| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
(RAR) and Their Phosphorylation Are Differentially Involved in Parietal Endodermal Differentiation of F9 Cells and Retinoid-Induced Expression of Target Genes
Institut de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP/Collège de France BP 163, 67404 ILLKIRCH Cedex CU de Strasbourg, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT RESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-null F9 cells
(RAR-/- cells), RA still efficiently
triggers RAR
-mediated primitive endodermal differentiation, but
parietal endodermal differentiation is markedly delayed. To investigate
the role of RAR1 activation functions AF-1 and AF-2 and of their
phosphorylation sites during RA- and cAMP-induced parietal
differentiation, cell lines reexpressing WT or mutated RAR1 were
established in RAR-/- cells. We have found
that the protein kinase A (PKA) phosphorylation site and the AF-2AD
core (helix 12) of RAR1 are required for efficient parietal
endodermal differentiation, whereas the AF-1 proline-directed kinase
phosphorylation site is dispensible. Interestingly, deletion of the
AF-1 activating domain (the A/B region), but not of the AF-2AD core,
generates a dominant negative mutant that abrogates primitive
endodermal differentiation when expressed in
RAR-/- cells. We also show that the RAR
AF-1 and AF-2 activation functions, but not their phosphorylation
sites, are involved in the induction of RA-responsive genes in a
differential promoter context-dependent manner. is the most potent biologically active form of
vitamin A. RA exhibits a wide range of activities and influences the
proliferation and differentiation of a variety of cell types. In that
context, F9 embryonal carcinoma (EC) cells represent a well established
cell-autonomous model system for investigating retinoid signaling
in vivo, as upon RA treatment and depending on culture
conditions, they differentiate into three distinct cell types
resembling primitive, parietal, and visceral endodermal extraembryonic
cells (1–4). RA- induced differentiation of F9 EC cells is
accompanied by an apoptotic response and a decrease in the rate of
proliferation, as well as by the induction of expression of a number of
genes (3, 5–7).
RA exerts its pleiotropic effects through two classes of nuclear
ligand-dependent transregulators: the retinoic acid receptors (RAR,
RARß, and RAR isotypes and their isoforms) activated by either
all-trans-RA or its 9-cis-isomer and the retinoid
X receptors (RXR, RXRß and RXR) activated by
9-cis-RA only (8–11). F9 cells express all RARs and RXRs
(12–14), and two strategies have been used to investigate their roles
in the response of F9 EC cells to RA treatment. Firstly, using
homologous recombination, we engineered F9 cells in which either the
RAR, the RAR, or the RXR gene, both the RAR and RXR
genes or both the RAR and RXR genes, are knocked out (6, 7, 15,
16). Secondly, wild-type (WT) and mutant F9 cells were treated with
pan-RXR- and RAR isotype (, ß, or )-selective retinoids (7,
17–20). It was established that RAR is indispensible for RA-induced
differentiation of F9 EC cells into primitive endoderm-like cells,
whereas RAR is additionally required for efficient parietal
endodermal differentiation that occurs in the presence of RA and cAMP
(20), and for which primitive endodermal differentiation is a
prerequisite. These studies also demonstrated that RAR/RXR heterodimers
are the functional units transducing the retinoid signal in
vivo (7, 19).
RARs and RXRs possess two transcriptional activation functions (AFs):
AF-1, located in the N-terminal A/B region, and AF-2, associated with
the ligand binding domain (LBD) (region E) and activated by agonistic
ligands (8, 21–24). The integrity of a conserved amphipatic -helix,
referred to as the AF-2AD core, is required for AF-2 activity (23,
25–28, and references therein). The AF-2AD core is located in the
C-terminal helix of the LBD (helix 12) that is indispensible for
the formation of the coactivator-binding surface generated during the
LBD transconformation triggered by ligand binding (8, 29–31).
In addition, RARs and RXRs are phosphorylated in their AF-1 domain that
contains sites for proline- directed kinases (20, 32, 33). In
RAR1, the phospho-rylated residue in the AF-1domain has been
identified as serine 77 (32). Furthermore, RAR1 can be
phosphorylated by PKA at serine 369 that is located in the LBD/AF-2
domain (34). Interestingly, RAR1 is phosphorylated at these two
residues in F9 cells, as well as in transfected COS cells (20, 34).
Similar phospho-rylation sites are present in other RARs and have
been shown to be phosphorylated in RAR2, which is the major RAR
isoform in F9 cells (34, 35). The role of RAR phosphorylation in
retinoid-induced events has been studied in
RAR-/- F9 cells, by establishing rescue cell
lines reexpressing either RARWT or RAR mutated at its
phosphorylation sites (20). The RAR AF-1 domain and the
proline-directed phosphorylation sites located in this domain were
found to be required for rescuing the differentiation of F9 cells into
primitive endoderm-like cells, whereas the PKA phosphorylation site was
dispensible (20). As we had previously established that overexpression
of RAR in RAR-/- cells could also restore
their primitive and parietal endodermal differentiation (14), the role
of RAR phosphorylation was similarly studied using overexpressed
RAR mutants. We concluded that both the RAR AF-1 and AF-2
phosphorylation sites were not required for allowing overexpressed
RAR to functionally replace RAR for primitive endodermal
differentiation, but they were apparently mandatory for parietal
differentiation (20).
Thus, RAR and its phosphorylation sites could be selectively
required for parietal endodermal differentiation. However, an
unequivocal demonstration of this RAR-selective function requires to
establish rescue cell lines reexpressing RAR (either WT or mutated)
in RAR-/- F9 EC cells that can still
differentiate into primitive endodermal cells (16) but whose parietal
endodermal differentiation is greatly delayed (20), thus providing a
model for directly analyzing the contribution of RAR AF-1 and AF-2
and of their phospho-rylation sites in parietal endodermal
differentiation. Furthermore, such lines also offer the possibility to
study which RAR mutations may generate dominant negatives,
preventing endogenous RAR from mediating primitive endodermal
differentiation. Rescue lines were therefore established with RAR
mutated at either one of its phosphorylation sites, as well as with
AF-1 or AF-2AD core deletion mutants. We demonstrate here that the PKA
phosphorylation site and the AF-2AD core (helix 12 of the LBD) of
RAR are required for parietal endodermal differentiation, whereas
the AF-1 proline-directed kinase phosphorylation site is dispensible.
None of the corresponding mutants behaved as dominant negatives,
preventing primitive endodermal differentiation, whereas deletion of
the AF-1 activating domain (the A/B region) generated such a dominant
negative mutant. The present study also shows that RAR AF-1 and AF-2
are differentially implicated in the induction of RA-responsive genes,
whereas their phosphorylation sites are not involved.
|
RESULTS |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Lacking Either
AF-1 or the AF-2AD Core or Mutated at Either the AF-1 or the AF-2
Phosphorylation Sites AF-1 and AF-2 are involved in
parietal endodermal differentiation, rescue lines stably expressing
wild-type RAR (RARWT line) and RAR lacking the A/B region
(RAR
AB line) or the AF-2AD core (RARAF2 line) were derived
from RAR-/- cells (see Fig. 1A
). Similarly, to investigate the role
played in this differentiation by phosphorylation of the AF-1 domain at
the proline-directed kinases sites (serines 74 and 77) and of the AF-2
domain at the PKA site (serine 369), rescue lines stably expressing
RAR mutated at these phosphorylation sites (RARS74/77A and
RARS369A lines, respectively, see Fig. 1A), were also
established.
|
WT
and of its mutant derivatives was determined after immunoprecipitation
and Western blotting, and compared with the expression of endogenous
RAR in WT F9 cells (Fig. 1B
). RARAB and RARS369A were
expressed in the selected rescue lines at levels similar to that of
RAR in WT F9 cells (Fig. 1B, lanes 1, 6 and 8). The RARWT and
RARAF2 rescue lines overexpressed the RAR protein relative to
endogenous RAR (Fig. 1B, lanes 3 and 7). Two RARS74/77A rescue
cell lines were selected that expressed the RAR protein at a level
either similar (clone 3) or higher (clone 14) than that of RAR in WT
F9 cells (Fig. 1B; lanes 4 and 5).
Role of RAR AF-1 and AF-2 and of Their Phosphorylation Sites in
F9 Cell Primitive and Parietal Endodermal Differentiation
When grown as monolayers in the presence of
all-trans-RA alone, WT F9 cells differentiate at 96 h
into primitive endoderm-like cells exhibiting a characteristic flat
triangular morphology (4, 20). Addition of cAMP along with RA, results
in the formation of parietal endoderm-like cells that have a rounded
and refractile appearance (1, 20).
WT F9 cells also differentiated into primitive endoderm-like cells when
treated for 96 h with a combination of a RAR agonist (BMS961)
and a pan-RXR agonist (BMS649) (36) (Fig. 2b). Parietal endodermal differentiation
occurred upon subsequent addition at 48 h of the RAR agonist
BMS753 (36), the pan-RXR agonist BMS649, and cAMP (Fig. 2c). No
parietal differentiation occurred at 96 h (or even later at
120 h) in the absence of BMS753, confirming that RAR is
required for parietal endodermal differentiation of WT F9 cells (Ref.
20 and data not shown).
|
-/-
cells (Fig. 2e) (16), but no parietal endodermal differentiation could
be seen at 96 h (Fig. 2f) (20). In fact, parietal differentiation
was markedly delayed and did not occur before 120 h (Ref. 20 , and
data not shown). Similar results were obtained with the RAR agonist
and cAMP, in the absence of the RAR agonist (data not shown). Thus,
in RAR-/- cells, RAR can substitute at
least to some extent for RAR, leading to a delayed parietal
endodermal differentiation.
Reexpression of RARWT (RARWT cell line) restored parietal
endoderm differentiation at 96 h (Fig. 2i). The two RARS74/77A
lines also differentiated efficiently into parietal endoderm-like cells
(Fig. 2l, and data not shown), indicating that RAR can efficiently
mediate this differentiation in the absence of AF-1 phosphorylation. In
contrast, the RARS369A rescue line did not differentiate into
parietal endoderm at 96 h (Fig. 2o). As in the case of the
parental RAR-/- cell line, parietal
endodermal differentiation was delayed and occurred only at 120 h
(data not shown). The same observation was made with the RARAF2
rescue line (Fig. 2u). Interestingly, the RARAB rescue line
retained a stem cell morphology in response to addition of BMS961 and
BMS649. Only a few differentiated cells were observed upon subsequent
addition of BMS753, BMS649, and cAMP (Fig. 2, p–r), thus suggesting
that deletion of the A/B region generates a RAR dominant negative
mutant that prevents endogenous RAR from mediating primitive
endodermal differentiation and subsequently suppresses parietal
endodermal differentiation.
The extent of differentiation of the various rescue lines was also
assessed by the expression of transcripts of two markers of primitive
endodermal differentiation, collagen IV (1), and laminin B1 (4, 37),
and of one marker specific for parietal endodermal differentiation,
thrombomodulin (2, 38). Semiquantitative RT-PCR analysis showed that
the induction of collagen IV and laminin B1, which was not affected in
RAR-/- cells (Fig. 3A, lanes 4–6), was similar in all
rescue cell lines, except the RARAB line where it was
significantly reduced (Fig. 3A, lanes 16–18), in agreement with the
absence of obvious morphological differentiation (see above).
|
) and BMS649 (selective for all RXRs) either
in the absence or presence of cAMP (Fig. 3B, lanes 2–4), thus
corroborating previous reports (2, 39). However, thrombomodulin was
induced upon subsequent addition of BMS753 (selective for RAR),
BMS649, and cAMP (Fig. 3B, lane 5), indicating that its expression is
indeed specific for parietal endodermal differentiation. Note that
there was no induction by BMS753 in the absence of the RAR agonist
(data not shown). The induction of thrombomodulin was only mildly
reduced in the RAR-/- cell line, in
agreement with the delayed morphological parietal endodermal
differentiation (Fig. 3B, lanes 8 and 9 and Fig. 3C, lane 4), and also
confirming that RAR can be, to some extent, functionally
replaced by RAR (20). The induction of thrombomodulin was restored
in the RARWT and RARS74/77A lines (Fig. 3C, lanes 6 and 8).
Interestingly, in the RARS369A and RARAF-2 lines, the
induction of this gene was mildly reduced, as in the case of the
RAR-/- line (Fig. 3C, lanes 12 and 14) and
in agreement with the delayed parietal endodermal differentiation. In
contrast, the induction of thrombomodulin was completely abrogated in
the RARAB rescue line (Fig. 3C, lane 10), thus corroborating the
lack of differentiation.
Effect of RAR AF-1 and AF-2 Activating Domains and of Their
Phosphorylation Sites on Expression of RA-Responsive Genes
Whereas knockout of the RAR gene in F9 cells results in a
marked reduction of the expression of several RA-responsive genes such
as Hoxa-1, HNF3, HNF1ß, Stra6, and Stra8 (19, 20), knockout of
RAR had no effect on their expression (Fig. 4, compare lanes 2 and 4; Fig. 5, compare lanes 6 and 7 and Fig. 6, lane 2). In fact, only Hoxb-1 and
CRABP II gene expression was significantly reduced in
RAR-/- cells (Fig. 4, compare lanes 2 and 4;
Fig. 5, compare lanes 6 and 7 and Fig. 6, lane 2) (16, 19). The
expression of these RA-responsive genes was investigated in the various
rescue cell lines using semiquantitative RT-PCR (for each gene, in the
linear range of the PCR amplification reaction) after treatment for
24 h with 100 nM RA or a synthetic retinoid selective
for RAR (BMS961) at 100 nM. The results that are
presented in Figs. 4 and 5 are summarized in Fig. 6.
|
|
|
WT (Fig. 6, lane 3). The phosphorylation site
mutants for proline-directed kinases and PKA (RARS74/77A and
RARS369A, respectively) were as efficient as RARWT (Fig. 6, lanes
4 and 5). RAR deleted for either the AF-2AD core or the A/B region
also rescued the RA-induced expression of Hoxb-1, but not that of
CRABPII (Fig. 6, lanes 6 and 7).
We previously concluded that the retinoid-induced expression of Hoxb-1
and CRABPII could be mediated by both RAR-RXR and RAR-RXR
heterodimers (19). However, the expression of both genes, in WT F9
cells, was less efficiently induced by a retinoid selective for RAR
(BMS961) than by RA (Fig. 5, compare lanes 6 and 11 and Fig. 6, compare
lanes 1 and 8). Interestingly, this latter retinoid was more efficient
in RAR-/- cells (Fig. 6, compare lanes 8 and
9), indicating that the presence of RAR was to some extent
preventing RAR to mediate the induction of Hoxb-1 (19), and CRABPII.
Note the down-effect of reexpressing RARWT (Fig. 6, lane 10), which
was not observed with the two deletion mutants (Fig. 6, lanes 11 and
12).
Genes whose expression was not affected in
RAR-/- cells, e.g. Hoxa-1,
HNF3, HNF1ß, Stra6, and Stra8, were also analyzed in the rescue
lines. Their induction was not significantly affected in RARWT,
RARS74/77A, and RARS369A rescue lines (Fig. 6, lanes 3–5). Note
that HNF3 expression was mildly reduced in the RARS74/77A line
corresponding to clone 14 (Fig. 4, lane 10), which expressed higher
RAR levels than the WT rescue line (Fig. 1B, lane 5). The
RARAB and RARAF2 deletion mutants did not affect the
RA-induced expression of Hoxa-1 but inhibited markedly those of HNF3
and HNF1ß (Fig. 6, lanes 6 and 7). On the other hand, only
RARAF2 mildly decreased the RA-inducibility of Stra6 and Stra8
(Fig. 6, lanes 6 and 7).
The RAR selective agonist (BMS961) was almost as efficient as
RA in inducing the expression of Hoxa-1, HNF3 and Stra8 in WT and
RAR-/- cells (Fig. 5, compare lanes 6 and
11; Fig. 6, compare lanes 1 and 8). Whereas Hoxa-1 expression was
unaffected by reexpressed RARAB and RARAF2, that of HNF3
was reduced and that of Stra8 was reduced only by RARAF2 (Fig. 6, lanes 11 and 12). Note that the basal level of expression of Stra8 was
also decreased by this mutant (Fig. 5, lane 5). In contrast, the
RAR-selective retinoid was more efficient in
RAR-/- cells than in WT F9 cells for
inducing not only the expression of Stra6 (19), but also that of
HNF1ß (Fig. 6, lanes 8 and 9), confirming that maximal induction of
some genes by RAR may be competed for by RAR (see above). Note,
in that respect, the effect of reexpressing RARWT (Fig. 6, lane 10).
Note also that in some experiments, the basal levels of HNF1ß
expression were mildly up-regulated in
RAR-/- cells (Fig. 5, lane 2). RARAF2
was as efficient as RARWT in preventing maximal induction of Stra6
by RAR but completely abrogated the activation of HNF1ß by BMS961
(Fig. 6, lane 12). RARAB also abrogated the activation of HNF1ß
but did not inhibit the activation of Stra6 by the RAR agonist (Fig. 6, lane 11).
We also analyzed the expression of another RA-inducible gene,
CYP26, whose expression has been shown to be mediated by RAR/RXR
heterodimers (40). Accordingly, it was only slightly decreased in
RAR-/- cells (Fig. 7, compare lanes 1–4). The induction of
CYP26, which was maximal at 96 h, was not affected in the various
rescue lines with the exception of the RARAB line where it was
markedly reduced (Fig. 7, lane 10).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
to F9 cell parietal endodermal
differentiation and the activation of RA-responsive genes, and also to
examine whether the activity of RAR could be modulated by
phosphorylation of these activating domains. To that end, RAR1
mutants lacking either the AF-1 domain or the AF-2AD core (helix 12),
or bearing mutations in either the AF-1 or AF-2 phosphorylation sites,
were reexpressed to WT levels in RAR-/-
cells. We demonstrate that both the AF-1 and AF-2 activating domains of
RAR contribute to parietal endodermal differentiation of F9
cells and to the expression of RA-responsive genes. We also demonstrate
that, whereas the phosphorylation sites located in the AF-1 activating
domain are not involved in these processes, the PKA phosphorylation
site located in the LBD is required for parietal endodermal
differentiation. Indeed, mutation of the PKA site abrogates parietal
endodermal differentiation at 96 h even in the presence of
cAMP.
The RAR AF-2 Activating Domain and Its PKA Phosphorylation Site
Are Required for Efficient Parietal Endodermal Differentiation
The AF-2AD core motif of RARs belongs to the C-terminal
-helix of nuclear receptor ligand binding domains (helix 12), which
is indispensible for the formation of the coactivator-binding surface
during the LBD transconformation triggered by ligand binding, thus
generating transcriptionally active receptors (8, 31). Numerous studies
have demonstrated that the integrity of the AF-2AD core is required for
efficient ligand-dependent interactions with coactivators in
vitro (for reviews, see Refs. 8, 41, 42). In the present
study, we demonstrate that the AF-2AD core of RAR and therefore its
AF-2 activation function, is required to mediate the effects of RA and
cAMP on F9 cells differentiation in vivo, as RAR deleted
for this domain is unable to restore parietal endodermal
differentiation. Note that this AF-2 activation function was also found
to be required for RAR and overexpressed RAR to restore
primitive endodermal differentiation in
RAR-/- cells (see Table 1).
|
S369A mutation) also abrogates the ability of RAR to
efficiently restore parietal endodermal differentiation, demonstrating
that this cAMP-dependent phosphorylation of RAR is additionally
required for this differentiation process. This is in accordance with
our previous report (20) showing that overexpression of RARS369A in
RAR-/- cells was also unable to restore
parietal endodermal differentiation (see Table 1), which suggested that
this mutant exerts a dominant negative effect on endogenous RAR,
preventing parietal differentiation. However, it is clear from our
present results that RARS369A has no dominant negative (dn) effect
on RAR, as the RARS369A line normally differentiates into
primitive endoderm-like cells.
Collectively these results demonstrate that the contribution of RAR
to parietal endodermal differentiation of F9 cells requires the
integrity of the AF-2 surface interacting with coactivators, as well as
the integrity of the PKA phosphorylation site located in the AF-2
domain. The effect of this cAMP-dependent RAR LBD phosphorylation is
unknown, but it may modulate the efficiency of ligand-dependent
coactivator/corepressor binding to RAR, and/or its dimerization with
RXR and thus its DNA binding properties to RA-response elements (34).
Interestingly, PKA-mediated phosphorylation of certain coactivators may
also modulate their interaction with RAR/RXR heterodimers (Refs. 41, 42, 43, 44
and references therein).
Phosphorylation of the AF-1 Activating Domain Is Not Required for
Parietal Endodermal Differentiation, but Deletion of this Domain
Generates a Dominant Negative Mutant for F9 Cell Differentiation
When reexpressed in RAR-/- cells,
RAR mutated at its proline-directed kinase phosphorylation site in
the N-terminal AF-1 domain (RARS74/77A) is as efficient as RARWT
at restoring parietal endodermal differentiation, indicating that
phosphorylation of the RAR AF-1 domain is not required for this F9
cells differentiation. In contrast, we concluded in our previous study
(see Table 1) that although the RAR AF-1 phosphorylation site was
apparently not required to allow overexpressed RAR to restore, at
the morphological level, the primitive endodermal differentiation of
RAR-/- cells, it was indispensible for
subsequent parietal endodermal differentiation. However, the expression
of several RA-responsive genes was not restored to WT levels in
RAR-/- cells rescued with RARS74/77A (see
Table 1). It is therefore possible that the phosphorylation of RAR
AF-1 is in fact indispensible in the RAR-/-
rescued cells to efficiently rescue primitive endodermal
differentiation at the molecular level, which is a prerequisite for
subsequent parietal differentiation, whereas this phosphorylation is
not required for parietal differentiation per se.
Expression of RAR lacking the A/B region in
RAR-/- cells (RARAB rescue line)
abrogates their differentiation into primitive endoderm-like cells.
This must reflect a ligand-independent (see Fig. 2) dn effect of
RARAB on the endogenous RAR, which is required for primitive
endodermal differentiation. This dn effect of RARAB was also
exerted on the expression of certain RA-responsive genes (see below and
Fig. 6). Note in this respect that we have previously shown (Table 1)
that RARAB overexpressed in RAR-/-
cells could not restore RA-induced primitive endodermal differentiation
and expression of most RA-induced genes, and thus could not
functionally replace RAR.
We conclude that the RAR AF-1 proline-directed phosphorylation
site is dispensible for parietal endodermal differentiation of F9
cells, whereas deletion of the A/B region (see Fig. 2, RARAB),
but not of the AF-2AD core (RARAF2), results in a dominant
negative receptor that blocks the RAR-mediated primitive endodermal
differentiation.
Differential Promoter Context-Dependent Effect of the AF-1 and AF-2
Activating Domains of RAR and of Their Phosphorylation Sites on
Expression of RA-Responsive Genes
Among the F9 cell RA-responsive genes that we have studied
here and previously (16, 19), the expression of only two of them
(Hoxb-1 and CRABPII) is reduced in RAR-/-
cells, whereas that of the others is not affected. In the present
study, we have investigated the role played by RAR AF-1 and AF-2
activating domains and their phosphorylation sites in the expression of
these RA-responsive genes. Our results, summarized in Fig. 6 and in
Table 1, lead to the following conclusions.
Firstly, the PKA site located in the AF-2 domain and the
proline-directed kinase site located in the AF-1 domain are not
required for restoring the RA-induced expression of Hoxb-1 and CRABPII
genes, which is decreased in RAR-/- cells.
This is in contrast with our previous observation showing that in
RAR-/- cells overexpressing RAR, the
phosphorylation of RAR1 AF-1 (RARS74/77A) is required for
RA-induced expression of some other RA-responsive genes (see Table 1).
Note in this respect that phosphorylation of RAR AF-1
(RARS66/68A) is also required for RA-induced expression of a subset
of responsive genes (see Table 1).
Secondly, the AF-1 domain and the AF-2AD core of RAR are
differentially required for RA-induced expression of target genes.
Indeed, RAR deleted for either of these two domains restores the
induction of Hoxb-1, but not of CRABPII. Note that a similar
promoter-context dependent requirement of the AF-1 activating domains
of RAR and RAR was also reported for RA-induced expression of
other responsive genes in RAR-/- rescue
cells (see Table 1).
Thirdly, both AF-1 and AF-2 deletion mutants of RAR behave as
dominant negatives in a promoter context-dependent manner, as the
induction of RA-responsive genes whose expression is not altered in
RAR-/- cells (Hoxa-1, HNF3, HNF1ß,
Stra6, Stra8, and CYP26) was differentially decreased upon reexpression
of RAR mutants deleted for either the A/B region or the AF-2AD core
(see RARAB and RARAF2 in Fig. 6). Note that promoter
context-dependent dn effects have also been observed with
reexpressed RARAF2 and overexpressed RARAF2 in
RAR-/- cells (see Table 1 and Ref. 46), in
keeping with the previous suggestion that AF-2 deletion mutants may
behave as dominant negatives for the transcription of RA-responsive
genes (23, 45).
Fourthly, depending also on the promoter context, the presence of
RAR differentially hinders the potential ability of RAR to
mediate RA-induced expression of a number of responsive genes. Indeed,
in several cases, the RAR-selective agonist BMS961 was found to be a
more efficient inducing ligand in RAR-/-
cells than in F9 WT cells (Fig. 6). These observations confirm that
functional redundancies between RARs can be artefactually generated
upon gene knockout (18, 20), and furthermore clearly show that the
extent of these redundancies is variable, being dependent on the
promoter context of the RA-responsive genes.
Conclusion
Using the cell autonomous RA-responsive F9 EC cell
differentiation system, we have demonstrated in our present and
previous studies (see Table 1) that the AF-1 and AF-2 activation
functions of both RAR and RAR, as well as their proline-directed
kinase and PKA phosphorylated forms, play important and distinct roles
in primitive and parietal endodermal differentiation and RA-induced
expression of responsive genes. Moreover, our data show that RAR and
RAR AF-1 and AF-2, and their phosphorylated forms exhibit distinct
functions that are differentially required, in a promoter
context-dependent manner, to elicit the induction of RA-responsive
genes expression. Interestingly, the dominant negative effect of some
AF-1 and AF-2 deletion mutants on RAR- and RAR-mediated induction
of expression of RA-responsive genes also exhibits a promoter-context
dependency. Finally, a similar promoter-context dependency is also
characteristic of the artefactual functional redundancy generated
between RAR and RAR upon RAR knockout in F9 cells. All of
these promoter-context dependencies certainly reflect the diversity of
the responsive complexes that can be combinatorially assembled on
polymorphic control regions with different receptor isoforms that are
differentially phosphorylated, interact with different coactivators,
and synergize with a variety of transregulators (41, 42).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 (RARWT), A/B region-truncated RAR
(RARAB), AF-2AD core-deleted RAR (RARAF2), or RAR
harboring mutations in phosphorylation sites were cloned into the
pD402A vector (gift of D. Lohnes, Institut de Recherches Cliniques de
Montréal, Montreal, Canada) which is driven by the PGK
promoter (47). The expression construct pD403A containing mRAR1 cDNA
cloned in pD402A, has been described (14). RARAB, RARS74/77A,
and RARS369A were isolated as EcoRI fragments from the
corresponding pSG5 constructs (22, 32, 34) and subcloned into the same
site of pD402A. RAR deleted for the AF-2AD core (amino acids
408–416) was constructed in pD402A, by replacing the WT
SacI/BamHI fragment in pD403A by the
corresponding fragment containing the deletion in the described pSG5
construct (23).
Cell Culture and Establishment of Stable Rescue Lines
Fig 9 cells were cultured as monolayers on gelatinized surfaces
as described (15). For differentiation studies,
105 cells were cultured in a 10-cm dish, and
treated with RA (100 nM) alone or in combination with cAMP
(250 µM) for 96 h, with a change of media after
48 h. Cells were also treated with a combination of a RAR
agonist (BMS961, 100 nM) and a pan-RXR agonist (BMS649, 1
µM), followed by the subsequent addition at 48 h of
a RAR agonist (BMS753, 100 nM), the pan-RXR agonist and
cAMP. Control cells were treated with vehicle alone (0.1% ethanol
final concentration). To establish rescue lines,
RAR-/- cells were electroporated with each
of the constructs displayed in Fig. 1A along with a plasmid conferring
resistance to puromycin (pD503; gift of D. Lohnes), in a ratio of 10:1.
After 24–36 h, cells were selected with 0.8 µg/ml of puromycin for
10 days as described (14) and analyzed for the presence and expression
of the transgene by Southern and Western blotting.
Cell Extracts, Immunoprecipitation and Immunoblotting of
RAR
Whole cell extracts were prepared (48) from rescue lines grown
as monolayers in the absence of RA, and RAR was immunoprecipitated
with Protein A-Sepharose cross-linked with monoclonal antibodies
directed to the F region of RAR [(Ab9(F)] (49). Proteins were
resolved by SDS-PAGE (10% acrylamide), electrotransferred onto
nitrocellulose filters and immunoprobed with polyclonal antibodies
against the F region of RAR [RP(F)] (49), followed by
peroxidase-labeled protein A and chemiluminescence detection, according
to manufacturer’s protocol (Amersham Pharmacia Biotech Europe GmbH, Saclay, France).
RNA Isolation and RT-PCR
RNA was isolated using the guanidinium thiocyanate method and
conditions for semiquantitative RT-PCR were as described (5). The
quantity of RNA used for RT-PCR in each reaction (2 µg) was
normalized with 36B4 transcripts (5, 50) which are unresponsive to
retinoid treatment. The RT-PCR oligonucleotides used for Hoxa-1,
HNF3, HNF1ß, Stra8, laminin B1, collagen type IV (1), Hoxb-1,
CRABPII, and 36B4 were as described (6, 20). The RT-PCR primers for
throm-bomodulin (51) were as follows: 5'-TGGAGCATGAGTGCTTCGC-3' and
5'-GGTGTTGTAGGTACTAGAGA-3'. The primers for CYP26 (52) were
5'-GCTCAAGCTCTGGGACCTGT-3' and 5'-CGATCACGAGCACGTAGCAC-3'. The
amplifications were done for 13 cycles (collagen IV), 15 cycles
(laminin B1, Stra8, and 36B4), 18 cycles (Hoxb-1, Hoxa-1, and HNF3),
20 cycles (thrombomodulin, CRABPII, Stra6, and CYP26) or 24 cycles
(HNF1ß). Aliquots were electrophoresed on a 1.5% agarose gel and
samples were transferred onto Hybond N membranes. The RT-PCR blots were
probed with cognate [32P]-labeled cDNA
fragments or with end-labeled oligonucleotide probes for thrombomodulin
and CYP26. The signals were quantified using a Bio-Imaging Analyzer
(BAS 2000, Fuji Photo Film Co., Ltd., Tokyo, Japan).
|
ACKNOWLEDGMENTS |
|---|
1) cDNAs and to P. Bouillet and D.
Lohnes for the gift of several plasmids and probes. We thank D.
Metzger for critically reading the manuscript. We also thank S. Vicaire
for DNA sequencing, the cell culture group for maintaining and
providing cells, the staff of oligonucleotide synthesis, and C.
Werlé, S. Metz, B. Boulay, and J. M. Lafontaine for
preparing the figures. |
FOOTNOTES |
|---|
This work was supported by funds from the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Santé et de la Recherche Médicale (INSERM), the Collège de France, the Hôpital Universitaire de Strasbourg, the Association pour la Recherche sur le Cancer (ARC), the Comité Départemental du Haut-Rhin de la Ligue Nationale Contre le Cancer, and Bristol-Myers Squibb.
1 Present address: Department of Biochemistry and Molecular Biology,
Box 1126, Mount Sinai School of Medicine, New York, New York
10029-6574. 
Received for publication March 27, 2000. Revision received June 2, 2000. Accepted for publication June 8, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-null F9 embryonal carcinoma cells are resistant to the
differentiation, anti-proliferative and apoptotic effects of retinoids.
EMBO J 15:4142–4155[Medline]
and ß retinoic acid receptors and a novel
receptor predominantly expressed in skin. Nature 339:714–717[CrossRef][Medline]
or RAR ß in RAR gamma-null F9 cells reveals
a partial functional redundancy between the three RAR types. Proc Natl
Acad Sci USA 92:7854–7858 function in F9 cells by gene disruption
results in aberrant Hoxa-1 expression and differentiation upon retinoic
acid treatment. Proc Natl Acad Sci USA 90:9601–9605 (RAR) and
RAR results in receptor-specific alterations in retinoic
acid-mediated differentiation and retinoic acid metabolism. Mol Cell
Biol 15:843–851[Abstract]
)-,
RARß-, or RAR-selective ligand in combination with a retinoid X
receptor-specific ligand. Mol Cell Biol 15:6481–6487[Abstract]
and RAR is indispensable for differentiation of F9 cells
upon retinoic acid and cAMP treatment. EMBO J 16:6452–6465[CrossRef][Medline]
activation function AF-1 through binding to the
general transcription factor TFIIH and phosphorylation by CDK7. Cell 90:97–107[CrossRef][Medline]
(RXR) by
activated c-Jun N-terminal kinases (JNKs). J Biol Chem 274:18932–18941 by protein kinase A. Mol Endocrinol 9:860–871 (RAR), phosphorylates its AF-1 activating domain through
cdk7. J Biol Chem 275:21896–21904 and retinoid X receptor .
J Biol Chem 273:2409–2415 inhibits retinoic acid-induced P19 cell
differentiation by binding to DNA. Exp Cell Res 225:35–43[CrossRef][Medline]
-helical core motif of RAR, RAR AF-2
activating domains is indispensable for RA- induced differentiation
of F9 cells. J Cell Sci, in press
: evidence for phosphorylation. Exp Cell Res 201:335–346[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  Y.-H. Han, H. Zhou, J.-H. Kim, T.-d. Yan, K.-H. Lee, H. Wu, F. Lin, N. Lu, J. Liu, J.-z. Zeng, et al. A Unique Cytoplasmic Localization of Retinoic Acid Receptor-{gamma} and Its Regulations J. Biol. Chem., July 3, 2009; 284(27): 18503 - 18514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. M. Huq, N.-P. Tsai, S. A. Khan, and L.-N. Wei Lysine Trimethylation of Retinoic Acid Receptor-{alpha}: A Novel Means To Regulate Receptor Function Mol. Cell. Proteomics, April 1, 2007; 6(4): 677 - 688. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Parrella, M. Gianni, M. Fratelli, M. M. Barzago, I. Raska Jr, L. Diomede, M. Kurosaki, C. Pisano, P. Carminati, L. Merlini, et al. Antitumor Activity of the Retinoid-Related Molecules (E)-3-(4'-Hydroxy-3'-adamantylbiphenyl-4-yl)acrylic Acid (ST1926) and 6-[3-(1-Adamantyl)-4-hydroxyphenyl]-2-naphthalene Carboxylic Acid (CD437) in F9 Teratocarcinoma: Role of Retinoic Acid Receptor {gamma} and Retinoid-Independent Pathways Mol. Pharmacol., September 1, 2006; 70(3): 909 - 924. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Futaki, Y. Hayashi, T. Emoto, C. N. Weber, and K. Sekiguchi Sox7 Plays Crucial Roles in Parietal Endoderm Differentiation in F9 Embryonal Carcinoma Cells through Regulating Gata-4 and Gata-6 Expression Mol. Cell. Biol., December 1, 2004; 24(23): 10492 - 10503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Hu, L. Cherbas, and P. Cherbas Transcription Activation by the Ecdysone Receptor (EcR/USP): Identification of Activation Functions Mol. Endocrinol., April 1, 2003; 17(4): 716 - 731. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Cherbas, X. Hu, I. Zhimulev, E. Belyaeva, and P. Cherbas EcR isoforms in Drosophila: testing tissue-specific requirements by targeted blockade and rescue Development, March 2, 2003; 130(2): 271 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bastien, S. Adam-Stitah, J.-L. Plassat, P. Chambon, and C. Rochette-Egly The Phosphorylation Site Located in the A Region of Retinoic X Receptor alpha Is Required for the Antiproliferative Effect of Retinoic Acid (RA) and the Activation of RA Target Genes in F9 Cells J. Biol. Chem., August 2, 2002; 277(32): 28683 - 28689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Boskovic, D. Desai, and R. M. Niles Regulation of Retinoic Acid Receptor alpha by Protein Kinase C in B16 Mouse Melanoma Cells J. Biol. Chem., July 12, 2002; 277(29): 26113 - 26119. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
