intervals after induction of differentiation were
harvested by scraping, aliquoted into cell wells (10H]Thymidine Incorporation cells
in 2 ml of medium/well), allowed to stabilize in a 37 °C CO incubator for 1 h, and then pulsed for 15 min with
[H]thymidine (TdR, 40 Ci/mmol, 0.2 µCi/ml).
The cells were then collected by centrifugation, washed extensively and
resuspended in PBS, and lysed with 0.1% SDS and 1 mM EDTA.
Measurement of acid-insoluble radioactivity was performed by
trichloroacetic acid precipitation on GF/C filters (Whatman), followed
by washes with 20% trichloroacetic acid, 70% ethanol, and 95% ethanol
and liquid scintillation counting of the filters.Immunoprecipitation and
Immunoblotting
Immunoprecipitation was performed with slight
modifications as described previously for these two enzymes (Simbulan, et al., 1993). Cells were washed twice with ice-cold PBS and
lysed with 1 ml of EBC buffer (50 mM Tris-HCl, pH 8.0, 120
mM NaCl, 0.5% Nonidet P-40, and 0.1 trypsin inhibitory unit of
aprotinin) for 20 min on ice. The lysates were then clarified by
centrifugation, aliquoted to 200 µl/sample (50 µg protein), and
pre-cleared with 10 µl/sample of protein A-Sepharose overnight at 4
°C. After centrifugation, the supernatants were rocked for 1 h with
0.5 ml NET-N buffer (20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, and 0.5% Nonidet P-40) containing a
mouse monoclonal anti-DNA polymerase antibody (ATCC cell line
SJK-132-20, 10 µl of ascitis/sample), and rocked for another 20 min
with 20 µl of a 1:1 mixture of protein A-Sepharose, freshly washed
and suspended in Tris-buffered saline with 10% bovine serum albumin.
The beads were washed five times with 1 ml of NET-N buffer, and the
immunoprecipitates on the beads were subsequently separated by
SDS-polyacrylamide gel electrophoresis and transferred to
nitrocellulose. After staining with Ponceau S stain to confirm
transfer, the blots were blocked, incubated with polyclonal rabbit
anti-PADPRP (1:1000), probed with peroxidase-labeled anti-rabbit IgG,
and detected by electrochemiluminescence.
PADPRP Antisense Vector and Cell
Transfection
With a human PADPRP cDNA probe, murine genomic and
cDNA partial clones encoding PADPRP were isolated previously (Huppi et al., 1989). A 1.1-kb fragment of murine cDNA, encoding the
DNA-binding domain and the amino-terminal portion of the
automodification region of PADPRP, was subcloned in the antisense
orientation into the expression vector pMAM-neo under control
of the MMTV promoter (Fig. 1). The antisense orientation of the
PADPRP cDNA in the resulting plasmid, pMAM-As, was verified by
restriction enzyme mapping. The pMAM-As construct was stably
transfected into 3T3-L1 preadipocytes by calcium phosphate
precipitation. After selection in G418 for 4 weeks, 50 resistant
colonies were isolated and screened for decreased PADPRP activity after
incubation with dexamethasone, as described previously for HeLa cells
(Ding et al., 1992). DNA was prepared from seven of the cell
lines and analyzed for integration of the antisense PADPRP sequence.
The 1.1-kb antisense cDNA fragment was detected after digestion of the
DNA with XhoI in all seven cell lines (not shown). The
hybridization intensity corresponding to the 1.1-kb fragment was
significantly greater than that observed for the endogenous murine
PADPRP gene, suggesting that pMAM-As was integrated in high copy
numbers.
Expression of PADPRP Antisense RNA in 3T3-L1 Cells and
Effects on Endogenous Transcripts
To characterize and quantify
the kinetics of antisense PADPRP RNA expression, we performed Northern
analysis with control 3T3-L1 cells and two transfected 3T3-L1 cell
lines (As 5 and As 6) after incubation with dexamethasone for various
times (Fig. 2). Blots were probed with a riboprobe that
hybridizes specifically to murine PADPRP antisense RNA (Fig. 2B). An antisense PADPRP RNA of 2.4-kb was
detected in clones As 5 and As 6 within 5 h after induction by
dexamethasone; thereafter, the amount of the antisense RNA decreased,
but was still detectable after 72 h (Fig. 2A). The
antisense RNA included a portion of the MMTV promoter region upstream
of the PADPRP cDNA insert, as well as SV40 sequences downstream of the
PADPRP cDNA (Fig. 1). Antisense RNA was not detected in the As 5
and As 6 cell lines in the absence of dexamethasone (Fig. 2A) indicating that the MMTV promoter was under
tight control in these cells. In addition, expression of a hybridizing
antisense transcript was not observed in control cells in the absence
or presence of dexamethasone (Fig. 2A). The decrease in
the amount of PADPRP antisense RNA in cell lines As 5 and As 6 between
5 and 72 h after dexamethasone treatment is consistent with several
observations showing that antisense RNA is unstable and that hybrid
sense-antisense duplex RNAs have short half-lives (Izant, 1984).
P-labeled single-stranded RNA probe that was synthesized
from a pGEM-3Z vector containing the 1.1-kb mouse PADPRP cDNA fragment
used to construct pMAM-As. The riboprobe was synthesized in the sense
orientation with T7 RNA polymerase.
Effect of Induction of PADPRP Antisense RNA on 3T3-L1
PADPRP Concentration
Total cellular protein was extracted from
antisense cell lines As 5 and As 6 as well as from control cells that
had been incubated with dexamethasone for various periods of time.
Proteins were analyzed by immunoblotting with polyclonal antibodies to
porcine PADPRP (Fig. 3); these antibodies were previously shown
to react with a 113-kDa protein, corresponding to full-length PADPRP,
as well as with 89- and 90-kDa proteins, which correspond to truncated
PADPRP molecules observed in murine cell extracts, not present in other
species examined (Ludwig et al., 1988). Because of the
additional PADPRP-derived degradation products, some variation in the
number of immunostained bands were occasionally encountered in contrast
to earlier studies with human cells (Ding et al., 1992). A
Ponceau stain of the immunoblot indicated that equivalent amounts of
protein were loaded onto the gel in each lane and that there was no
overall cellular protein degradation during PADPRP antisense RNA
induction (data not shown).
40, 60, and 80% after 48, 72, and 96 h of exposure to
dexamethasone, respectively (Fig. 3B). The 113-kDa
full-length PADPRP protein and the truncated 89-kDa protein decreased
proportionally with induction of antisense RNA synthesis. To verify
that the decrease in PADPRP concentration was not attributable to
decreased growth rate, we showed that the amounts of PADPRP in cells
incubated for 0 and 72 (As 6) or 96 (As 5) h in the absence of
dexamethasone were virtually identical (Fig. 3B).
Effect of Antisense RNA Synthesis on PADPRP
Activity
PADPRP activity was measured directly in sonicated
extracts of control and antisense cells at various times after
induction by dexamethasone. The sonication procedure had previously
been established to maximize the number of endogenous DNA strand
breaks, which are required for PADPRP activity. In the PADPRP assay,
activity is proportional to the number of PADPRP molecules because
automodified PADPRP is the major product of the reaction (Cherney et al., 1985). PADPRP activity decreased by 50-60% in As
5, As 6, and As 7 cells after 48 h of dexamethasone treatment (Table 1). A maximal decrease in activity of 75% was observed
after 72 h of exposure to dexamethasone. The extent of inhibition of
PADPRP activity was essentially consistent with both the corresponding
decreases in the concentrations of PADPRP protein (Fig. 3) and
PADPRP mRNA (data not shown).
Effect of Differentiation of 3T3-L1 Preadipocytes on the
Concentration of PADPRP
Several studies have shown that PADPRP
activity changes during the differentiation of 3T3-L1 preadipocytes
into adipocyte-like cells. However, the observed patterns of changes
have differed, depending on the nature of the polymerase assay used
(Pekala et al., 1981; Lewis et al., 1982; Pekala and
Moss, 1983; Janssen and Hilz, 1989). Because of these differences, we
have reexamined the effect of 3T3-L1 cell differentiation on PADPRP by
both activity measurement as well as immunoblot analysis. Control
3T3-L1 cells were allowed to grow for 2 days after reaching confluency,
and differentiation was then initiated by exposure to insulin,
dexamethasone, and methylisobutylxanthine. Cells were collected daily
and examined for both morphological and biochemical characteristics of
adipocytes. The cells progressed through two additional rounds of cell
division, which is a prerequisite for differentiation (Villarreal,
1991). After 2 days, differentiating cells adopted an overall spherical
morphology and after 3 days synthesized triglyceride droplets (see
below). A transient 2-fold increase in PADPRP activity was apparent
during the first 2 days of differentiation (Fig. 4).
Effect of Reduced PADPRP Activity on Preadipocyte Growth,
Morphology, and Differentiation
To investigate whether decreased
PADPRP activity as a result of PADPRP antisense RNA induction would
trigger 3T3-L1 differentiation, we exposed control cells and antisense
cell lines As 5 and As 6, 2 days after achieving confluency, to
dexamethasone. Dexamethasone had no effect on control 3T3-L1 cell
morphology and the cells remained quiescent. Furthermore, dexamethasone
treatment alone did not induce As 5 and As 6 cells to differentiate.
However, in the presence of dexamethasone, As 5 and As 6 cells appeared
spindle-shaped, and the boundaries between cells were clearly
discernible; in contrast, the membrane processes of control cells, as
well as of As 5 and As 6 cells in the absence of dexamethasone,
overlapped, and the boundaries between cells were not visible (data not
shown).
H]-TdR incorporation into DNA.
Post-confluent control and antisense cells were induced to
differentiate, harvested at indicated time intervals, pulse labeled for
15 min with [H]TdR (0.2 µCi/ml), and the acid
insoluble radioactivity was then measured by TCA precipitation and
liquid scintillation counting as described under ``Materials and
Methods.''
DNA Synthesis During the Initial Stages of
Differentiation
Based upon the results on cell proliferation
shown in Fig. 7, the rate of DNA replication during the early
stages of differentiation was measured in samples pulse labeled for 15
min with [H]TdR (Fig. 8). In control
cells, DNA synthesis is initiated after approximately 12 h after
induction of differentiation, peaked at 24 h, and declined to 30% or
less by 30 h. In contrast, negligible [H]TdR
incorporation was noted in antisense cells during the same time frame.Nuclear Co-localization of Induced PADPRP and DNA
Synthesis by Immunocytochemical Staining
To further examine the
expression of PADPRP and the onset of DNA replication (Fig. 8),
we compared the cellular localization of PADPRP and incorporation of
BrdU into newly replicated DNA after 1 day of differentiation by
immunocytochemical staining (Fig. 9). In control cells, prior to
differentiation (zero time), there was only limited incorporation of
both BrdU and PADPRP protein expression observed (Fig. 9A); however, after 24 h of induction of
differentiation, there was a significant incorporation of BrdU in the
nuclei of control cells, consistent with the cell division data
presented in Fig. 7and the [H]TdR
incorporation data (Fig. 8). Concomitantly, there was a marked
expression of PADPRP protein localized to the nuclei of control cells
24 h after induction of differentiation, also consistent with Western
data presented in Fig. 4.
Association of DNA Polymerase
To initiate studies on PADPRP's role(s) in
biological events occurring during the early stages of differentiation,
we studied the interaction of PADPRP with DNA polymerase with PADPRP During
Differentiation (DNA pol
). This approach was suggested by earlier data showing that
purified PADPRP binds to and stimulates the activity of
immunoaffinity-purified calf thymus or human DNA pol by about
10-60-fold in a dose-dependent manner (Simbulan et al.,
1993). The effect is quite specific since the presence of PADPRP has no
such effect on the in vitro activities of DNA polymerases
, ,
, and
, as well as DNA primase. The stimulation
of DNA pol
activity appears to be due to direct PADPRP binding,
since a complex of PADPRP and DNA pol co-immunoprecipitates with
monoclonal DNA pol antibody (Simbulan et al., 1993). at zero
time and 24 h after induction of differentiation. Crude extracts
derived from these cells were immunoprecipitated with a monoclonal
antibody to DNA pol . The immunocomplex was then isolated by
addition of protein A-Sepharose beads and separated by SDS-gel
electrophoresis, followed by Western blotting using antibody to PADPRP.
associate
during early stages of differentiation, as determined by
co-immunoprecipitation. Both control and antisense cells were treated
essentially as described in the legend to Fig. 9. At zero time
(prior to differentiation) and 24 h after induction of differentiation,
cells were collected, washed, lysed, and aliquots of the cell extracts
(50 µg of protein) were immunoprecipitated with a monoclonal
anti-DNA pol antibody, as described under ``Materials and
Methods.'' The immunocomplex was then separated by SDS-gel
electrophoresis, transferred to nitrocellulose by Western transfer,
probed with a rabbit anti-PADPRP antibody, and detected by
electrochemiluminescence. Lanes 1, 3, 4, 7, and 8 are samples from control cells, at zero time (lane 1), or 24 h after induction of differentiation (lanes 3, 4, 7, and 8); lanes 3 and 4 are duplicate determinations. In lane 7,
control cell extracts were incubated with only protein A-Sepharose,
prior to electrophoresis, and, in lane 8, preimmune serum was
substituted for the anti-DNA pol antibody. Lanes 1, 5, and 6 represent samples from antisense cells,
either at zero time (lane 1) or at 24-h differentiation (lanes 5 and 6); lanes 5 and 6 are
duplicate determinations.
,
consistent with the low levels of PADPRP in cells prior to
differentiation (Fig. 4). However, after 24 h of induction of
differentiation (lanes 3 and 4) binding of DNA pol
and PADPRP was observed, as shown by the immunostained band for
murine PADPRP at 113 kDa. Co-immunoprecipitation of the two
enzymes did not occur if only protein A-Sepharose was present (lane
7), nor if preimmune serum replaced anti-DNA pol
(lane
8). As anticipated, we did not observe the co-immunoprecipitation
of PADPRP with DNA pol in antisense cells, either at zero time or
after 24 h of differentiation (lanes 2, 5, and 6). The data in Fig. 8Fig. 9Fig. 10further support our earlier
results using purified enzymes (Simbulan et al., 1993) and
suggest that PADPRP and DNA pol may associate at some stage
during the round of DNA replication required for differentiation.
-actin or to ferritin heavy chain
has previously been shown to have no influence on the differentiation
of these cells (Wenz et al., 1992). Alternatively, Fang-Tysr
and Lane(1992) have shown that antisense expression of an RNA
associated with adipocyte differentiation (C/EBP) prevented expression
of a number of adipocyte specific mRNAs and also prevented
differentiation. The observation that the increases in both PADPRP
protein and activity occur at the same time as cellular proliferation
at the onset of differentiation does not directly prove a correlation
between these two events. However, the increased stability of PADPRP
mRNA during DNA replication (Bhatia et al., 1990a), as well as
the high concentrations of poly(ADP-ribosyl)ated intermediates at the
middle and end of S phase (Kidwell and Mage, 1976; Wong et
al., 1983), supports the concept that the increase in PADPRP
expression during the early stages of differentiation may be critical
for the differentiation of preadipocytes because of a role of the
enzyme in a stage in the cell cycle prior or during DNA replication. in crude extracts
derived from control cells during the initial stages of
differentiation, but this was not observed in the antisense cells (Fig. 10). In this regard, we previously showed that PADPRP
binds to purified calf thymus DNA pole and stimulates its
activity by about 10-60-fold in a dose-dependent manner. It was
also observed that the stimulatory activity produced by this physical
association of PADPRP and DNA pol is lost when PADPRP is
automodified. Apparently, in the presence of PADPRP, the saturation
curve for the DNA template primer becomes sigmoidal; at very low
concentrations of DNA, PADPRP inhibits the reaction in competition with
template DNA, while at higher DNA doses the reaction is significantly
stimulated by increasing the V
of the reaction,
which suggests an allosteric effect of PADPRP on the activity of DNA
pol (Simbulan et al., 1993). and DNA ligase, both
of which exists in a 21 S multienzyme complex for DNA synthesis, can be
localized to distinct ``granular-like foci,'' whereas DNA pol
, not associated with the complex, appears to be diffusely
distributed in the nucleus. We have also previously shown that, in
crude extracts of calf thymus, a portion of PADPRP exists in a 400-kDa
as well as a large 700-kDa complex, containing DNA pol (Simbulan et al., 1993). We anticipate that the 3T3-L1 antisense cell
lines characterized in our current study may help facilitate a better
understanding of the significance of the association of PADPRP and DNA
pol and also what other biological roles poly(ADP-ribosyl)ation
may participate in during differentiation.
), polymerase
; kb, kilobase pair(s).
We thank Jane Murphy for help with the experiments,
Dr. Helmuth Hilz, University of Hamburg, for the generous gift of
antibody to murine PADPRP, and Dr. Shonen Yoshida, Nagoya University,
for the antibody to DNA pol .
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Bai, S. M. Houten, A. Huber, V. Schreiber, M. Watanabe, B. Kiss, G. de Murcia, J. Auwerx, and J. M.-d. Murcia Peroxisome Proliferator-activated Receptor (PPAR)-2 Controls Adipocyte Differentiation and Adipose Tissue Function through the Regulation of the Activity of the Retinoid X Receptor/PPAR{gamma} Heterodimer J. Biol. Chem., December 28, 2007; 282(52): 37738 - 37746. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ertsey, C. J. Chapin, J. A. Kitterman, and L. M. Scavo Ontogeny of Poly(ADP-Ribose) Polymerase-1 in Lung and Developmental Implications Am. J. Respir. Cell Mol. Biol., June 1, 2004; 30(6): 853 - 861. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Masuno, T. Kidani, K. Sekiya, K. Sakayama, T. Shiosaka, H. Yamamoto, and K. Honda Bisphenol A in combination with insulin can accelerate the conversion of 3T3-L1 fibroblasts to adipocytes J. Lipid Res., May 1, 2002; 43(5): 676 - 684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Simbulan-Rosenthal, D. S. Rosenthal, R. Luo, J.-H. Li, J. Zhang, and M. E. Smulson Inhibition of poly(ADP-ribose) polymerase activity is insufficient to induce tetraploidy Nucleic Acids Res., February 1, 2001; 29(3): 841 - 849. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. R. Sallmann, M. D. Vodenicharov, Z.-Q. Wang, and G. G. Poirier Characterization of sPARP-1. AN ALTERNATIVE PRODUCT OF PARP-1 GENE WITH POLY(ADP-RIBOSE) POLYMERASE ACTIVITY INDEPENDENT OF DNA STRAND BREAKS J. Biol. Chem., May 12, 2000; 275(20): 15504 - 15511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Nozaki, M. Masutani, M. Watanabe, T. Ochiya, F. Hasegawa, H. Nakagama, H. Suzuki, and T. Sugimura Syncytiotrophoblastic giant cells in teratocarcinoma-like tumors derived from Parp-disrupted mouse embryonic stem cells PNAS, November 9, 1999; 96(23): 13345 - 13350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Boulares, A. G. Yakovlev, V. Ivanova, B. A. Stoica, G. Wang, S. Iyer, and M. Smulson Role of Poly(ADP-ribose) Polymerase (PARP) Cleavage in Apoptosis. CASPASE 3-RESISTANT PARP MUTANT INCREASES RATES OF APOPTOSIS IN TRANSFECTED CELLS J. Biol. Chem., August 13, 1999; 274(33): 22932 - 22940. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Simbulan-Rosenthal, D. S. Rosenthal, R. Luo, and M. E. Smulson Poly(ADP-ribosyl)ation of p53 during Apoptosis in Human Osteosarcoma Cells Cancer Res., May 1, 1999; 59(9): 2190 - 2194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Orlicky, J. G. Lieber, C. L. Morin, and R. M. Evans Synthesis and accumulation of a receptor regulatory protein associated with lipid droplet accumulation in 3T3-L1 cells J. Lipid Res., June 1, 1998; 39(6): 1152 - 1161. [Abstract] [Full Text]
|
|
|
|
|
|
C. M. Simbulan-Rosenthal, D. S. Rosenthal, S. Iyer, A. H. Boulares, and M. E. Smulson Transient Poly(ADP-ribosyl)ation of Nuclear Proteins and Role of Poly(ADP-ribose) Polymerase in the Early Stages of Apoptosis J. Biol. Chem., May 29, 1998; 273(22): 13703 - 13712. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-A. Laniel, G. G. Poirier, and S. L. Guerin Nuclear Factor 1 Interferes with Sp1 Binding through a Composite Element on the Rat Poly(ADP-ribose) Polymerase Promoter to Modulate Its Activity in Vitro J. Biol. Chem., June 1, 2001; 276(23): 20766 - 20773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. M. Patel and M. D. Lane Mitotic Clonal Expansion during Preadipocyte Differentiation: Calpain-mediated Turnover of p27 J. Biol. Chem., June 2, 2000; 275(23): 17653 - 17660. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |