| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Gene
European Molecular Biology Laboratory (C.G., G.F., V.S-B.,
F.G.) D-69117, Heidelberg, Germany
National Diagnostic
Centre (C.G.) University College Galway, Ireland
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
protein (ER-) in chicken tissues is demonstrated: the
previously reported receptor (cER- form I), which has a size of 66
kDa, and a new form (cER- form II), which lacks the N-terminal 41
amino acids present in form I and thus gives rise to a protein of 61
kDa. Whereas the 66-kDa protein is the translation product of several
cER- mRNAs (A1–D), the cER- protein isoform II is encoded
by a new cER- mRNA (A2), which is transcribed in vivo
from a specific promoter that is located in the region of the
previously assigned translation start site of the cER- gene. SI
nuclease mapping analysis reveals that cER- mRNA A2 is liver
enriched. The resulting cER- forms I and II differ in their ability
to modulate estrogen target gene expression in a promoter- and cell
type-specific manner. Whereas cER- form I activates or represses in
a strictly E2-dependent manner, the truncated
form is characterized by a partial transactivating or repressing
activity in the absence of its ligand. Comparison of the N-terminal
coding regions of different vertebrate ER- reveal a conservation of
the translation start methionine of the protein ER- form II in other
oviparous species but not in mammals. The expression of two classes of
ER- transcripts encoding the two ER- receptor forms in the liver
of Xenopus laevis and rainbow trout is demonstrated.
Therefore, the existence of two functionally different protein isoforms
produced from the ER- gene is probably a common and specific feature
in oviparous species. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The physiological changes induced by estrogens often result from modifications in the expression patterns of specific genes and are mediated through specific nuclear proteins, the estrogen receptors (ERs). These receptors belong to a large family of ligand-activated transcription factors whose members, the steroid, thyroid hormone, and retinoic acid receptors, regulate gene expression by interacting either in a protein/DNA manner with cognate DNA sequences called responsive elements (Ref. 5 and references therein) or in a protein/protein manner with other transcriptional factors (6). Structure-function analysis of these receptors reveals that they are modular proteins composed of a DNA-binding domain (region C) and a hormone-binding domain (region E) (7, 8). In addition, two other regions have been shown to mediate their capability to transactivate target genes. A transcriptional activation function 1 (AF-1) has been located in the A/B domain and functions as an independent transcription activating domain in yeast and mammalian cells. A hormone-inducible transcription activating domain, AF-2, is present in the hormone-binding domain. Both AF-1 and AF-2 are required for optimal stimulation of transcription, but the relative contributions of the two varies in a cell- and promoter-specific manner (9, 10, 11).
Diversity in the family of receptors that respond to the same ligand is
generated in many ways. In some instances, receptors are encoded by
separate genes present at discrete genetic loci. These include the
retinoic acid receptors RAR, -ß, and -
(12, 13, 14, 15) and RXR,
-ß, and - (16, 17), the thyroid hormone receptors, TR and TRß
(18), and the two estrogen receptors, ER and ERß, found in mammals
(19, 20, 21). For the same gene, differential promoter usage and
alternative splicing may also generate receptor isoforms, differing
either at the N terminus (22, 23, 24) or at the C terminus (25). A relevant
illustration of this is the progesterone receptor (PR), which exists as
two isoforms, PR-A and PR-B. These forms are produced from distinct
promoters and differ at their N-terminal end. The human PR-A lacks the
first 164 amino acids of PR-B while sharing the remainder of the coding
region of the gene. An indication of the possible biological
significance of protein isoforms lies in the demonstration that the two
PR isoforms have different efficiencies in inducing
progesterone-responsive element-mediated responses depending on the
promoter and the cellular contexts (26, 27, 28).
Both human and chicken ER genes have been shown recently to generate
several mRNA variants (A–F, hER mRNAs; and A–D, cER mRNAs) by
alternative splicing of upstream exons to a common site upstream of the
translation start site (29, 30). All of these transcripts differ in
their 5'-untranslated regions (5'-UTR) but code for the same ER
protein. In this paper we report the existence of a second cER
protein referred to as cER form II (61 kDa in size). This form lacks
the first 41 amino acids present at the N terminus of the previously
characterized full-length cER form I (66 kDa in size). The I and II
forms of the cER receptor can be produced in vitro and
are also detected in vivo in oviduct and liver tissues. We
show that each cER protein isoform originates from distinct mRNA
classes (mRNAs A1–D and A2) that are transcribed in
vivo from different promoters. The tissue-specific expression of
both cER- mRNA species A1–D and A2 was analyzed and shown to vary
in the different chicken tissues tested. Electrophoretic mobility shift
assays demonstrated that cER- form II is able to bind to an
estrogen-responsive element (ERE) in vitro, but in contrast
to cER- form I, the N- terminal truncated form can, to a limited
extent, modulate estrogen-responsive promoter activity in an
E2-independent manner in vivo. Finally,
comparison of the amino acid sequence of the N-terminal coding regions
of different vertebrate ER- proteins indicates that the two ER-
forms exist in other oviparous species but not in mammals. We show that
in the frog (Xenopus laevis) and rainbow trout
(Oncorhynchus mykiss), distinct ER mRNA transcripts may
encode both ER protein forms I and II. The possibility that the two
ER forms could play different roles in the control of gene
expression by estrogens in oviparous species is discussed.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Transcripts Whose 5'-Ends
Are Located Downstream of the Previously Assigned Translation Start
Site mRNAs was compared in two main
estrogen-responsive tissues in chicken, oviduct, and liver, by a SI
nuclease mapping analysis using probes that covered either the
5'-extremity of the gene including the transcription start site of the
A1 mRNA isoform (+1) and exon 1B–1D acceptor splice site in exon 1A
(+154) (probe A), or a part of the region coding for the
hormone-binding domain (probe B) (Fig. 1A
). Quantitative analysis of the results
showed that the intensity of the cER-specific signal obtained with
total RNA from oviduct was comparable using both probes (data not
shown). However, a clear difference was found in the signal level from
the two probes when liver total RNA was used. The signal detected by
probe B mapping a part of the hormone-binding domain was greater in
intensity than that found using probe A mapping the 5'-extremity of the
gene, although both probes were similar in their size and specific
activity (data not shown). These data indicated that new cER- mRNA
variant(s), resulting from either transcriptional initiation(s) or an
alternative splicing event(s) that occur between the two regions mapped
by probes A and B, contribute in part to the total cER mRNA content
in liver.
|
gene
previously delimited, another SI-nuclease mapping experiment was
performed using probe C, which was complementary to cER- mRNA A1
sequences from +158 to +892, as schematically depicted in Fig. 1A
. A
protected fragment of 734 nucleotides, corresponding in size to the
fully protected probe, was obtained from total RNA of both laying hen
oviduct and liver tissues, while no signal was visible with yeast total
RNA used as a negative control (Fig. 1B). This fragment was expected
from many previous experiments and results from a protection of cER
mRNA isoforms A1–D. In addition to this band and a smear of minor
specific bands that are probably caused by hybridization of cER
mRNAs with partially degraded SI probes (see the pattern of the free
probe), two other major specific products of 607 and 588 nucleotides,
respectively, mainly present in liver RNA sample, were detected. The
5'-extremities of these protected fragments are located in a region
downstream from the initiator methionine position, at +285 and +304 in
exon 1A. The fact that the same two sites were subsequently mapped by a
primer extension analysis (Fig. 1C), again most notably in the liver
RNA sample, excludes the possibility that they resulted from a splicing
event. Therefore, the new mRNA species (called cER mRNA A2 to
distinguish them from the previously characterized cER mRNA A1) are
transcriptionally initiated within exon 1A of the cER gene at
positions +271, +285, and +304.
Examination of the cDNA sequence downstream of the newly determined
transcription start sites showed the presence of an ATG codon at
position +347/9 (methionine +42), which is in-frame with the remainder
of the cER open reading frame (Fig. 1A). Analysis of the sequence
surrounding this ATG (5'-GCGACATGT-3') revealed
a favorable Kozak sequence for translation initiation (31). Therefore,
this ATG could function as a translation initiation codon for cER-
mRNA A2 and thus would give rise to a truncated 547 amino acid cER
protein with a predicted size of 61 kDa.
The level and pattern of distribution of cER- mRNA A2 were analyzed
in a panel of chicken tissues using SI probe C. The results show that
these transcripts are mainly expressed in liver and ovary (Fig. 1D). It
should be noted however that a weak expression of cER mRNA A2 was
observed in a minority of the analyzed oviduct RNA samples (data not
shown). The fully protected fragment that is specific to the remainder
of the cER mRNA isoforms (
A1–D cER mRNAs) was visible as
expected in oviduct, liver, ovary, and lung. Using probe C, it was not
possible to find a significant expression level of A1–D cER
mRNAs in kidney and testis. Densitometry of the signals obtained in
this experiment allowed a quantitative comparison of the expression
levels of A2 cER mRNA with those found for A1–D cER mRNAs.
Summarized in a table below Fig. 1D, these data are expressed as a
percentage of the total cER mRNA expression detected in oviduct
using the S1 probe C. This comparative analysis revealed a tissue
specificity in the level of expression of the A2 cER mRNA isoform.
The liver is the main site of expression for this isoform, although
weak expression was detected in ovary and occasionally in oviduct.
A Functional Promoter Maps the Region of the Previously Assigned
Translation Start Site of the cER Gene
To investigate whether sequences located upstream of the
transcription start site of cER mRNA A2 exhibited promoter activity,
three fragments from -503 to +183, -503 to +381, and +32 to +381 were
generated by PCR using as a template a genomic 
clone containing
exon 1A and 3 kb of sequence upstream of the 5'-end of the cER cDNA
(32). These fragments were subcloned upstream of the luciferase
reporter gene of the pGL2 basic vector, thus generating the reporter
vectors, pGL-503/+183, pGL-503/+381, and pGL+32/+381, as
schematically illustrated in Fig. 2A. A
reporter plasmid pGL(GH4) containing 802 bp of nonspecific DNA was also
constructed for use as a size control for the transfection experiments.
These vectors and the parental vector pGL2 basic were transiently
transfected into chicken embryo fibroblast (CEF) cells. Compared with
the size control and the empty parental vector, each cER
gene-specific fragment was able to initiate transcription as determined
by an increase in luciferase activity (Fig. 2B). The highest activity
was observed with pGL+32/+381, demonstrating that the region located in
the vicinity of the transcription start sites for cER mRNA A2
contains a functional promoter, called promoter A2 hereafter. When this
region was extended to include sequence from promoter A1
(pGL-503/+381), a 1.5-fold decrease in promoter activity was observed.
This effect may be attributed to the putative silencer elements located
between -437 and -384 in this second promoter (32). Finally, promoter
A2 was able, in part, to counterbalance the activity of this silencer
as a further 1.5-fold reduction in activity was observed when promoter
A2 region was removed from construction pGL-503/+381 giving rise to
pGL-503/+183 (Fig. 2B).
|
Protein Isoforms I and II Are Produced in
Vivo mRNA A2 showed
the presence of an ATG codon at position +347/9 (methionine +42), which
is in frame with the remainder of the cER open reading frame (Fig. 1A). To test whether this in-frame ATG could function as a translation
initiation codon for a truncated cER protein with a predicted size
of 61 kDa, cER- cDNAs were inserted in pSG5 to generate the
expression vectors pSG cER- I and pSG cER- II. The expression
vector pSG cER II, which contained sequences from +308 to +2038, was
expected to generate a cER protein in which translation was
initiated at methionine 42, while vector pSG cER I, composed of the
original cER cDNA (from +158 to +2038), should produce cER
proteins initiating at methionine 1, and methionine 42 if there is
leakiness in the translation initiation (Fig. 3A). The hER cDNA expression vector
(HEO) was used as a positive control (33). Analysis of the expression
of constructs pSG cER- I and HEO in a rabbit reticulocyte lysate
system in the presence of radioactively labeled methionine showed
66-kDa ER protein, as expected (Fig. 3B). Interestingly, a second
smaller cER protein was observed from the expression vector pSG
cER I whose size correlated to a truncated cER receptor protein
starting at methionine 42, a similar leakiness of translation is noted
for HEO with an alternative ATG used to generate a minor protein band.
The second cER form was the only form produced by the expression
vector pSG cER II, thus confirming the initiating translational
functionality of the in-frame ATG codon at position +347/9 (methionine
+42) (Fig. 3B). This new protein was called cER form II relative to
the full-length receptor cER- form I.
|
protein (34). The result confirmed the expression
in both tissues of two cER proteins of the expected molecular mass,
66 and 61 kDa (Fig. 3C). To compare the levels of the two cER forms,
the protein signals from the immunoblot were quantified by
densitometry. An 8- to 10-fold excess of the cER form II over form I
was observed in the liver, whereas cER protein I was the major form
expressed in the oviduct (5 times higher). Finally, the presence of
another protein, recognized by this H 222 antibody, should be noted.
This protein of apparent molecular mass of 45 kDa corresponds most
likely to initiation of translation at an internal methionine (Meth.
170) within the coding region of the cER cDNA.
cER Form II Protein Binds Specifically to an ERE
As the DNA-binding domain of cER protein form II is identical
in sequence to the original cER, it was expected to bind with
similar efficiency to the same type of estrogen response element (ERE),
namely an inverted repeat spaced by three nucleotides (GGTCAnnnTGACC).
Electrophoretic mobility shift assays with both chicken ER receptors
were conducted to confirm this assumption. Figure 4 shows that cER protein isoforms
produced by the rabbit reticulocyte lysate system are indeed able to
form retarded complexes in the presence of the radiolabeled consensus
ERE from the apoVLDL II promoter (35). The specificity of these
complexes was confirmed by competition experiments using equal, double,
or a 10-fold excess of unlabeled consensus apoVLDLII-ERE, since a
reduction and ultimate elimination of the ER-ERE complex signal were
observed. In addition, a 10-fold excess of a mutated ERE had no effect.
The complexes had different mobilities depending on the extract used.
The slower migrating complex (A) was obtained from the extract
producing cER- form I (66 kDa), whereas the faster one (C) is
produced from cER form II (61 kDa) extract. Finally, an intermediary
mobility complex was generated when both receptor forms were
synthesized simultaneously from the rabbit reticulocyte lysate.
Therefore, complex A and C presumably correspond to homodimers of the
cER form I and II, respectively, while complex B is probably due to
the formation of cER form I and II heterodimers. Also worth
emphasizing is the fact that when both receptor forms are coproduced,
no cER form I homodimer formation is visible. This is probably
accounted for by a cER form I/II ratio of approximately 1:2.
|
Form I and II.
As expected from previous work (37), the ability of E2 to
induce the VTG II promoter was found to be cell specific since a
hormone- and receptor-dependent induction was observed only in LMH
cells (Fig. 5). No difference in the
transactivation efficiency (10- to 20-fold increase) of VTG II promoter
was detected between cER- form I and II proteins in these cells
(Fig. 5). However, results showed that whereas no transactivation of
the VTG II promoter by either of the two liganded cER- forms was
observed in CEF cells, the unliganded cER- form II was able to
repress the basal activity of VTG II promoter (3-fold decrease).
Results of a similar nature were obtained using the second chimeric
estrogen-responsive apoVLDL II gene. This construct was induced in the
presence of E2 to a similar level (100- to 200-fold
increase) in both cell types in the presence of the two cER- forms I
and II (Fig. 5). In contrast, whereas the transactivation of the
apoVLDLII-CAT reporter gene by cER- form I is hormone dependent in
both cell lines, the N-terminal truncated cER- form II is able to
partially transactivate it in a significant manner in the absence of
its ligand (10- to 20-fold increase). It should be noted that, in LMH
cells, the induction of apoVLDLII-CAT activity measured in the presence
of estradiol, but in the absence of cotransfected cER- receptor, may
be explained by the presence of endogenous cER- in this
liver-derived cell line. We conclude from these data that the cER-
form II, unlike cER- form I, possesses a hormone-independent
transactivation activity that functions in a promoter-dependent
manner.
|
gene (3, 38), it was also interesting to
study the effect of the two cER- protein isoforms on the promoters
that control their production. Therefore, the reporter constructs
pGL(GH4), pGL-503/+183, pGL-503/+381, and pGL +32/+381 described
earlier (see Fig. 2A) were transfected with the expression vector pSG5,
pSG cER- I, pSG cER- II, or pSG RAR (negative control) in CEF
cells. The data showed that each of the three promoter fragments was
down-regulated in a ER--dependent manner as shown in Fig. 6. Moreover, as previously found for the
apoVLDLII-CAT reporter gene, the truncated form of the cER- was able
to function in absence of estradiol whereas the full-length receptor
required the ligand to down-regulate the transcriptional activity of
the cER- promoter fragments. As all three promoter constructs
reacted in a similar manner to the presence of receptor, we deduced
that the cis-acting element mediating this effect was
probably in a region common to all three promoter fragments and is
therefore between +32 and +183 in the cER- gene.
|
Form II Receptor Is Conserved in
Oviparous Species but Not in Mammals proteins demonstrates that the first initiator methionine codon,
called ATG1, is conserved in all species analyzed with the exception of
rainbow trout (Fig. 7). Translation from
ATG1 results in the production of the classical 66-kDa ER- protein.
In contrast, the original rtER- cDNA, cloned from a rainbow trout
liver cDNA library, encodes an N-terminal truncated ER- protein that
is translated from an initiator methionine similarly positioned to the
second translation start site (ATG2) used to produce the liver-enriched
cER- form II protein. ATG2 is also conserved in X. laevis
but is absent in human, mouse, and rat ER- receptors. In mammals,
this downstream methionine codon was replaced by a valine that is
unable to initiate translation (Fig. 7). These data suggested the
likely existence of previously unidentified ER- protein forms in
rainbow trout and X. laevis, equivalent to the 66- and
61-kDa cER- forms, respectively. If indeed this was the case, then
the production of these two ER- protein forms from a single ER-
gene might be a conserved feature of oviparous species.
|
transcripts
in X. laevis and rainbow trout that might encode two protein
forms of ER-, similar to the classical full-length cER- form I
and the N-terminal-truncated cER- form II. The SI probes for the
chicken and X. laevis ER- genes used in this experiment
were designed to cover exon 1 and a part of its 5'-flanking region as
depicted in Fig. 8A (42, 45). Regarding
the rainbow trout ER- gene, it was previously shown that the
assigned translation start site was located in the second exon, which
was separated from the untranslated first exon by an 846-bp intron 1
(46, 47). However, the analysis of the 5'-genomic flanking sequence of
exon 2 revealed an in-frame initiator methionine, 45 amino acids
upstream of the previously characterized initiator codon (this upstream
translation start site in intron 1 will be called ATG1 and the
initiator codon in exon 2 will be renamed ATG2 hereafter) (see Fig. 8A). This new organization of potential translation start points in the
rtER- gene was therefore comparable to that described herein for the
chicken. To ascertain whether this 5'-intronic flanking sequence of
exon 2 was transcribed, a specific rainbow trout SI probe was designed
to protect this region and a part of exon 2 (see Fig. 8A). Finally, to
allow for easier comparison of the SI nuclease mapping results, the
three ER- probes were designed to contain the same distance (286
nucleotides) between their 3'-end and ATG2.
|
confirmed those described earlier. Four protected
fragments representing the transcription start site of the cER- mRNA
A1 transcript, the splice site at position +154, and the two
transcription start sites for cER- mRNA A2 were detected. In the
case of X. laevis, two protected fragments of approximately
520 and 350 nucleotides in size were also specifically observed after
SI nuclease analysis of liver RNA from females in vitellogenesis (Fig. 8B). The 5'-extremity of the longest was positioned upstream of the
translation initiation codon (ATG1) position, whereas the 5'-end of the
shorter fragment was located in a region between ATG1 and the putative
initiator methionine codon ATG2. These two positions were also mapped
by a primer extension experiment (performed as described earlier for
the chicken ER- gene) (data not shown), thereby demonstrating that
they correspond to transcription start sites of distinct xER- mRNAs,
which potentially encode different xER- protein forms of 66 and 61
kDa. It should be noted that no fragments corresponding to a protection
which extends to the previously described transcription start site of
the xER- gene were detected (45). Neither was a partial protection
of the Xenopus SI probe as far as a position equivalent
to the alternative splice site found in exon 1 of the cER- gene
observed, even though sequence analysis reveals a good candidate
acceptor site (ctgttttcag/GTG) at a similar position in exon 1 of
xER- gene. Finally, as for the two previous oviparous species
investigated, the results of the SI nuclease mapping analysis of
rainbow trout liver RNA pointed to the existence of transcripts that
potentially encoded the different ER- protein forms. As expected,
the previously described rtER- mRNA (43, 44), as well as alternative
splicing variants (47), was partially protected by the SI probe up to
the 5'-extremities of exon 2 due to the noncomplementarity of their
5'-end (the untranslated exon 1) to the probe (see asterisks
in Fig. 8B). This transcript has the potential to encode the previously
described rtER- protein (N-terminal truncated form). The SI nuclease
mapping experiment showed, in addition, the existence of new rtER-
transcripts which initiate in a region of intron 1 at
approximately 40 and 100 nucleotides upstream of the predicted in-frame
initiator methionine (ATG1), as confirmed by a primer extension
experiment (data not shown). The translational product of this new
class of transcripts should be a longer form of rtER- protein
equivalent to the 66-kDa cER- receptor.
The conclusion of these SI nuclease mapping analysis is that the liver
tissue of the oviparous species tested, chicken, X. laevis,
and rainbow trout, express distinct classes of ER- mRNAs able to
generate either the normal ER- protein or an N-terminal truncated
form.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
is known to be widely distributed in reproductive as well
as nonreproductive tissues, thereby mediating estradiol action on
various important physiological functions that range from female
sexual development and reproduction, to liver, fat, and bone cell
metabolism (1, 2, 3). It is obvious that the expression of the ER-
gene should be subject to a variety of controls to ensure that the
correct amount of ER- protein is available in the correct cells at
the correct time. Therefore, the elucidation of the molecular
mechanisms controlling the tissue-specific pattern of ER- gene
expression should provide a starting point to understanding the
pleiotropic effects of its ligand in a wide range of biological
processes. Recent studies on the structure and organization of the
human and chicken ER- genes showed that these genes are complex
genomic units exhibiting alternative splicing and promoter usage in a
tissue-specific manner (29, 30). In both species, all identified ER-
mRNA isoforms diverged in their 5'-UTR sequences upstream from the
translational initiation codon and therefore encoded a common ER-
protein of 66 kDa in size.
In this present study, we demonstrated that the chicken ER- gene is
able to produce a second cER- protein of 61 kDa in size, called
cER- form II. As shown by its in vitro production using
the rabbit reticulocyte lysate system, this second form results from
initiation of translation at methionine 42 and thus lacks the 41 amino
acids present at the N-terminal domain of the 66-kDa cER- form I.
Two mechanisms might be involved in the production of these two
proteins. Both isoforms may be translated from the same transcripts as
a consequence of the leaky ribosome-scanning mechanism (31), or their
production could arise from distinct mRNA species, each of which
encodes a different protein. Whereas in vitro and in
vivo translation of the previously identified cER- mRNAs
(A1–D) generated both forms of cER- protein (see Fig. 3, B and C,
oviduct sample), which indicates that the leaky ribosome-scanning
mechanism may occur, SI nuclease mapping and primer extension analysis
of liver total RNA allowed the existence of another class of cER-
mRNA transcripts (A2) to be shown, in agreement with the second
mechanism. This new class of cER- transcripts is transcribed from
positions +285 and +304 in exon 1A, between the translation initiation
methionine 1 and methionine 42. Therefore, these transcripts are unable
to encode the cER- protein form I. Their translational product is
the truncated receptor form II. Sequence analysis of the 5'- UTR region
of A2 cER- mRNA revealed that this region is devoid of short open
reading frames (sORFs) in contrast to the 5'-UTRs of A1–D cER-
transcripts, which contain at least one sORF (30). The significance of
these sORFs remains to be elucidated, but similarly placed sORFs in
other messages such as the GCN4 and the BCR/ABL oncogene mRNA have been
shown to be involved in the posttranscriptional control of their
expression (48, 49). Therefore, further experiments should be
informative on the existence of a differential turnover and/or
translational regulation of the two classes of cER- mRNAs.
The 5'-flanking genomic region of the new A2 cER- transcripts was
shown to be able to promote the transcription of a luciferase reporter
gene in transient transfection experiments performed in the CEF cell
line, thus providing evidence that a previously unknown promoter (pA2)
was located in exon 1A regions coding for the N-terminal part of
cER- protein form I. These data demonstrated clearly that cER-
form II can be specifically generated from transcripts distinct from
those encoding form I and that a unique promoter is responsible for the
production of cER- form II only. In contrast to the previously
characterized pA1 promoter, which contains a well positioned typical
TATA sequence, with potential CAAT box sequences close to the start
site of transcription (32), the promoter of A2 cER- transcripts is
devoid of any obvious TATA or CAAT-box sequences. Computer- assisted
analysis of promoter A2 sequences failed to identify any consensus EREs
(50). This is in contrast to pA2 transfection results in CEF cells,
which demonstrated clearly its down-regulation by E2 in an
ER--dependent manner and therefore suggests that protein/protein
interaction is involved. Comparable data were previously obtained for
the pA1 promoter (30). As pA1 and pA2 promoters are located close to
each other, both promoters may be similarly affected by the
cis-acting element(s) involved in the autoregulation as
suggested by transfection studies herein. These results have to be
integrated with the reports that cER- gene expression is differently
regulated by estrogen in liver and oviduct tissues. Estrogen increases
cER- expression level in the liver (3), whereas cER- mRNA
expression is down-regulated in the oviduct (38). Interestingly,
promoter pA1 is the main promoter used to transcribe the cER- gene
in oviduct tissue while promoter pA2 activity is predominant in the
liver and weak or absent in oviduct. Therefore, the mechanism by which
a single ligand, estradiol, mediates such opposing effects in distinct
tissues probably involves complex interactions between
cis-acting element(s) and tissue-specific factor(s). Further
studies, such as the analysis of the DNAse I-hypersensitive sites,
followed by a more detailed promoter characterization using the
in vivo footprinting technique would be informative in
identifying sequences involved in this cell-specific expression of the
cER- mRNAs.
Analysis of the pattern of expression of the two classes of cER-
mRNAs (A1–D and A2) revealed that their relative levels vary in the
different chicken tissues. As previously reported (30), the highest
amount of A1–D cER- mRNA class was detected in the oviduct, and
lower amounts were present in liver and ovary, two other chicken female
reproductive tissues tested. In contrast, expression of the second
cER- mRNA class was mainly observed in liver, was weakly detected in
ovary, and, in most of the samples studied, was absent in oviduct. The
consequence at the protein level of this differential distribution
pattern of the two cER- mRNA classes should be that tissues
expressing the first class of cER- mRNAs will only contain both
cER- protein forms as the result of the leaky scanning mechanism
(31) while tissues that coexpress A2 cER- mRNA transcripts will
produce the truncated receptor as the main cER- protein form.
Therefore, the presence of A2 cER- mRNA class in a tissue should
allow the ratio between the two cER- proteins to be changed in favor
of form II. Immunoblot analysis of chicken oviduct and liver nuclear
extracts confirmed this hypothesis since both isoforms were detected in
oviduct and liver tissues with the truncated cER- form II
predominantly produced in liver tissue where it accounted for
approximately 80% of the total immunoreactive ER-.
Two distinct classes of ER- mRNAs potentially encoding the normal
and N-terminal truncated form of the ER- protein were also detected
by SI nuclease mapping analysis in the liver of other oviparous species
such as X. laevis and rainbow trout. In contrast to a
previous investigation (51), which proposed that the production of the
two forms of xER- was the result of the leaky ribosome-scanning
mechanism (31), the present study provides evidence that a new class of
xER- mRNAs initiate downstream of the previously determined
translation start site (ATG1) in a region coding for the N-terminal
part of xER- protein form I. Likewise, the initially characterized
rainbow trout ER- mRNA encodes a protein equivalent to the
N-terminal truncated receptor form II. The present study
demonstrated the existence of a second class of rtER- mRNAs in
rainbow trout, whose translation product is a receptor possessing 45
additional amino acids at its N-terminal domain. Western blot analysis,
previously performed using an antibody raised against the rainbow trout
ER- protein, confirmed the presence of two receptor forms in rainbow
trout liver (52). However, the origin of these two proteins was not
determined in that study (52). Comparison of the N-terminal primary
amino acid sequence of vertebrate ER- proteins reveals the
conservation of the second translation start codon (ATG2) in oviparous
species (chicken, X. laevis, and rainbow trout) but not in
mammals (human, mouse, and rat). These data indicate that the existence
of two different protein isoforms produced from the ER- gene is
probably a common and specific feature in oviparous species and
suggests that the two ER- forms could play different roles in the
control of gene expression by estrogens in these species.
The two ER- protein forms I and II identified in this report show
similarity with the A and B forms of the chicken and human PR (26, 53, 54). The PR form A is an N-terminally truncated variant of PR form B
that may arise by translation initiation at an internal methionine
codon of the PR mRNA (55) or by translation of a specific PR mRNA (54).
Several groups have reported that PR form A and B have distinct
functions and influence gene transcription differently depending on the
promoter and cell context. For instance, Tora et al. (26)
found that chicken PR form B is entirely inactive on the chicken
ovalbumin promoter, in a setting in which form A is a strong
transactivator, while the MMTV promoter was activated more efficiently
by form B than form A. In addition, human PR form A is a stronger
transactivator than form B on the tyrosine aminotransferase promoter
(27). Likewise, the present study showed that the two chicken ER-
forms differ in their ability to modulate transcription activity of
estrogen target genes in a promoter- and cell type-specific manner.
Whereas cER- form I activates or represses in a strictly
E2-dependent manner, the truncated form is characterized by
a partial activity in the absence of its ligand. Supporting this result
is the previous observation that the rainbow trout ER-, which is
equivalent to the N-terminally truncated cER- form II, exhibits a
basal transcriptional activity in yeast, while no activity was measured
for the human ER- (full-length ER- form) (56). This functional
difference between the human and rainbow trout ER-, or between the
two chicken ER- forms, is probably due to the presence or absence of
the N-terminal A domain in the receptor protein. Further structure
function studies will be necessary to elucidate the exact role played
by this A domain in mediating target gene activity and especially its
impact on the two transcription activation functions (AF-1 and AF-2) of
the ER.
The evolutionarily conserved production of a functionally different
ER- isoform in the liver of oviparous species (but not in mammals)
suggests a key role for this form in the production of very-low-density
lipoproteins and/or egg yolk proteins such as vitellogenin.
Vitellogenesis is an important estrogen-initiated process unique to the
liver of oviparous vertebrates (57). Previous studies aimed at
providing insights into the molecular events involved in vitellogenesis
and its regulation have employed the full-length ER- cDNA whose
translational product is the receptor form I (66 kDa) (3, 58). The data
presented in this paper demonstrate that this is not a true
representation of the in vivo hepatic environment where the
main ER- mRNA isoform expressed encodes an ER- protein devoid of
its N-terminal 41 amino acids. Gene transfer technology may be helpful
in providing an answer to the interesting question of whether there are
physiological consequences resulting from blocking the production of
the truncated ER- form II by mutation of methionine at position 42
(ATG2).
In conclusion, the production of two ER- forms from a single ER-
gene in oviparous species provides two different transacting regulatory
proteins which, in addition to their respective spatial distributions,
differ in their ability to modulate the activity of estrogen-responsive
genes. This new ER- complexity may account, to a large extent, for
the pleiotropic effects of the corresponding ligand, estradiol, in a
wide range of physiological processes occurring in oviparous
vertebrates.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Modified SI Nuclease Mapping and Primer Extension
Modified SI nuclease protection and primer extension procedures
were followed as described by Flouriot et al. (59, 60).
These two methods involved the use of biotinylated single-stranded DNA
templates to prepare highly labeled single-stranded DNA probes or long
primers by extension from specific primers with the T7 DNA polymerase
in the presence of [-32P]dCTP (3000 Ci/mmol). These
probes or long primers are then hybridized with the appropriate RNA
sample and subjected either to an S1 nuclease digestion or to a reverse
transcriptase extension, respectively.
To prepare the template used to make probe C (from +158 to +892) (see
Fig. 1A), an RT-PCR reaction was performed. The 5'- and 3'-primers used
for the amplification were S1 (5'-ACTGCCAGCTGCCGATCTTG-3') and S2
(5'-ATAGTACACTGGTTAGTGGCAG-3'), respectively. The RT-PCR product was
subcloned downstream of T7 and upstream of M 13 reverse primer in the
TA cloning vector pCR 2.1 (Invitrogen, San Diego,
CA). A PCR was then performed using a biotinylated T7 primer with M 13
reverse primer.
The origin of the chicken (probe D), X. laevis, and rainbow
trout probe templates was genomic PCR products obtained by
amplification of the regions from -232 to +633 (32) (see Fig. 8A),
from -243 to +632 (42, 45) (see probe D in Fig. 8A), and from -424 in
intron 1 to +508 in exon 2 (46, 47) (see Fig. 8A). The DNA fragments
were then subcloned in the pCR 2.1 vector (Invitrogen)
downstream of T7 and upstream of M 13 reverse primer. A PCR reaction
was then performed using a biotinylated T7 primer with M 13 reverse
primer.
The template for the long cER- primer was obtained by RT-PCR using
the 5'-biotinylated S3 (5'-GCAACAAGACAGGAGTTTTTAACTA-3') and the
3'-primer S4 (5'-CTGTAGAA-GGCTGGAGGAGCAGCT-3').
All biotinylated PCR products were bound to streptavidin-coated magnetic beads (DynAl) as recommended by the manufacturer (Dynabeads, Dynal, Hamburg, Germany), and the nonbiotinylated DNA strands were removed in 0.1 M NaOH. The S1 probes (C and D, Xenopus and rainbow trout probes) and the long primer were obtained by extending their respective primer [S2, S5 (5'-CCCTCATCCCAAAGCTGCCCTGT-3'), X1 (5'-TTACTGCGAAAGTGCCCTGCTCAC-3'), RT1 (5'-CCAGGTAGTATGACTGGCTGG-3') and S6 (5'-ATGGATGAAGGGTGAGAGCTG-3'), respectively] annealed to the corresponding biotinylated single-stranded template. After elution of the single-stranded DNA probes by alkaline treatment and magnetic separation, 105 cpm of the probe or primer were coprecipited with 100 or 50 µg, respectively, of total RNA and then dissolved in 20–30 µl of hybridization buffer (80% formamide, 40 mM piperazine-N,N'-bis(2-ethanesulfonic acid), pH 6.4, 400 mM NaCl, 1 mM EDTA, pH 8), denatured at 70 C for 10 min, and hybridized overnight at 55 C. The S1 digestions and the reverse transcriptase extension were carried out as previously described by Ausubel et al. (61), and the samples were electrophoresed through denaturing polyacrylamide/urea gels.
Expression Vector Preparation
pSG cER- I was prepared as described by Griffin et
al. (30). By the same approach, the expression vector pSG cER-
II was made by directionally cloning the cER- coding region from
+308 to +2038 into the parental expression vector pSG5 (62). HEO [pSG5
expression vector containing the complete hER- cDNA (33)] and pSG
RAR [pSG5 expression vector containing the mouse retinoic acid
receptor cDNA] were gifts from P. Chambon.
In Vitro Transcription and Translation
In vitro transcription and translation were
accomplished with the TNT-coupled Reticulocyte Lysate system from
Promega Corp. (Madison, WI) following the manufacturers
directions. pSG5 recombinant expression vectors pSG cER- I, pSG
cER- II, and HEO were used as templates for transcription with T7
RNA polymerase followed by translation to generate cER- I, cER-
II, and hER- proteins. Translation efficiency was checked by
incorporating [35S] methionine. Two microliters of the
radioactive translation product were run on a 10% SDS-PAGE gel as
outlined by Laemmli (63). The gel was dried and autoradiographed. Cold
methionine was used in the in vitro transcription and
translation of proteins for electromobility shift assays and Western
blot analysis.
Nuclear Extract Preparation
Nuclear extracts were prepared from freshly killed egg-laying
hen oviduct and liver tissues using the procedure of Sierra et
al. (64). Protein concentration was determined using the Bradford
protein assay solution from Bio-Rad Laboratories, Inc.
(Richmond, CA).
Western Blot Analysis
Nuclear protein from egg-laying hen oviduct and liver tissues
and 7.5 µl of in vitro transcription and translation mix
were subjected to SDS-PAGE. Proteins were denatured at 95 C for 15 min
and resolved on a 10% SDS polyacrylamide gel next to prestained
broad-range protein standards from Bio-Rad Laboratories, Inc. and electrotransferred to Immobilon nitrocellulose membrane
(Millipore Corp., Bedford, MA). The membrane was blocked
in TS (10 mM Tris, pH 7.4, 0.5 M NaCl)
containing 10% (wt/vol) nonfat dry milk powder. The membrane was
incubated with primary antibody (2 µg/ml)-anti hER- monoclonal
antibody H 222 kindly provided by Dr. G. L. Greene (34) in TS
containing 3% nonfat dry milk powder for 1 h at room temperature.
Incubation with secondary peroxidase-coupled goat antirat antibody was
performed under the same conditions. ER- proteins were visualized by
chemiluminesence using the ECL system from Amersham Pharmacia Biotech (Arlington Heights, IL) according to the manufacturers
instructions. Signals were quantified by densitometry.
Electrophoretic Mobility Shift Assay
Chicken ER- I and ER- II proteins were prepared by
in vitro transcription and translation as described above.
One microliter of in vitro translated product was
preincubated at 0 C for 20 min in binding buffer (10% glycerol, 10
mM Tris, pH 7.5, 1 mM MgCl2, 0.5
mM DTT, 0.5 mM EDTA, 0.1 M KCl) in
the presence of 80 mM Mg/Spermidine, 2 mg of poly dI:dC,
2.5 mg sonicated salmon sperm DNA, and 10 mM
Na2HPO4. The samples were then incubated with 1
ng of radioactive oligonucleotide probe [50,000 dpm] end labeled with
[-32P] ATP (3000 Ci/mM) using T4 polynucleotide
kinase. Competition was performed by premixing different concentrations
of unlabeled competitor oligonucleotide with radioactively labeled
oligonucleotide before addition to the binding reaction. Protein-DNA
complexes were separated from free probe by nondenaturing
electrophoresis on a long 5% polyacrylamide gel in 0.25 x TBE.
The gel was prerun at 4 C for 60 min followed by 5 h running at
600 V. After electrophoresis the gel was fixed for 30 min in a 10%
methanol/acetic acid solution and dried before autoradiography.
Sequence of the consensus ERE 30-bp oligonucleotide was derived from
the 5'-flanking region of the chicken apoVLDL II gene (-186 to -156)
(65). The nucleotide sequence was 5'-ctgtgctcaGGTCAG-
ACTGACCttccatta-3' with the wild-type consensus ERE sequence
shown in uppercase letters. The sequence of a mutant version
of this oligonucleotide (m) was 5'-ctgtgctca-
GGACACTGTGTACttccatta-3'. The mismatches are
underlined. Both oligonucleotides were used as
double-stranded DNA for the electrophoretic mobility shift assay.
Promoter Construct Preparation
To create the luciferase reporter plasmids pGL-503/+183,
pGL-503/+381, and pGL+32/+381, PCR was used to amplify, from a genomic
clone containing 3 kb of sequence upstream of the 5'-end of the
cER- cDNA (33), regions from -503 to +183, -503 to +381 and +32 to
+381, respectively. For each construction, the two synthetic primers
used for the amplification [pAI (5'-
TGCGCTGGTACCTCTTTTACATTCTTCAATTTCTG-3') and pAII
(5'-AGTGCGAAGCTTCAACAGCAAGATCGGCAGCTGG-3'), pAI and pBII
(5'-CGTGCGAAGCTTTAAAAACTCCTGTCTTGTTGCTT-3') or pCI
(5'-GTTCATGGTACCCCAGTGCTCACCCTGCATTTGT-3') and pBII,
respectively] were designed to introduce 5' KpnI and 3'
HindIII restriction sites (underlined within the
primer sequence) at the ends of the PCR products. The amplified
fragments were directionally cloned into the polylinker of the pGL2
basic plasmid (Promega Corp.) upstream of a luciferase
reporter gene. pGL(GH4) reporter plasmid containing 802 bp of
nonspecific DNA was also constructed for use as a size control for
transfection experiments.
The following CAT-containing reporter plasmids were kind gifts: The apoVLDL II gene promoter fragment from -900 to +1455 (apoVLDLII-CAT) from M. Evans (35) and the chicken vitellogenin II promoter fragment from -1133 to +11 from J. Burch (36). This was subsequently subcloned into pGL2 basic vector to give the construct VTGII-LUC.
Cell Culture and Transient Transfection Assays
CEF cells (a gift from T. Graf, Heidelberg, Germany) were
maintained in DMEM supplemented with 5% FCS, 1% chicken serum, 10
mM HEPES, pH 7.4, 100 U/ml penicillin, and 100 µg/ml
streptomycin at 37 C in a 5% CO2 incubator. LMH cells
(ATCC, Manassas, VA) were grown in Weymouth’s MB/251
medium with 10% FCS, L-glutamine, 100 U/ml penicillin, and
100 µg/ml streptomycin.
CEF cells were transiently transfected using the DNA/calcium phosphate
coprecipitation method (66). LMH cells were transiently transfected as
described by Binder et al. (37). In all transfection
studies, 6-cm dishes containing 5 x 105 cells were
transfected with a total of 10 µg of DNA per dish [5 µg reporter
plasmid, 1 µg of expression vector, 0.1 µg of internal control
(EF-1-CAT or pCMV-tk-LUC) (67), and carrier DNA to 10 µg
(pBluescript)]. Medium was changed 6 h before transfection. After
16 h incubation with the DNA/calcium phosphate precipitate, the
medium was aspirated and cells washed twice with PBS, and fresh
serum-stripped phenol red free medium was added. Transfected cells were
cultured for 24 h in the absence or presence of 10-8
M 17ß-estradiol before harvesting for luciferase and CAT
assays. Luciferase assays were performed as outlined by Brasier and Ron
(68) on 20% of the lysate. The CAT activity was assayed for with the
ELISA kit from Roche Molecular Biochemicals (Mannheim,
Germany) using 20% of the lysate. The reporter gene activity
values were normalized for transfection efficiency according to the
activity of the cotransfected reference control (EF-1-CAT or
pCMV-tk-LUC).
|
FOOTNOTES |
|---|
This work was supported by the Irish American Partnership fellowship to Caroline Griffin, the Irish Cancer Society, and by an European Molecular Biology Laboratory long-term fellowship to Gilles Flouriot.
Received for publication January 14, 1999. Revision received May 6, 1999. Accepted for publication May 24, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and
ß genes. EMBO J 8:429–433[Medline]
and ß retinoic acid receptors and a novel
receptor predominantly expressed in skin. Nature 339:714–717[CrossRef][Medline]
and ß. Proc Natl Acad Sci USA 93:5925–5930 with specific patterns of expression. Proc Natl Acad Sci USA 87:2700–2704 are generated by alternative splicing and
differential induction by retinoic acid. EMBO J 10:59–69[Medline]
mRNA isoforms
generated by alternative splicing and promoter usage. Endocrinology 139:4614–4625 2u
globulin gene in liver and spleen nuclear extracts. Mol Biol Med 7:131–146[Medline]
This article has been cited by other articles:
 |
|
 |
|
  M. Longo, B. Peruzzi, D. Fortunati, V. De Luca, S. Denger, G. Caselli, S. Migliaccio, and A. Teti Modulation of human estrogen receptor {alpha} F promoter by a protein kinase C/c-Src-dependent mechanism in osteoblast-like cells J. Mol. Endocrinol., December 1, 2006; 37(3): 489 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Wilhelms, C. G. Scanes, and L. L. Anderson Lack of estrogenic or antiestrogenic actions of soy isoflavones in an avian model: the Japanese quail. Poult. Sci., November 1, 2006; 85(11): 1885 - 1889. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Merot, R. Metivier, G. Penot, D. Manu, C. Saligaut, F. Gannon, F. Pakdel, O. Kah, and G. Flouriot The Relative Contribution Exerted by AF-1 and AF-2 Transactivation Functions in Estrogen Receptor {alpha} Transcriptional Activity Depends upon the Differentiation Stage of the Cell J. Biol. Chem., June 18, 2004; 279(25): 26184 - 26191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kos, G. Reid, S. Denger, and F. Gannon Minireview: Genomic Organization of the Human ER{alpha} Gene Promoter Region Mol. Endocrinol., December 1, 2001; 15(12): 2057 - 2063. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Denger, G. Reid, M. Kos, G. Flouriot, D. Parsch, H. Brand, K. S. Korach, V. Sonntag-Buck, and F. Gannon ER{alpha} Gene Expression in Human Primary Osteoblasts: Evidence for the Expression of Two Receptor Proteins Mol. Endocrinol., December 1, 2001; 15(12): 2064 - 2077. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Menuet, I. Anglade, G. Flouriot, F. Pakdel, and O. Kah Tissue-Specific Expression of Two Structurally Different Estrogen Receptor Alpha Isoforms along the Female Reproductive Axis of an Oviparous Species, the Rainbow Trout Biol Reprod, November 1, 2001; 65(5): 1548 - 1557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Griffin, G. Flouriot, P. Sharp, G. Greene, and F. Gannon Distribution Analysis of the Two Chicken Estrogen Receptor-Alpha Isoforms and Their Transcripts in the Hypothalamus and Anterior Pituitary Gland Biol Reprod, October 1, 2001; 65(4): 1156 - 1163. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Métivier, F. G. Petit, Y. Valotaire, and F. Pakdel Function of N-Terminal Transactivation Domain of the Estrogen Receptor Requires a Potential {alpha}-Helical Structure and Is Negatively Regulated by the A Domain Mol. Endocrinol., November 1, 2000; 14(11): 1849 - 1871. [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 |
