| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
/Signal Transducers and Activators of Transcription (STAT) Pathway by Hypermethylation at a STAT-binding Site in the p21WAF1 Promoter Region1
Department of Pathology, University of Arkansas for Medical Sciences and Arkansas Children’s Hospital, Little Rock, Arkansas 72202 [B. C., L. H., V. H. S., D. M. P.], and Department of Pathology and Laboratory Medicine, St. Jude Children’s Research Hospital, Memphis, Tennessee 38105 [J. J. J.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
. In this study, we examined CpG methylation at
the p21WAF1 promoter region in rhabdomyosarcomas
(RMSs) using Southern blot analysis with the
methylation-sensitive restriction enzyme HpaII.
Sis-inducible element (SIE)-1, a STAT-responsive element located
upstream of the p21WAF1 CpG island, was
completely methylated at an internal CpG in 13 of 26 (50%) primary RMS
tumors and 2 of 5 RMS cell lines. In contrast, all normal tissues
examined showed a partial methylation pattern at SIE-1. Complete
methylation within SIE-1 strongly correlated with decreased
p21WAF1 mRNA expression in RMS. We further
studied the effects of SIE-1 hypermethylation on
p21WAF1 induction by STAT activation. CpG
methylation within SIE-1 significantly inhibited binding of activated
STAT1 in electrophoretic mobility shift assays and abrogated
STAT-mediated transcription activation in response to IFN- in
luciferase reporter gene assays. Activation of STAT1 in response to
IFN- resulted in increased p21WAF1 expression
and growth suppression in RMS cells containing unmethylated SIE-1 but
failed to induce p21WAF1 or growth inhibition in
RD and A673 cells, both of which were completely methylated within
SIE-1. However, demethylation at SIE-1, induced by a demethylating
agent 5-aza-2'-deoxycytidine, reactivated
p21WAF1 expression and restored the
responsiveness to IFN- in RD cells. Our results indicate a mechanism
by which altered DNA methylation in the p21WAF1
promoter region, by precluding STAT1 binding to SIE-1, directly
inhibits the p21WAF1 induction and cell growth
regulation through the IFN-/STAT signaling pathway in RMS cells. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
signal transduction pathway (10, 11, 12)
. In a number
of cell types, activation of STAT1 in response to IFN- or other
extracellular factors correlates with up-regulation of
p21WAF1 expression and inhibition of cell growth
(9
, 11
, 12)
. However, the role of the IFN-/STAT pathway
in the growth regulation of soft tissue cancer cells is unclear. One characteristic of cancer cells is a failure to undergo growth arrest and differentiation. RMS, the most common soft tissue malignancy in childhood, is believed to originate from skeletal muscle precursor cells and is characterized by expression of several well-known myogenic regulatory factors, including MyoD (13) . Although expression of MyoD and other myogenic factors typically correlates with terminal differentiation, untreated RMS cells typically fail to arrest or to differentiate into normal muscle (14) . In RMS cell lines with abnormally low levels of p21WAF1 protein, overexpression of an exogenous p21WAF1 gene results in growth arrest but does not lead to cell differentiation (15 , 16) . These observations suggest that p21WAF1 plays an inhibitory role in the proliferation of RMS cells and that the suppression of p21WAF1 expression may provide an advantage for tumor growth.
Methylation of DNA at the CpG dinucleotides is a postreplication event
catalyzed by the DNA (cytosine-5)-methyltransferase (17)
,
which establishes normal methylation patterns during embryogenesis and
reproduces these patterns during replication of adult cells (18
, 19)
. DNA methylation, an important mechanism of epigenetic gene
regulation, is involved in genomic imprinting, X chromosome
inactivation, aging, mutagenesis, regulation of tissue-specific gene
expression during development, and latency of viral infection
(18
, 20, 21, 22)
. Alterations in the normal pattern of DNA
methylation, including an overall decrease in the genomic content of
the 5'-methylcytosines (23
, 24)
and hypermethylation of
tumor suppressor genes (25)
, are frequently found in
cancer cells. For example, hypermethylation at the p16 and
p15 CpG islands frequently correlates with loss of gene
expression in various malignancies (26, 27, 28, 29)
. Previous
studies have shown that although decreased
p21WAF1 expression is often detected in cancer
cells, mutations and allelic loss of the p21WAF1
gene are rarely detected (30
, 31)
. In light of our recent
findings that altered DNA methylation is present in RMS tumors and that
the DNA methyltransferase expression is increased in both embryonal and
alveolar subtypes of this cancer (32
, 33)
, we hypothesized
that abnormal methylation in the p21WAF1 promoter
region might play a role in the down-regulation of this gene in RMS
cells. We now show that hypermethylation of the
p21WAF1 gene at the proximal STAT-binding site,
SIE-1, correlates with decreased p21WAF1 expression
and inhibits the STAT-mediated induction of p21WAF1
expression in response to IFN- in RMS cells. Induced demethylation
at SIE-1, on the other hand, reactivates p21WAF1
expression and restores the growth-suppressive effects of IFN- in
resistant tumor cells that were hypermethylated within SIE-1.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Cell Culture.
Human RMS cell lines RD, A673, A204, Rh30, and HS729 were purchased
from the American Type Culture Collection (ATCC 136-CCL, CRL-1598,
CRL-7900, CRL-2061, and HTB-153) and grown in DMEM supplemented with
10% FCS (Life Technologies, Inc., Gaithersburg, MD). Human
IFN- and 5-aza-2'-deoxycitidine (5-aza-CdR) were purchased from
Sigma Chemical Co. (St. Louis, MO). To determine STAT activation and
p21WAF1 induction, cells were incubated in 100 ng/ml
IFN- for 1 h to 4 days. The 5-aza-CdR treatment was performed
as described previously (34
, 35)
. Cells were incubated in
0.3, 0.5, 1, 2, or 3 µM 5-aza-CdR for 1–3 days
and then in normal medium for 3 days. At the end of each treatment,
cells were trypsinized and divided for extraction of nuclear proteins
and total RNA. To determine the effects of 5-aza-CdR on cell
proliferation, RD cells were plated in 96-well plates (1
x 104 cells/well), treated with various
concentrations of 5-aza-CdR for 48 h, and then cultured for 1–4
days in normal medium or medium containing 100 ng/ml IFN-. Cell
proliferation was assessed with a cell proliferation assay kit
(Promega, Madison, WI) and detected with a SpectraMax 250 plate reader
(Molecular Devices, Sunnyvale, CA).
Methylation Analysis with Southern Blot.
The Southern blot methylation analysis was performed as described
previously (32)
. Genomic DNA was extracted from tumor and
normal tissue specimens with the Puregene DNA extraction kit (Gentra
Systems, Inc., Minneapolis, MN). DNA (5 µg) was digested with 40
units of MspI or HpaII (New England Biolabs,
Beverly, MA) at 37°C for 16 h, followed by an additional 10
units of each enzyme for 8 h. The digested DNA fragments were
separated by electrophoresis in 1.2% agarose gels and blotted onto
Hybond plus nylon membrane (Amersham Pharmacia Biotech,
Arlington Heights, IL) overnight. Hybridization probes U12 (nt –786 to
–387), U37 (nt –161 to 518), and U64 (nt –571 to 518) were
prepared by PCR amplification and subcloning into the pCR2.1 vector
(Invitrogen, Carlsbad, CA). Approximately 100 ng of the plasmid insert
DNA were 32P-labeled by random priming. All
hybridizations were performed in 2x Denhardt’s solution, 100 µg/ml
denatured salmon sperm DNA, and 6x SSC at 65°C for 24 h.
Washing was performed at 65°C in 3x, 2x, 1x, 0.5x, and 0.1x SSC,
each for 1 h. Blots were then exposed to X-ray film at –80°C
for 16–72 h. The degree of CpG methylation at SIE-1 was determined by
quantifying the intensity of the 808-bp methylation fragment relative
to the 496-bp unmethylated band with a Cyclone phosphorimager (Packard,
Meriden, CT).
RPA.
Total RNA was extracted with the Purescript RNA extraction kit
(Gentra). RNA concentration was determined by spectrophotometry. PCR
primers p21F (5'-GGG GAC AGC AGA GGA AGA C-3') and p21R (5'-CGG CGT TTG
GAG TGG TAG A-3'), synthesized based on the published
p21WAF1 mRNA sequence (GenBank accession no.
U03106), were used to generate a 159-bp fragment using Rh30 cDNA as a
template. A 169-bp ß-actin cDNA fragment was also generated as the
internal control. The fragments were subcloned into the pCR2.1 vector
by TA cloning. Inserts were sequenced to assess orientation and to
eliminate any clones with mutations attributable to subcloning.
Antisense probes for RPA were generated by linearizing the plasmid with
HindIII and using a T7 polymerase in vitro
transcription kit (Ambion, Austin, TX) in the presence of
[
-32P]UTP. Probes were purified by gel
electrophoresis. Duplex RPA reactions using probes for
p21WAF1 and ß-actin were performed with an RPA III
kit (Ambion), as recommended by the manufacturer. At least 1 µg of
total RNA was used in each reaction. Electrophoresis was performed in
5% polyacrylamide/8 M urea gels. The intensity
of the p21WAF1 and the ß-actin band was quantified
with a Cyclone phosphorimager (Packard), and the ratio was determined
for each reaction. The mean p21WAF1/ß-actin ratio
(arbitrarily set as 100%) and the SE in the normal muscle samples were
used to assess levels of p21WAF1 expression in the
tumors. Normalized values below 3 SE were considered to be decreased
expression.
Direct Sequencing Analysis.
Genomic regions containing exons 1, 2, and 3 of the
p21WAF1 gene were amplified with primers U3 (5'-AGG
TGC TCC AGG TGC TTC-3') and U7 (5'-ACT TGT AAT CCC GCT CTC C-3'),
primers E2F (5'-TGA GGT GAC ACA GCA AAG-3') and E2R (GAG AAT CCT GGT
CCC TTA C-3'), and primers E3F (5'-GGT GCG GTG ATG GAT AAA-3') and E3R
(5'-GAC TAA GGC AGA AGA TGT A-3'), respectively. The PCR products were
directly sequenced with the fmol cycle sequencing
system (Promega, Madison, WI). The sequencing reactions were
analyzed using a 6% denaturing polyacrylamide gel. The gels were
transferred onto blotting paper, dried, and exposed to Hyperfilm-MP
(Amersham Pharmacia Biotech) for 2–4 h at –80°C.
EMSA.
Nuclear extracts were prepared as described (36)
and
quantified with the Bradford method. To prepare the double-strand
oligonucleotides used as probes or competitors in EMSA, sense and
antisense strands of the oligonucleotides were synthesized separately
(Integrated DNA Technologies, Coralville, IA), annealed to
complimentary strands, electrophoresed in polyacrylamide gel, and then
purified (36)
. The methylated SIE-1 (Met-SIE-1)
oligonucleotide was prepared by incubating 20 µg of unmethylated
SIE-1 (UM-SIE-1) with 80 units of SssI CpG methylase (New
England Biolabs) for 4 h at 37°C. Then the reaction was heated
at 65°C for 30 min to inactivate the methylase, purified by PAGE, and
concentrated with Centricon 3 microconcentrators (Amicon, Danvers, MA).
EMSA probes were prepared by end labeling each double-strand
oligonucleotide with [-32P]ATP, and portions
equivalent to 20,000 cpm were used in each reaction. Binding reactions
were performed in a total volume of 20 µl in 10
mM Hepes (pH 7.9), 80 mM
KCl, 1.5 mM MgCl2, 1
mM EDTA, 1 mM DTT, 5%
glycerol, and 50 µg/ml double-strand salmon sperm DNA. Nuclear
extracts (2 µg/reaction) were incubated for 15 min on ice in the
presence or absence of unlabeled competitor oligonucleotides, followed
by addition of the end-labeled probe and a 15-min incubation on ice.
Supershift reactions were performed by incubating an antibody against
STAT1 or an antibody against STAT3 (Santa Cruz Biotechnology,
Santa Cruz, CA) with nuclear proteins for 1 h at room temperature
prior to addition of the 32P-labeled probe. The
samples were fractionated in native 4.5% polyacrylamide gels at 160 V
for 2 h in 0.25x Tris-borate EDTA. After electrophoresis, the gel
was transferred to 3 MM paper, dried, and exposed to X-ray
films. DNA-protein complexes were quantified by phosphorimaging.
Reporter Constructs, Transient Transfection, and Luciferase
Assay.
Various lengths of the p21WAF1 5'-end region were
amplified by PCR and subcloned into the pCR2.1 vector (Invitrogen). The
sequences of the inserts and their orientations were examined by direct
sequencing analysis. The reporter constructs were prepared by excising
the p21WAF1 inserts from the pCR2.1 vector and
subcloning into the pGL3-basic reporter plasmid (Promega), followed by
standard CsCl purification. The methylated plasmids (Met-pGL3-U12 and
Met-pGL3-basic) were obtained by incubating 40 µg of CsCl-purified
plasmid DNA with 100 units SssI CpG methylase (New England
BioLabs) and S-adenosylmethionine according to the
manufacturer’s instructions. The constructs were then precipitated by
ethanol and resuspended in 1x Tris-EDTA. Complete CpG
methylation of the plasmid DNA was confirmed by HpaII
digestion and gel electrophoresis. The final concentration of the
methylated plasmids was determined by comparing to the CsCl-purified
pGL3-U12 and pGL3-basic by electrophoresis in an ethidium
bromide-stained agarose gel. Twenty-four h prior to transfection,
approximately 1 x 105 cells HeLa
cells (ATCC) cultured in DMEM supplemented with 10% fetal bovine serum
(Life Technologies, Inc.) were plated in 12-well culture plates. Cells
were transfected with 2 µg of pGL3 reporter plasmid and 50 ng of the
pSEAP control plasmid (Clontech, Palo Alto, CA) using Transfast
(Promega) in a 1:1 ratio. Forty-eight h after transfection, the medium
was removed, and cells were incubated for additional 6 h in fresh
medium in the absence or presence of 100 ng/ml IFN-. Then, the cells
were harvested and assayed for luciferase activity using the LucLite
luciferase assay kit (Packard). Transfection efficiency was normalized
relative to the SEAP activity determined in the culture medium. Each
experiment was repeated at least twice in triplicate, and the variation
between experiments was less than 15%.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
). A CpG site at nt –692 relative to the transcription
start site, located within a previously identified STAT-binding site
SIE-1 (5'-CTTCCCGGAAG-3'; Ref. 9
), was present
in both unmethylated and methylated states in all normal tissues,
including skeletal muscle (n = 34), heart
(n = 5), lung (n =
6), liver (n = 8), kidney
(n = 5), pancreas (n
= 5), spleen (n = 5), and peripheral
lymphocytes (n = 26). This "incomplete"
methylation pattern was indicated by the presence of both an 808-bp
fragment attributable to CpG methylation at nt –692 and a 496-bp
fragment that was a result of unmethylated CpG sites at nt –692 and
–196 on Southern blot (Fig. 1B
). The degree of methylation
at this CpG site, as evaluated by the intensity of the 808-bp
methylation fragment relative to the 496-bp unmethylated band, was
consistent within each tissue type but varied among different tissue
types, ranging from 2.7:1 in lung to 1:15 in liver (Fig. 1B
). In the normal skeletal muscle samples, the ratio of
methylated to unmethylated SIE-1 was 3.1–3.5:1 (Fig. 1C
).
The other HpaII sites in the 1.5-kb region, including 11
sites clustered in the CpG-rich region surrounding the
p21WAF1 exon 1, were unmethylated in all normal
tissues examined in this study (data not shown).
|
).
The rest of the tumors (6 embryonal and 7 alveolar) showed the
incomplete methylation pattern similar to that of normal
skeletal muscle. In addition, we examined five RMS cell lines
(A204, RD, A673, Rh30, and HS729), for CpG methylation in the
p21WAF1 upstream region. Complete methylation at
SIE-1 was found in RD and A673, whereas A204, Rh30, and HS729 cells
contained both unmethylated and methylated CpG site within SIE-1 (Fig. 1C
). The hypermethylated CpG within SIE-1, which was found
in a relatively CpG-sparse region upstream of the
p21WAF1 CpG island, was the only methylation
alteration identified in this study. No alteration of CpG methylation
pattern was detected in the primary RMS samples or in the RMS cell
lines with regard to the downstream HpaII sites that span
the CpG-rich region of the p21WAF1 gene (not shown).
Correlation between SIE-1 Methylation and the
p21WAF1 mRNA Expression in RMS.
To determine whether the aberrant CpG methylation in the
p21WAF1 promoter region correlates with decreased
p21WAF1 expression in RMS tumors, we analyzed levels
of the p21WAF1 mRNA transcripts in the same set of
tumor and normal muscle samples using RPA. Levels of the
p21WAF1 mRNA were normalized to an internal standard
ß-actin, and decreased expression was defined as less than 3 SE of
the mean expression level in the normal muscle samples. All 13 tumors
with completely methylated SIE-1 showed decreased
p21WAF1 expression; in contrast, decreased
p21WAF1 expression was detected in only one of the
six embryonal tumors and none of the seven alveolar tumors with
incomplete SIE-1 methylation. The correlation between SIE-1
hypermethylation and decreased p21WAF1 expression in
RMS tumors is summarized in Table 1
, with representative RPA results shown in Fig. 2
. Compared to the mean expression in normal muscle, the average
p21WAF1 levels were significantly lower in both
tumor subtypes with complete methylation at SIE-1, whereas the
difference between the incompletely methylated tumors and normal muscle
was not statistically significant (Table 1)
. Consistent with the
results from the primary tumor samples, decreased
p21WAF1 expression was observed in RD and A673, the
two cell lines with complete methylation at SIE-1 (Fig. 2D
).
Of the five RMS cell lines examined, Rh30, an alveolar RMS cell line
with the least degree of CpG methylation at SIE-1, showed the highest
p21WAF1 expression (Fig. 2D
). These
results indicated that complete CpG methylation within SIE-1 strongly
correlated with decreased constitutive expression of
p21WAF1 in RMS cells.
|
|
A variant or the C T variant at the 20th bp after the stop
codon. These results further supported an epigenetic mechanism for the
decreased p21WAF1 expression in RMSs.
CpG Methylation within SIE-1 Inhibits DNA Binding of STAT1.
Abnormal CpG methylation in the promoter region, when present within
regulatory elements, could potentially interfere with binding of
specific transcription factors that recognize these motifs
(37)
. To determine whether CpG methylation within SIE-1
interferes with STAT binding, we compared the STAT binding abilities of
a 27-bp oligonucleotide (nt –703 to –677) containing SIE-1 in
the unmethylated (UM-SIE-1) and methylated (Met-SIE-1) forms, using
EMSA. We first examined the abilities of UM-SIE-1 and Met-SIE-1 to
compete with a STAT consensus sequence S1 (8)
in binding
to activated STAT proteins, using competition EMSA reactions with
nuclear proteins from RD cells treated with 100 ng/ml IFN- for
24 h (Fig. 3A
). Activation of STAT1 was induced by IFN-, as indicated
by the diminished DNA-protein complex and the formation of a supershift
band when an antibody against STAT1 was added to the reaction (Fig. 3A
, Lane 15). The addition of an antibody against STAT3 had
no effect on complex formation (Fig. 3A
, Lane 16),
indicating that STAT3 was not activated. In competition EMSA reactions,
addition of 100-fold excess of unlabeled S1 or UM-SIE-1 inhibited
complex formation on an equimolar basis, suggesting that the
unmethylated SIE-1 binds activated STAT1 as strongly as the S1
consensus sequence (Fig. 3A
, Lanes 3–6). However, addition
of the same amount (100-fold excess) of unlabeled Met-SIE-1 resulted in
lesser inhibition of DNA-protein complex formation. Under identical
conditions, similar inhibition was observed when the Met-SIE-1
competitor was increased to 500-fold excess (Fig. 3A
, Lanes
7–9). Similar results were observed by using nuclear proteins
from IFN--treated A673 or A204 cells or by substituting the S1 probe
with a high-affinity STAT-binding probe M67 (38)
in the
reactions (not shown). We then directly compared the abilities of
unmethylated and methylated SIE-1 to bind activated STAT1 in EMSA
reactions, using 32P-labeled UM-SIE-1 or
Met-SIE-1 probes prepared to have equal specific activities (Fig. 3B
). Nuclear proteins from RD cells treated with 100 ng/ml
IFN- for 1 h to 4 days were used in these reactions.
Binding of STAT1 to UM-SIE-1 was observed throughout the IFN-
treatment (Fig. 3B
, Lanes 1–6). In contrast, binding of
activated STAT1 to the methylated SIE-1 probe was significantly
inhibited. Under identical conditions, the STAT complexes formed with
Met-SIE-1 were in average 5.2 times less intense than with the UM-SIE-1
probe (Fig. 3B
, Lanes 7–12). EMSA reactions using nuclear
proteins from IFN--treated A673 or A204 cells yielded similar
results (not shown). Therefore, CpG methylation within SIE-1
significantly inhibited binding of activated STAT1 to this site.
|
, and analyzed for luciferase
activity (Fig. 4
). IFN- induced promoter activities of the constructs containing the
SIE-1 site. As shown in Fig. 4
, the promoter activity of two
SIE-1-containing constructs, pGL3-U18 and pGL3-U12, increased by 3- and
2.8-fold, respectively, with the IFN- treatment. Removal of the
SIE-1 site from the reporter construct (pGL3-U68) virtually eliminated
the response to IFN- stimulation, indicating that SIE-1 was an
essential cis-acting element for IFN--induced activation
of the p21WAF1 promoter. We then determined the
promoter activity of the SssI-modified reporter construct
pGL3-U12, because the CpG dinucleotide within SIE-1 is the only CpG in
the U12 fragment. CpG methylation of pGL3-U12 (Met-pGL3-U12) reduced
its promoter activity by an average of 66% (P
< 0.001, Student’s t test) and completely inhibited
the induction of transcription activation by IFN- (Fig. 4
). In
contrast, methylation of the control plasmid (Met-pGL3-basic) resulted
in less than 12% decrease of the average luciferase activity, a loss
that was statistically insignificant (P >
0.05, Student’s t test). These results suggested that
decreased STAT-binding to methylated SIE-1 inhibited STAT-mediated
transcriptional activation.
|
and other extracellular
factors has been correlated with induction of
p21WAF1 mRNA expression and cell growth suppression
(9
, 11
, 12)
. In light of the results described above, we
hypothesized that hypermethylation within SIE-1 could result in
deficient STAT-mediated up-regulation of p21WAF1
expression and resistance to the antiproliferative effects of IFN-
in RMS cells. To test this hypothesis, we studied STAT activation,
p21WAF1 expression, and cell growth response in RD,
A673, and A204 cells treated with 100 ng/ml IFN- for 1 h to 3
days. Using EMSA, STAT activation was detected in all three cell lines,
evidenced by the presence of the STAT complexes in all treated cells
(Fig. 5A
). We then performed RPA analysis to determine the induction
of the p21WAF1 mRNA expression by IFN- (Fig. 5B
). IFN--treated A204 cells showed significantly
increased p21WAF1 expression. As shown in Fig. 5B
, the increase of the p21WAF1
expression was 4.5-fold after a 6-h IFN- treatment and 3.4-fold
after a 1-day treatment. In contrast, in RD and A673 cells, both of
which were completely methylated within SIE-1, no increase of
p21WAF1 expression was induced at any time of
IFN- treatment (Fig. 5B
). To determine whether induction
of p21WAF1 correlated with suppression of cell
growth, we analyzed the growth responses of A204, RD, and A673 cells to
IFN- treatment. The growth of A204 cells was inhibited by IFN-
treatment, whereas RD and A673 cells were resistant to the treatment
and continued to proliferate in the presence of IFN- (Fig. 6
). Thus, p21WAF1 induction and cell growth
suppression correlated with STAT activation in A204 cells, which
contained unmethylated SIE-1; however, the responses to IFN- were
inhibited in RD and A673 cells because of complete methylation within
SIE-1.
|
|
.
). Demethylation at SIE-1 resulted in significantly
increased p21WAF1 expression, which was further
induced by IFN- in the 5-aza-CdR-treated RD cells (Fig. 7B
). The effect of the demethylation treatment on cell
growth was determined with a cell proliferation assay, which measured
the growth responses of 5-aza-CdR-treated and control RD cells to
IFN-. IFN- did not affect the proliferation of control RD cells
but did result in growth inhibition of the 5-aza-CdR-treated RD cells
(Fig. 8
). The differences in cell proliferation rate between RD cells treated
with 5-aza-CdR only and those treated with 5-aza-CdR followed by
IFN- were statistically significant (Fig. 8
). Therefore, induced
demethylation at SIE-1 by 5-aza-CdR reactivated
p21WAF1 expression and restored the responsiveness
to IFN- in the previously resistant RD cells.
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
-induced
expression of the p21WAF1 mRNA in RMS cells.
Expression of the p21WAF1 gene is regulated by
p53-dependent and p53-independent mechanisms (30)
. MyoD, a
basic helix-loop-helix protein, activates p21WAF1
expression in a p53-independent pathway and induces cell cycle
withdrawal during normal muscle differentiation, a process accompanied
by down-regulation of the activity of CDKs (39
, 40)
.
Previous studies have indicated that both the p53-dependent and the
MyoD-mediated pathways of p21WAF1 induction are
inactivated in RMS cells (15)
. Our study shows that
complete CpG methylation at nt –692 in the SIE-1 element is a frequent
occurrence in both the embryonal and the alveolar subtypes of RMS
tumors. CpG methylation within SIE-1 inhibited binding of activated
STAT1, resulting in abrogation of the STAT-mediated induction of
p21WAF1 expression and the growth suppression in
response to IFN-. More importantly, induced demethylation at SIE-1
by 5-aza-CdR treatment reactivated p21WAF1
expression and restored cellular responsiveness to IFN-. Although it
is unclear whether a decrease in p21WAF1 expression
contributes to tumor progression in vivo, our findings have
demonstrated that altered DNA methylation is a new mechanism for
down-regulation of the p21WAF1 expression in RMS
cells.
In our study, hypermethylation was detected at a single CpG site within
SIE-1, which is present in a subregion with very few other CpG sites
upstream of the p21WAF1 CpG island. Previous studies
have shown that methylation affects promoter regions with and without
CpG islands in a differential manner (25
, 41) . Except for
X-linked inactivated genes and autosomal imprinted genes, promoters
that contain CpG islands are virtually always unmethylated in normal
tissues. It has been well established that hypermethylation of promoter
CpG islands leads to inactivation of tumor suppressor genes in cancer
cells, as demonstrated by inactivation of the VHL gene
caused by regional hypermethylation in renal carcinoma
(42)
. In contrast, CpG sites in promoter regions without
CpG islands are variably methylated in normal cells, often in
tissue-specific patterns that reflect the transcription status of the
gene (41)
. In these promoter regions, methylation of
specific CpG sites in or near transcription regulatory motifs can block
accessibility of the transcription factors. For example, a
SnaBI CpG site in the IFN- gene promoter was completely
methylated in neonatal T cells and thymocytes that had low or
undetectable IFN- expression but was substantially hypomethylated in
adult T cells with greater levels of IFN- mRNA in response to
activation (43
, 44)
. In vitro methylation of
this SnaBI CpG site, which was found within a regulatory
element critical for IFN- gene transcription, was shown to block
binding of nuclear proteins in IFN--expressing T cells
(44)
. Recently, several groups have reported that
site-specific methylation alterations may result in perturbed gene
expression in cancer cells by affecting binding of transcription
factors. Hypermethylation at cAMP-responsive element in the
promoter regions of BRCA1 and NF1 inhibits
binding of the cAMP-responsive element binding protein
(37
, 45)
. Similarly, CpG methylation within the ATF-like
and the RBF1 recognition sites in the Rb1 promoter region prevents
binding of these factors and results in decreased Rb
expression (46)
. In our study, hypermethylation within the
SIE-1 element resulted in reduced STAT1 binding, leading to inhibition
of the STAT-mediated induction of the p21WAF1
expression. These results not only support the notion that methylation
alteration within regulatory elements in non-CpG island regions can
directly affect transcription regulation but also demonstrate that
altered DNA methylation can directly impair signal transduction in
cancer cells.
Our study indicates that the CpG dinucleotide within SIE-1 is maintained at a partially methylated state in normal tissues in a tissue-specific pattern. In RMS cells, hypermethylation within SIE-1 reflects a gain in the degree of methylation at this CpG site. It is not known how methylation patterns at this CpG site were established for different cell types during development, nor is it clear to what degree the specific methylation patterns are involved in regulating p21WAF1 expression in normal tissues. Although complete CpG methylation within SIE-1 strongly correlates with decreased constitutive p21WAF1 expression in RMS cells, it is unlikely that transcription abrogation is caused by altered methylation of a single CpG located at 692 bp upstream from the transcription start site. The 11 HpaII sites spanning the p21WAF1 CpG island are invariably unmethylated both in normal tissues and in RMS cells; however, because Southern blot analysis examines only a portion of the CpG sites in the p21WAF1 promoter, it is possible that abnormal methylation is present at additional, non-HpaII CpG sites in this region. To put the methylation alteration at SIE-1 into a more defined context, future studies are necessary to determine the complete CpG methylation profile of the p21WAF1 upstream region. It is also possible that other epigenetic mechanisms, such as changes in histone acetylation and heterochromatin structure, are involved in regulating p21WAF1 expression. Therefore, although our study indicates a strong correlation between hypermethylation within SIE-1 and decreased constitutive p21WAF1 expression in primary RMS tumors, the role of CpG methylation in p21WAF1 transcription regulation merits further elucidation.
SIE-1 is the most proximal STAT-binding site among the three SIE elements in the p21WAF1 promoter region and is likely the only one directly affected by abnormal DNA methylation, because SIE-2 and SIE-3 do not contain CpG dinucleotides (9) . STAT1 was shown to bind all three SIEs in in vitro protein-DNA binding experiments (9) ; however, how these SIE sites interact with activated STAT1 in vivo is unknown. Our study indicates that deficient protein binding to SIE-1 caused by hypermethylation is sufficient to inhibit the STAT1-mediated transcriptional activation of p21WAF1. These results have raised questions such as whether SIE-2 and SIE-3 bind activated STAT1 in vivo and how they function in the presence of hypermethylated SIE-1. Further investigation is needed to elucidate the relationship of all three SIE elements to the STAT-mediated p21WAF1 up-regulation.
Regulation of p21WAF1 expression through methylation
of a STAT-binding site adds a new facet to our understanding of the
interplay between DNA methylation, cell cycle control, and tumor
progression. p21WAF1 has been shown to form
quaternary complexes with cyclin, CDK, and the proliferating cell
nuclear antigen and to function as a negative regulator of cell
proliferation by inhibiting CDK activity and blocking the
G1-S transition (2
, 47)
. On the
other hand, p21WAF1 might also negatively
regulate DNA methylation, as it was shown to compete with DNA
(cytosine-5)-methyltransferase for binding to proliferating cell
nuclear antigen (48)
, thus antagonizing the DNA
methyltransferase function. Our results indicate that the
p21WAF1 gene is subject to methylation regulation at
the transcription level and is a target of aberrant methylation in RMS
cells. Decreased p21WAF1 expression has been found
in a number of human cancers and appears to be associated with a more
aggressive phenotype of malignant melanoma (49, 50, 51, 52, 53)
;
therefore, the role of abnormal p21WAF1
methylation in the development and/or progression of these tumors
is an intriguing question. Furthermore, our findings may have important
therapeutic implications, because 5-aza-CdR treatment reactivates
p21WAF1 expression and restores the responsiveness
to STAT activation by inducing demethylation at SIE-1 in tumor cells
that were resistant to IFN-. The possible application of this
observation to cancer chemotherapy deserves further exploration.
|
ACKNOWLEDGEMENTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
1 Supported by Arkansas Science and Technology
Authority Grant 98-B-34, American Cancer Society Institutional Research
Grant IRG-91-012-06, National Cancer Institute Grants P30 CA217657 and
5 U10 CA24507–09, the American Lebanese-Syrian Associated Charities,
and the Sam Walton and Ed Harms Research Awards from the University of
Arkansas for Medical Sciences Medical Research Endowment Fund. 
2 Current address: Division of Laboratory Systems, Public
Health Practice Program Office, Centers for Disease Control and
Prevention, 4770 Buford Highway, N.E., Mailstop G25, Atlanta, GA 30341.
3 To whom requests for reprints should be
addressed, at Slot 820, Arkansas Children’s Hospital, 800 Marshall
Street, Little Rock, AR 72202. Phone: (501) 320-1307; Fax:
(501) 320-3912; E-mail: WBMX95A{at}prodigy.com
4 The abbreviations used are: CDK,
cyclin-dependent kinase; STAT, signal transducers and activators of
transcription; RMS, rhabdomyosarcoma; RPA, RNase protection assay;
EMSA, electrophoretic mobility shift assay; SIE, sis-inducible element;
5-aza-CdR, 5-aza-2'-deoxycytidine; nt, nucleotide(s); SEAP, secreted
alkaline phosphatase.
Received 11/ 3/99. Accepted 4/10/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
. Proc. Natl. Acad. Sci. USA, 93: 7673-7678, 1996. induces cell growth inhibition by Fas-mediated apoptosis: requirement of STAT1 protein for up-regulation of Fas and FasL expression. Cancer Res., 58: 2832-2837, 1998. gene. J. Immunol., 153: 3603-3610, 1994.[Abstract]
gene correlates with its expression by primary T-lineage cells. Eur. J. Immunol., 25: 426-430, 1995.[Medline]
This article has been cited by other articles:
|
|
 |
|
  L. Escoubet-Lozach, I-L. Lin, K. Jensen-Pergakes, H. A. Brady, A. K. Gandhi, P. H. Schafer, G. W. Muller, P. J. Worland, K. W.H. Chan, and D. Verhelle Pomalidomide and Lenalidomide Induce p21WAF-1 Expression in Both Lymphoma and Multiple Myeloma through a LSD1-Mediated Epigenetic Mechanism Cancer Res., September 15, 2009; 69(18): 7347 - 7356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Klampfer, J. Huang, S. Shirasawa, T. Sasazuki, and L. Augenlicht Histone Deacetylase Inhibitors Induce Cell Death Selectively in Cells That Harbor Activated kRasV12: The Role of Signal Transducers and Activators of Transcription 1 and p21 Cancer Res., September 15, 2007; 67(18): 8477 - 8485. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Riuzzi, G. Sorci, and R. Donato RAGE Expression in Rhabdomyosarcoma Cells Results in Myogenic Differentiation and Reduced Proliferation, Migration, Invasiveness, and Tumor Growth Am. J. Pathol., September 1, 2007; 171(3): 947 - 961. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Egger, A. M. Aparicio, S. G. Escobar, and P. A. Jones Inhibition of Histone Deacetylation Does Not Block Resilencing of p16 after 5-Aza-2'-Deoxycytidine Treatment Cancer Res., January 1, 2007; 67(1): 346 - 353. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nishigaki, C. Hanson, T. Ohashi, A. Spadaccini, and S. Ruscetti Erythroblast Transformation by the Friend Spleen Focus-Forming Virus Is Associated with a Block in Erythropoietin-Induced STAT1 Phosphorylation and DNA Binding and Correlates with High Expression of the Hematopoietic Phosphatase SHP-1. J. Virol., June 1, 2006; 80(12): 5678 - 5685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Gartel and S. K. Radhakrishnan Lost in Transcription: p21 Repression, Mechanisms, and Consequences Cancer Res., May 15, 2005; 65(10): 3980 - 3985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Grandemange, P. Seyer, A. Carazo, P. Becuwe, L. Pessemesse, M. Busson, C. Marsac, P. Roger, F. Casas, G. Cabello, et al. Stimulation of Mitochondrial Activity by p43 Overexpression Induces Human Dermal Fibroblast Transformation Cancer Res., May 15, 2005; 65(10): 4282 - 4291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Chatterji, J. E. Riby, T. Taniguchi, E. L. Bjeldanes, L. F. Bjeldanes, and G. L. Firestone Indole-3-carbinol stimulates transcription of the interferon gamma receptor 1 gene and augments interferon responsiveness in human breast cancer cells Carcinogenesis, July 1, 2004; 25(7): 1119 - 1128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Staalesen, B. Leirvaag, J. R. Lillehaug, and P. E. Lonning Genetic and Epigenetic Changes in p21 and p21B Do Not Correlate with Resistance to Doxorubicin or Mitomycin and 5-Fluorouracil in Locally Advanced Breast Cancer Clin. Cancer Res., May 15, 2004; 10(10): 3438 - 3443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ying, G. Srivastava, Z. Gao, X. Zhang, P. Murray, R. Ambinder, and Q. Tao Promoter hypermethylation of the cyclin-dependent kinase inhibitor (CDKI) gene p21WAF1/CIP1/SDI1 is rare in various lymphomas and carcinomas Blood, January 15, 2004; 103(2): 743 - 746. [Full Text] [PDF]
|
|
|
|
|
|
N. Sato, N. Fukushima, A. Maitra, H. Matsubayashi, C. J. Yeo, J. L. Cameron, R. H. Hruban, and M. Goggins Discovery of Novel Targets for Aberrant Methylation in Pancreatic Carcinoma Using High-Throughput Microarrays Cancer Res., July 1, 2003; 63(13): 3735 - 3742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-G. Zhu, K. Srinivasan, Z. Dai, W. Duan, L. J. Druhan, H. Ding, L. Yee, M. A. Villalona-Calero, C. Plass, and G. A. Otterson Methylation of Adjacent CpG Sites Affects Sp1/Sp3 Binding and Activity in the p21Cip1 Promoter Mol. Cell. Biol., June 15, 2003; 23(12): 4056 - 4065. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Nanni, G. Nicoletti, C. De Giovanni, S. Croci, A. Astolfi, L. Landuzzi, E. Di Carlo, M. Iezzi, P. Musiani, and P.-L. Lollini Development of Rhabdomyosarcoma in HER-2/neu Transgenic p53 Mutant Mice Cancer Res., June 1, 2003; 63(11): 2728 - 2732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. He, M. J. J. Ronis, and T. M. Badger Ethanol Induction of Class I Alcohol Dehydrogenase Expression in the Rat Occurs through Alterations in CCAAT/Enhancer Binding Proteins beta and gamma J. Biol. Chem., November 8, 2002; 277(46): 43572 - 43577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Shen, Y. Kondo, J.-P. Issa, G. Garcia-Manero, J. Roman-Gomez, J. A. Castillejo, A. Torres, and A. Jimenez Lack of p21CIP1 DNA methylation in acute lymphocytic leukemia Blood, October 16, 2002; 100(9): 3432 - 3433. [Full Text] [PDF]
|
|
|
|
|
|
A. Widschwendter, S. Tonko-Geymayer, T. Welte, G. Daxenbichler, C. Marth, and W. Doppler Prognostic Significance of Signal Transducer and Activator of Transcription 1 Activation in Breast Cancer Clin. Cancer Res., October 1, 2002; 8(10): 3065 - 3074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Hong, Hyeon.-A. Kim, G. L. Firestone, and L. F. Bjeldanes 3,3'-Diindolylmethane (DIM) induces a G1 cell cycle arrest in human breast cancer cells that is accompanied by Sp1-mediated activation of p21WAF1/CIP1 expression Carcinogenesis, August 1, 2002; 23(8): 1297 - 1305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Roman-Gomez, J. A. Castillejo, A. Jimenez, M. G. Gonzalez, F. Moreno, M. d. C. Rodriguez, M. Barrios, J. Maldonado, and A. Torres 5' CpG island hypermethylation is associated with transcriptional silencing of the p21CIP1/WAF1/SDI1 gene and confers poor prognosis in acute lymphoblastic leukemia Blood, April 1, 2002; 99(7): 2291 - 2296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. TEGEDER, J. PFEILSCHIFTER, and G. GEISSLINGER Cyclooxygenase-independent actions of cyclooxygenase inhibitors FASEB J, October 1, 2001; 15(12): 2057 - 2072. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-L. Santiago-Raber, B. R. Lawson, W. Dummer, M. Barnhouse, S. Koundouris, C. B. Wilson, D. H. Kono, and A. N. Theofilopoulos Role of Cyclin Kinase Inhibitor p21 in Systemic Autoimmunity J. Immunol., October 1, 2001; 167(7): 4067 - 4074. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
