| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
E. A. Doisy Department of Biochemistry and Molecular Biology Saint Louis University Health Sciences Center St. Louis, Missouri 63104
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The recent discovery of the cAMP response element modulator (CREM)
proteins suggests that the activity of the functional CREB dimer is
regulated not only by its phosphorylation state, but also by
protein/protein interactions (6, 7). CREM
, CREMß, CREM
, and
inducible cAMP early repressor (ICER) are alternatively spliced
isoforms of the CREM gene, whose protein products function to repress
CREB activity either by the formation of a nonfunctional CREB:CREM
heterodimer or by competition between CREM and CREB homodimers for
binding to a CRE. The ICER isoform utilizes a distinct promoter located
in an intron and is itself induced by cAMP (6). The induction of ICER
was originally identified in neuroendocrine cells but more recently has
been shown to occur during some stages of lymphocyte development (8).
The cAMP inducibility of the ICER promoter may be due to the presence
of four cAMP-autoregulatory response elements (CAREs) located in
two clusters of two elements each (6). Each cluster was shown to exert
cAMP inducibility in JEG-3 choriocarcinoma cells and was able to bind
the ICER protein in vitro. Moreover, ICER is proposed to
down-regulate its own expression and to down-regulate the expression of
cAMP-inducible genes. Thus, CREM/ICER may function to attenuate the
transcriptional response to cAMP.
The cellular effects of agents that increase cAMP are varied. For example, cAMP stimulates proliferative pathways in some endocrine cells and antiproliferative pathways in many immune cells. Increases in cAMP levels in developing thymocytes block differentiation by blocking cell cycle progression and by inducing apoptosis in a subset of cells (9, 10). Furthermore, cAMP-induced apoptosis in thymocytes is preceded by a G1 block in the cell cycle (11) and proceeds through a mechanism requiring activated gene expression (12, 13). The molecular bases of cAMP cell cycle arrest and apoptosis are poorly understood.
In this study, we examined the effect of forskolin, an activator of adenylate cyclase, on the expression of the immediate early genes c-fos, junB, and ICER in the apoptotic thymocyte. The c-fos and junB protooncogenes have previously been shown to be cAMP-responsive, and each have been implicated in apoptosis induced by a variety of stimuli (14, 15, 16, 17, 18). Moreover, junB expression has been implicated as a negative regulator of cell growth in neuronal cells, fibroblasts, and mammary carcinoma cells (19, 20). ICER regulation by cAMP was detected recently in developing and mature human T lymphocytes, and it was suggested that ICER may function in the circadian control of lymphocyte development and activity (8) similar to its proposed role in the pineal gland (21, 22). Because of the importance of cAMP signaling in the developing thymocyte, it was of considerable interest to determine whether these immediate early genes were induced during cAMP-induced apoptosis. We observed a rapid and transient induction of the c-fos and junB protooncogenes and a delayed induction of ICER expression. The kinetics of ICER expression was similar for the thymoma cells and a neuroendocrine cell line; however, there were qualitative differences that may be explained by thymoma-specific complexes bound to the ICER promoter. Moreover, ICER induction precedes DNA fragmentation in the WEHI7.2 cells and thus may represent an early step in the apoptotic cascade. The induced endogenous ICER expression correlated with the attenuated expression of c-fos and a reduced responsiveness to a second treatment of forskolin. Thus, ICER may function in the attenuation of the cAMP response, in an early transcriptional refractory period, as well as in apoptosis of thymocytes.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). An increase in steady-state CREM
mRNA levels was detected within 1.5 h. A maximal 10-fold increase
in CREM RNA levels was observed at 3.5 h, after which time
expression was reduced essentially to basal levels by 24 h. This
gradual decline after CREM induction is consistent with negative
autoregulation by CREM/ICER on ICER gene transcription (6).
c-fos and junB were also induced by forskolin,
although the kinetics of induction are significantly different than
CREM. c-fos and junB steady-state mRNA levels
increased within 20 min and were maximal at 1.5 h. Transcript
levels were attenuated to basal levels by 3.5 h and were reduced
to below basal levels by 6 h. Thus, the increase in CREM
expression correlated with the decrease in c-fos and
junB.
|
). Moreover, in the
cycloheximide-treated cells, the return of c-fos and
junB mRNA to basal levels was inhibited. Similarly,
cycloheximide was ineffective at blocking the induction of CREM
transcripts. Three hours of cycloheximide and forskolin cotreatment led
to a 26-fold induction of CREM whereas forskolin alone induced CREM
expression 10-fold. These data are consistent with the hypothesis that
the ICER protein represses its own expression in a negative
autoregulatory loop and suggests that CREM/ICER may also function to
attenuate the cAMP-induced expression of c-fos and
junB.
The relative kinetics of induced gene expression and cell death were
compared. Because WEHI7.2 cells excluded vital dyes for up to 16 h
after treatment with forskolin (D. Dowd, unpublished observations), we
used DNA fragmentation as a marker of apoptosis. During apoptosis, the
DNA is cleaved into double-stranded fragments whose lengths are
multiples of 180 bp, the length associated with the nucleosome (25, 26). DNA cleavage was observed after 3 h of forskolin treatment,
and the concentration of fragments increased over time (Fig. 1C). Thus,
the induction of CREM/ICER gene expression precedes DNA fragmentation
by approximately 1.5 h.
ICER Is the CREM Isoform Induced by Forskolin
The CREM gene is spliced to yield multiple
isoforms of the protein, which can differ by the presence or absence of
the N terminus and by the DNA-binding domain used (6, 7). Using a
full-length CREM cDNA probe, we detected induction of four
transcripts: 2.3, 1.6, 1.4, and 1.1 kb. To specifically identify which
of the CREM products were regulated, we used probes specific for the N
terminus and C terminus of CREM (Fig. 2A). The 3'-CREM probe, which encodes
C-terminal sequences common to all CREM and ICER isoforms, demonstrated
an induction pattern virtually identical to the full- length probe with
an approximately 10-fold induction in each of the four transcripts. In
contrast, the 5'-probe, specific for CREM, -ß, -, and -
but
absent in the ICER isoforms, demonstrated multiple transcripts with no
induction of expression of the 2.3-kb band and a 1.6-fold induction of
the smaller 1.4- and 1.1-kb species. JEG-3 choriocarcinoma cells
treated with forskolin demonstrated only a single transcript
hybridizing to the 5'-sequence that was not regulated by cAMP. A
pattern of induced transcripts similar to WEHI7.2 was observed when the
3'-ICER-specific probe was used (Fig. 2B), consistent with that
previously reported (6). It is interesting to note that although the
kinetics of induction appear similar in WEHI7.2 and JEG-3, the pattern
of transcripts was slightly different. Whereas the level of the 2.3-kb
transcript increased 10-fold in WEHI7.2 cells, it represented only a
small fraction of the regulated transcripts in the JEG-3 cells.
|
). Probe I was used to detect DBD I, specifically.
A DBD I-specific fragment was protected in ethanol-treated cells (lane
4), which comigrated with the protected band resulting from
hybridization with CREM. This band was induced in cells treated with
forskolin (lane 5). These observations were confirmed using probe II, a
probe that contained a portion of DBD II and noncoding sequence between
DBD I and DBD II. This noncoding sequence is spliced from ICER II, and
thus this probe generates two products of different sizes when
hybridized to CREM/ICER isoforms encoding DBD I, DBD II, and the
intervening sequence (e.g. ICER I, CREM ; lane 8) or
those encoding only DBD II (e.g. ICER II, CREM ß; lane 9).
We observed a forskolin-induced increase in the protected band
comigrating with DBD I, and no detectable band corresponding to the
exclusive use of DBD II (compare lanes 10 and 11). Thus, these data
indicate that the ICER isoform induced by forskolin in WEHI7.2 cells
utilizes DBD I.
CARE3/4 Directs cAMP-Induced Expression in WEHI7.2 Cells
Because of the observed differences in transcript pattern between
WEHI7.2 thymocytes and JEG-3 choriocarcinoma cells (Fig. 2, A and B),
ICER promoter activity was examined in transient expression studies in
these two cell lines. The ICER native promoter, containing
approximately 400 bp of 5'- untranslated sequence, was isolated by PCR
amplification of genomic DNA and cloned into a CAT reporter vector.
This ICER-CAT construct conveyed cAMP- inducible CAT activity in
WEHI7.2 and JEG-3 cells (Fig. 3). The
ICER promoter encodes a cluster of four CRE-like elements (termed
cAMP-autoregulatory elements, CAREs) that direct cAMP inducibility in
JEG-3 choriocarcinoma cells (6). The 62-bp cluster of four CAREs
conferred cAMP inducibility to the thymidine kinase (tk) promoter (27),
as did two copies of the CREs from the -glycoprotein hormone
promoter (1, 27, 28). This CARE1–4 construct was positively responsive
to forskolin in both cell lines. The CAREs were further divided into
two segments, CARE1/2 and CARE3/4, and tested for cAMP inducibility in
WEHI7.2 and JEG-3 cells. CARE1/2 demonstrated cAMP inducibility in
JEG-3 cells, in agreement with the observations of Molina et
al. (6), but it was ineffective in WEHI7.2 cells. In contrast,
CARE3/4 was responsive to forskolin in both JEG-3 and WEHI7.2 cells.
However in WEHI7.2 cells, the level of CAT activity measured utilizing
the CARE3/4tkCAT construct was 3-fold less than that obtained with
CARE1–4tkCAT. This suggests that although CARE1/2 was unresponsive to
cAMP, it augments the activation of CARE3/4. CARE3 and CARE4 were
examined separately for the ability to induce expression from a
CAREtkCAT expression vector. Both CARE3 and CARE4 were capable of
sustaining forskolin-activated transcription in WEHI7.2 and JEG-3 cells
(data not shown).
|
). Whereas CREB and CREM bound
avidly to the CARE3/4 oligonucleotide (Fig. 4A, lanes 9–14), binding
of equivalent amounts of extracts to CARE1/2 was greatly reduced (lanes
2–7). Note that at the highest levels of CREB or CREM protein
extracts, virtually all of the CARE3/4 probe is bound, whereas only a
small fraction of the CARE1/2 probe is bound by these proteins. When
CARE3/4 was used as the probe, both the CREB and CREM extracts resulted
in two major bands of retarded mobility. These complexes may correspond
to the occupancy of one CARE (band I) and two CAREs (band II). CARE1/2
demonstrated only a single major band with the CREB and CREM extracts
comigrating with CREB band I and CREM band I, respectively. These data
suggest that only one of the two CARE sites is occupied at these
protein concentrations.
|
). As before, the level of CREB protein used
resulted in the appearance of two slowly migrating bands (lane 1),
presumably due to the occupancy of one or both CAREs. The amount of
CREM extract used in this experiment produced primarily one band,
corresponding to the occupancy of one CARE (lane 9, Fig. 4B; and lane
12, Fig. 4A). A 20-fold excess of unlabeled CARE3/4 was effective at
eliminating CREB-specific band II (lane 6) and reducing the
CREM-specific band by 50% (lane 14), whereas a 100-fold excess
completely eliminated binding of labeled CARE3/4 to extracts containing
recombinant CREB or CREM (lanes 7 and 15). A 100-fold excess of CARE1/2
was an ineffective competitor when CREB or CREM protein was bound to
CARE3/4 (lanes 4 and 12). When CREB was bound to the CARE3/4 probe,
there did appear to be slight reduction in band II with 100-fold excess
of CARE1/2. This suggests that there was a small competition effect of
the CARE1/2 oligonucleotide reducing the amount of probe in which both
CARE sites were occupied. Both CARE3 and CARE4 were efficient at binding CREB and CREM. Indeed, the binding affinities of CREB and CREM to each of these sequences were similar. Gel mobility shift DNA-binding assays demonstrated that both CARE3 and CARE4 were effective competitors for CREB and CREM binding to a 32P-labeled CARE3 probe (data not shown). The ability of CREB to bind to both CARE3 and CARE4 is consistent with the ability of both of these elements to confer cAMP responsiveness on a heterologous reporter system (Ref. 6 and data not shown).
To further address the question of differential behavior of
CARE-driven transcription in WEHI7.2 and JEG-3 cells, we examined the
binding of nuclear proteins to the CARE1/2 and CARE3/4 oligonucleotides
(Fig. 5). Increasing concentrations of
nuclear extracts were incubated with the CARE3/4 oligonucleotide and
examined by gel mobility shift analysis (Fig. 5A). Extracts from both
cell lines formed a protein/DNA complex that comigrated with the
baculoviral-expressed CREB protein. Its identity as CREB was confirmed
with the use of an anti-CREB antibody, which resulted in a supershifted
protein/antibody/DNA complex. The WEHI7.2 extract also formed a complex
of slower mobility (denoted by the open arrow, lanes 3–5),
which was absent in the JEG-3 extract (lanes 7–9). This complex was
not supershifted by the anti-CREB antibody. However, the supershifted
CREB band comigrated with this unique band, making it difficult to
determine whether the antibody adversely affected it.
|
). Both
WEHI7.2 and JEG-3 extracts formed complexes that comigrated with
baculoviral-expressed CREB (filled arrow), however, these
complexes did not interact with the anti-CREB antibody, suggesting that
the complexes did not contain CREB. As with the CARE3/4 probe, WEHI7.2
extracts formed two additional protein/DNA complexes, which were absent
in the JEG-3 extracts (open arrows). The three complexes
formed by the WEHI extracts were competed with unlabeled CARE1/2
oligonucleotide (Fig. 5C), but only the fastest migrating complex
(denoted by the asterisk) was competed by the CARE3/4
oligonucleotide, suggesting that this protein complex can bind to both
CARE1/2 and CARE3/4. In contrast, the remaining two slower migrating
bands were unaffected by CARE3/4, suggesting that they are specific for
the CARE1/2 oligonucleotide and are not capable of binding to a typical
CRE. These data support the conclusion that the protein/CARE1/2 complex
that is seen in both WEHI and JEG extracts is not CREB, as CREB binding
the CARE1/2 would be competed by the CARE3/4 oligonucleotide.
Induction of c-fos Is Reduced after a Second Treatment
of Forskolin
The overexpression of ICER in transient expression studies blocks
the cAMP-induced expression of a number of cAMP-responsive promoter
constructs (8, 29, 30, 31). Moreover, ectopic expression of the CREM
transcriptional repressor in WEHI7.2 cells significantly inhibits the
responsiveness of a c-fos promoter to cAMP in agreement with
data reported earlier by Foulkes et al. (Ref. 32; and data
not shown). Although these experiments provide functional information
on CREM and ICER bioactivity, they do not address whether such an
inhibitory activity occurs in the context of native cellular levels of
endogenous ICER. We addressed this question by examining whether the
cAMP-induced expression of native ICER altered gene regulation after a
second stimulation of adenylate cyclase. Because activation of
adenylate cyclase by forskolin does not desensitize the enzyme with
respect to subsequent treatments (33, 34), we were
able to reactivate adenylate cyclase by removing the first treatment
and stimulating a second time with forskolin (see Materials and
Methods for details). Forskolin was added to WEHI7.2 cells for
3.5 h to achieve maximal levels of ICER expression. Then, cells
were harvested and resuspended in fresh media, and a second treatment
of forskolin was added. Gene induction in the presence of increased
ICER was then monitored (Fig. 6;
quantification in Table 1). As shown
previously, c-fos expression was induced approximately
3-fold at 0.5 h after the initial incubation with forskolin, and
the levels of c-fos dropped below basal levels by 3.5
h. In contrast, maximal levels of ICER message were observed at
3.5 h. At 3.5 h, cells were harvested and resuspended in
fresh media, and a second treatment with forskolin was immediately
initiated. After 0.5 h of additional incubation, c-fos
levels were only modestly induced and were not significantly higher
than basal levels. This observation could not be explained by a
reduction in CREB activation as the CREB protein was phosphorylated
after the second treatment to levels similar to the initial treatment
(Fig. 6 and Table 1). Thus, induction of endogenous
ICER severely impaired c-fos expression.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The ICER protein exists as multiple isoforms. RNase protection assays revealed that the ICER protein(s) induced in WEHI7.2 cells encoded DBD I. Isoforms that lacked DBD I and that used DBD II exclusively were not detected. However, the Northern blot analysis suggests the presence of multiple transcripts; therefore we cannot rule out the possibility that there are additional isoforms that we were unable to distinguish by these experiments.
Although ICER was induced in both WEHI7.2 and JEG-3 cells, there were
three differences in the expression and regulation of CREM/ICER between
the cell lines. First, 5'-sequences of CREM hybridized to multiple
transcripts in WEHI7.2 cells whereas only a single transcript was
observed for JEG-3 cells. These transcripts were not regulated by
forskolin and may represent cell-specific differences in the basal
expression of various CREM isoforms. The significance of this finding
remains to be elucidated. Second, the levels of the 2.3-kb transcript
increased more in the WEHI7.2 cells relative to the JEG-3 cells. The
2.3-kb transcript represented only a minor species in JEG-3 cells
treated with forskolin, whereas it was a major isoform in the WEHI
cells. Third, regulation in JEG-3 cells occurred through both CARE1/2
and CARE3/4. Moreover, CARE3/4 was sufficient in JEG-3 cells for
maximal induction equivalent to CARE1–4. In WEHI7.2 cells, CARE3/4 was
able to support cAMP-induced transcription whereas CARE1/2 was
nonfunctional in transient transfection assays. Although CARE3/4
conferred cAMP sensitivity to the tk promoter, it was not sufficient
for full activity. Other sequences appear to be required for maximal
induction. In this regard, CARE1/2 augmented the activity of CARE3/4 to
produce the maximal effect. One possible explanation for this
observation is that these sequences serve as binding sites for other
proteins that are required for maximal activation. Gel mobility shift
experiments using nuclear extracts demonstrate the presence of multiple
complexes associated with CARE1/2 that are unique to WEHI7.2. Although
CARE1/2 does not support cAMP-mediated gene expression in WEHI7.2
cells, it may recruit specific proteins to the area, which leads to the
synergistic effect on transcription observed with CARE1–4 in WEHI7.2
cells. The absence of these complexes in JEG-3 cells may explain the
lack of synergy of CARE1/2 and CARE3/4 in these cells. This would
suggest the presence of cell type-specific factors in thymocytes that
function in the regulation of ICER transcription in a manner different
from choriocarcinoma cells.
How CARE1/2 functions in cAMP-mediated transcription in JEG-3 cells is currently unclear. CREB from nuclear extracts did not bind to these elements, in agreement with our results demonstrating a low affinity of CREB for these sequences. However, a nuclear protein/DNA complex was formed that was common to both JEG-3 and WEHI extracts. This protein(s) may be responsible for the transcriptional activation observed. The absence of a response in WEHI7.2 cells may be due to the presence of other complexes not observed in JEG-3 extracts that may be inhibiting the response to cAMP. There are more than 12 different CREB/activating transcription factor (ATF) family members that bind to similar CRE-like sequences and may have a higher affinity for these sequences than does CREB. Thus the presence or absence of these may represent an additional level of transcriptional control and specificity. We are currently attempting to identify these factors to determine their potential role in the response to cAMP.
The kinetics of cAMP-mediated induction and attenuation of c-fos transcription in WEHI7.2 cells are typical of immediate early genes: rapid induction followed by transcriptional attenuation and down-regulation. Subsequent to the rapid induction of c-fos, we detected a reduction in mRNA levels to basal levels within 3 h, decreasing below basal levels within 6 h after forskolin treatment. It is proposed that transcriptional attenuation after cAMP induction of the somatostatin gene requires phosphatase activation and dephosphorylation of CREB (2, 3, 4). Although this mechanism is certainly of importance in the attenuation of expression of genes such as somatostatin, its role in the attenuation of c-Fos expression in WEHI7.2 cells is unclear. Significant CREB phosphorylation is detected after 3 h of forskolin treatment (1), a time when c-fos and junB mRNA levels have returned to basal. CREM transcript levels also are increased at 3.5 h after forskolin treatment, suggesting that CREB remains functional at this time point. Thus, it is unlikely that dephosphorylation of CREB plays a predominant role in the attenuation of c-fos expression in WEHI7.2 cells. It has been suggested that c-Fos may act to feedback and negatively regulate its own expression after stimulation with growth factors (35, 36, 37). This attenuation occurs through the palindromic sequence known as the dyad symmetry element, which is also the element that confers phorbol ester and serum responsiveness. Thus, the activation and repression of fos expression by phorbol ester and serum occur through the same DNA element. In comparison, the dyad symmetry element is not responsive to cAMP, whereas the CREs located in the promoter are responsive. Although it is possible that Fos may negatively regulate cAMP-induced transcription, it is unlikely, since these transcription factors act through separate response elements.
Another potential candidate for the transcriptional attenuation is ICER. It is suggested that ICER functions to attenuate the transcriptional response to cAMP (6, 8, 22, 29, 30, 31). In the WEHI7.2 cells, the induction of ICER correlates with the return of c-fos and junB expression to basal levels, suggesting that ICER may be involved in the attenuation phase of the response. Cotreatment of cells with forskolin and cycloheximide prevents the transcriptional attenuation of c-fos and junB and indicates that new protein synthesis is required for the rapid return of these oncogene transcripts to basal levels (reviewed in Ref.38). This is consistent with the idea that one component of the superinduction is the lack of ICER repression.
In addition to transcriptional attenuation, overexpression of ICER in
transient reporter assays prevents the induction of a variety of
cAMP-responsive genes (8, 29, 30, 31), and expression of the CREM , ß
and transcriptional repressors can inhibit cAMP induction of
c-fos in transient transfection assays (Ref. 32 and data not
shown). While these studies are paramount in understanding the role for
this family of transcriptional repressors, the levels of CREM/ICER in
the cells are superphysiological and thus do not represent levels of
protein actually occurring in the cell under normal physiological
conditions. Thus, we asked whether induced levels of native ICER were
sufficient to prevent the regulation of cAMP-induced gene expression at
time periods subsequent to transcriptional attenuation of the
c-fos gene. WEHI cells were treated with forskolin for
3.5 h, a time period required for induction of ICER and
attenuation of c-fos expression. Cells were then treated a
second time with forskolin and c-fos levels were measured.
Although c-fos levels increased, the relative levels
increased only to slightly above basal levels. The reduced induction of
c-fos could not be explained by the down-regulation of the
cAMP signaling machinery because the CREB protein was phosphorylated
after the second treatment to levels similar to the initial treatment.
These data indicate that: 1) adenylate cyclase was activated; 2) the
phosphatases responsible for attenuating the effects of CREB activation
did not prevent its reactivation; and 3) PKA was active; thus the
refractory period due to PKA down-regulation had not begun. Thus, other
factors appear to be responsible for preventing the relative levels of
the c-fos protooncogene from rising above the basal levels
upon subsequent activation.
In this regard, the production of transcriptional attenuators/repressors such as ICER may be required for early stages in the rapid down-regulation of cAMP-induced genes and the reduced responsiveness to subsequent activation. The dephosphorylation of CREB may prevent further activation of gene expression when cAMP levels remain elevated. At later time periods, down-regulation of the catalytic subunit of PKA functions to make the cell refractory to stimulation by the same agent. This concerted effort to reduce Fos in the cells after stimulation and keep levels low may be required to prevent oncogenesis. Deregulated expression of c-fos results in neoplastic transformation of rat fibroblasts (reviewed in Ref.38). Thus, ICER may function as an antioncogene by preventing protooncogenes such as c-fos from being expressed at high levels for an extended period of time, a condition supporting neoplasia.
In WEHI7.2 thymoma cells, the induction of ICER immediately precedes the detection of apoptosis-associated DNA cleavage. Thus, ICER may function in the early steps of apoptosis in cells that have an intact death cascade. The growth-promoting effects of protooncogenes might be dominant to death factors and block apoptosis. By down-regulating protooncogenes such as c-fos, ICER may be allowing for the death program to proceed. Moreover, ICER may function in part by inhibiting cell growth by down-regulating cell cycle genes. For example, cyclin A expression and cell cycle progression are induced by cAMP. ICER may function to return the level of cyclin A to its noninduced state and return the cells to the G0/G1 stage of the cell cycle (39). Clearly, the stable expression of ICER sense and antisense cDNA in WEHI7.2 cells will help to address the issue surrounding the possible involvement of ICER in both transcriptional attenuation and the apoptotic cascade, and such studies will be instrumental in enhancing our understanding of cAMP-induced gene expression leading to apoptosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cell Culture
WEHI7.2 thymoma cells were grown in DMEM containing 10% calf
bovine serum, 0.063 g/liter penicillin, 0.1 g/liter streptomycin. JEG-3
cells were grown in MEM containing 10% calf bovine serum, 0.063
g/liter penicillin, 0.1 g/liter streptomycin. All cells were grown at
37 C in an atmosphere of 6% CO2 and 90% humidity.
Northern Blot Analysis
Northern analysis was performed as described previously (40)
with the following modification: mRNA was purified using a PolyATract
mRNA isolation system (Promega, Madison, WI), according to
manufacturer’s instructions, and 3 µg mRNA were subjected to
electrophoresis. Quantification of Northern blots was performed using a
PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
The analysis of gene expression after two treatments with forskolin was performed as follows. Five aliquots of WEHI7.2 cells were pooled and one aliquot (control, 0 h) was removed for Northern and Western analyses. The remaining cells were treated with 10 µM forskolin and aliquots removed for analysis after 0.5 h (+forskolin, 0.5 h) and 3.5 h (+forskolin, 3.5 h). The final two samples were divided, and one sample continued to incubate for an additional 0.5 h (+forskolin, 4 h). The other sample was harvested, and the cells were resuspended in fresh media, 10 µM forskolin was added, and the cells were incubated 0.5 h (+forskolin, 3.5 h; +forskolin, 0.5 h). Individual pools were divided for the analysis of phospho-CREB by Western blot and for mRNA isolation and Northern blot analysis.
DNA Fragmentation
DNA fragmentation was assessed by the methods of Wyllie et
al. (26). Briefly, cells were treated with 10 µM
forskolin for the indicated time points, harvested by centrifugation at
900 g, and lysed on ice in an excess volume of hypotonic
buffer consisting of 5 mM Tris-HCl, pH 7.5, 5
mM EDTA, and 0.5% Triton X-100. The lysates were
centrifuged at 13,000 x g, and the supernatant
fraction containing soluble DNA fragments were extracted two times with
phenol-chloroform-isoamyl alcohol (25:24:1) and once with
chloroform-isoamyl alcohol (24:1). The DNA in the supernatants was
precipitated in 66% ethanol and 300 mM sodium acetate, pH
5.3. Precipitates were resuspended in 10 mM Tris-HCl, pH
8.0, 1 mM EDTA and incubated for 30 min at 25 C with 25
µg/ml RNase. Soluble DNA fragments from equivalent number of cells
were separated by electrophoresis on a 1.5% agarose gel and visualized
by ethidium bromide staining.
RNase Protection
RNA was analyzed according to manufacturer’s directions using
the RPA II Ribonuclease Protection Assay Kit (Ambion, Austin, TX). The
cDNAs corresponding to CREM probe I [nucleotides 554–653 of the CREM
cDNA sequence (7)] and probe II (nucleotides 841-1158) were
constructed in pBluescriptKS+ and 32P-labeled
probes were made using MAXIscript (Ambion). The mouse ß-actin
probe was supplied by Ambion and was used as a control for
normalization of RNA concentration. CREM and CREMß cDNAs were
cloned into pSG5 and unlabeled RNA made using MAXIscript. Briefly 2
µg total RNA were hybridized to a 32P-labeled RNA probe,
single-stranded RNA was digested, and the product was separated on a
standard urea gel. The gel was dried and products were visualized by
autoradiography.
Gel Mobility Shift Analysis
Gel mobility shift analysis used procedures and conditions
described previously (41). The CARE1/2 probe consists of the following
sense-strand sequence corresponding to positions -154 to -121 in the
ICER promoter (6): TTTCAGTGAGCTGCACATTGATGGCAGTGATAGG. The CARE3/4
probe consists of the following sense strand sequence corresponding to
positions -120 to -92: CTGGTGACGTCACTGTGATGTCAGTGCTC.
CREM-containing extracts were obtained by transfecting COS-7 cells
with pSG5-CREM (7) and preparing nuclear extracts by the method of
Shapiro et al. (42). Control COS-7 extracts were obtained in
a similar manner from cells transfected with the pSG5 plasmid lacking
the cDNA insert. Murine CREB cDNA was subcloned into the pVL1392
polyhedrin transfer plasmid for expression in a baculovirus system.
Recombinant baculovirus was isolated and plaque purified by standard
procedures (43). Sf-9 cells were infected for 48 h with CREB
recombinant baculovirus. Whole cell extracts were prepared by
sonication of infected cell pellets in 10 mM Tris-HCl (pH
7.8), 1 mM EDTA, 0.3 mM zinc acetate, 5
mM dithiothreitol, 0.3 M KCl, followed by
centrifugation at 200,000 x g for 30 min. WEHI7.2 and
JEG-3 nuclear extracts were prepared by the method of Shapiro et
al. (42). Supershift experiments were performed using an anti-CREB
antibody (New England Biolabs, Beverly, MA).
Construction of Reporter Plasmids
The ICER promoter from -378 to +1 from the translational start
site was isolated by PCR amplification of genomic DNA. DNA primers were
designed corresponding to position -378 to -365 and +1 to -13, and
standard PCR conditions were employed for amplification of genomic DNA
(44). The amplified sequence was subcloned into the ptkCAT plasmid (27)
in which the tk promoter was removed. CARE1/2tkCAT and CARE3/4tkCAT
were constructed using complimentary DNA oligonucleotides corresponding
to base pairs -154 to -121 and -120 to -92, respectively, and
subcloned into ptkCAT. The sequence of all constructs was confirmed
using a Sequenase kit (USB Corp., Cleveland, OH) according to
manufacturer’s recommendation.
Transient Transfection Assay
Logarithmically growing cells were harvested and resuspended in
DMEM, 0.5% calf bovine serum at a concentration of 3 x
106 cells/ml. Cells were transfected with the indicated
reporter plasmid (2 µg) for 5 h using Lipofectamine reagent
(Life Technologies, Gaithersburg, MD), according to manufacturer’s
recommendations. Then cells were diluted with DMEM, 10% calf bovine
serum to 6 x 105 cells/ml and treated with forskolin
or ethanol vehicle for 16 h. Cell extracts were prepared and
assayed for CAT activity as described (45).
Western Blot (Immunoblot) Analysis
Western blot analysis with antiphosphorylated CREB was performed
as previously described (1). Phosphorylated CREB was quantitated by
densitometric analysis of the resulting autoradiogram.
|
ACKNOWLEDGMENTS |
|---|
cDNA, Paolo Sassone-Corsi for the CREM and CREMß cDNA
vectors, and M. Greenberg for the antiphosphorylated CREB antibody. We
would like to thank Paul MacDonald for help with baculovirus-mediated
expression of CREB and for helpful discussions. Special thanks go to
Darcy Krueger for critical reading of the manuscript and to Monique
Heitmeier and Michelle Benoit for technical assistance. |
FOOTNOTES |
|---|
This work was supported by NIH Grant AI-35910 (to D.R.D.).
Received for publication September 9, 1997. Revision received December 12, 1997. Accepted for publication December 30, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit genes. Mol Endocrinol 5:243–255-subunit gene: dependence
on cyclic AMP-inducible elements. Mol Cell Biol 7:3994–4002This article has been cited by other articles:
 |
|
 |
|
  L. Shi, M. L. Ko, and G. Y.-P. Ko Rhythmic Expression of MicroRNA-26a Regulates the L-type Voltage-gated Calcium Channel {alpha}1C Subunit in Chicken Cone Photoreceptors J. Biol. Chem., September 18, 2009; 284(38): 25791 - 25803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Hu, I. V. Lund, M. C. Gravielle, D. H. Farb, A. R. Brooks-Kayal, and S. J. Russek Surface Expression of GABAA Receptors Is Transcriptionally Controlled by the Interplay of cAMP-response Element-binding Protein and Its Binding Partner Inducible cAMP Early Repressor J. Biol. Chem., April 4, 2008; 283(14): 9328 - 9340. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fink, J. Acimovic, T. Rezen, N. Tansek, and D. Rozman Cholesterogenic Lanosterol 14{alpha}-Demethylase (CYP51) Is an Immediate Early Response Gene Endocrinology, December 1, 2005; 146(12): 5321 - 5331. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Euskirchen, T. E. Royce, P. Bertone, R. Martone, J. L. Rinn, F. K. Nelson, F. Sayward, N. M. Luscombe, P. Miller, M. Gerstein, et al. CREB Binds to Multiple Loci on Human Chromosome 22 Mol. Cell. Biol., May 1, 2004; 24(9): 3804 - 3814. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Sussman ICER-Capades: Putting Cardiac Cyclic AMP Signaling "On Ice" Circ. Res., July 11, 2003; 93(1): 6 - 8. [Full Text] [PDF]
|
|
|
|
|
|
J. R. Mead, T. R. Hughes, S. A. Irvine, N. N. Singh, and D. P. Ramji Interferon-gamma Stimulates the Expression of the Inducible cAMP Early Repressor in Macrophages through the Activation of Casein Kinase 2. A POTENTIALLY NOVEL PATHWAY FOR INTERFERON-gamma -MEDIATED INHIBITION OF GENE TRANSCRIPTION J. Biol. Chem., May 9, 2003; 278(20): 17741 - 17751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Bodor, L. Feigenbaum, J. Bodorova, C. Bare, M. S. Reitz Jr, and R. E. Gress Suppression of T-cell responsiveness by inducible cAMP early repressor (ICER) J. Leukoc. Biol., June 1, 2001; 69(6): 1053 - 1059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Krueger, E. A. Warner, and D. R. Dowd Involvement of Thyrotroph Embryonic Factor in Calcium-mediated Regulation of Gene Expression J. Biol. Chem., May 5, 2000; 275(19): 14524 - 14531. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Zauli, P. Secchiero, L. Rodella, D. Gibellini, P. Mirandola, M. Mazzoni, D. Milani, D. R. Dowd, S. Capitani, and M. Vitale HIV-1 Tat-mediated Inhibition of the Tyrosine Hydroxylase Gene Expression in Dopaminergic Neuronal Cells J. Biol. Chem., February 11, 2000; 275(6): 4159 - 4165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ubeda, M. Vallejo, and J. F. Habener CHOP Enhancement of Gene Transcription by Interactions with Jun/Fos AP-1 Complex Proteins Mol. Cell. Biol., November 1, 1999; 19(11): 7589 - 7599. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Krueger, D. Mao, E. A. Warner, and D. R. Dowd Functional Analysis of the Mouse ICER (Inducible cAMP Early Repressor) Promoter: Evidence for a Protein That Blocks Calcium Responsiveness of the CAREs (cAMP Autoregulatory Elements) Mol. Endocrinol., July 1, 1999; 13(7): 1207 - 1217. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
2.
