| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Physiology University of Manitoba Winnipeg, Manitoba, Canada R3E 3J7
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- and ß-subunit genes (1, 2, 3, 4)
and potentially has a role in the placental expression of some other
genes (5, 6). Its involvement in the regulation of these several genes
and its role in the developmental regulation of other cell types has
led to speculation that AP-2 may participate in a developmental cascade
during trophoblast differentiation (1). The GATA (7, 8), Fos/Jun (AP-1)
(9), and TEF (10, 11) families of transcription factors have also been
implicated in the placental expression of some genes. None of these
factors, however, are expressed only in placenta, suggesting that there
are further protein/protein interactions that give specificity. The
recently characterized basic helix-loop-helix transcription factors
Mash-2 (12) and Hand-1 (13, 14, 15, 16, 17) (previously called Hxt, eHAND, Thing-1)
have been shown to be essential for determination of specific placental
cell types; no target genes have been identified for either factor,
although there is indirect evidence that mouse placental lactogen I
(PLI) may be regulated by Hand-1 (15). The limited data available make
it difficult to decide whether there may be a specific transcription
factor or combinations of factors important for placental-specific gene
expression or whether factors will vary from gene to gene. It would be
particularly useful to be able to study the transcriptional regulation
of multiple genes expressed in a specific placental cell type, thereby
allowing direct comparisons of the cis- and
trans-acting factors involved. The PRL-related genes expressed in the placentas of rats and mice represent an important resource for such studies. This is a large gene family expressed according to developmentally distinct expression patterns involving both specific placental cell types and temporal expression (18, 19, 20). PLI and PLII are the archetypes of these genes, being originally identified as placental proteins that could bind to the PRL receptor and that were highly expressed either early (PLI) or late (PLII) in pregnancy (21, 22, 23, 24). We and others (25, 26) have shown by in situ hybridization studies in rat placenta that, at early times after implantation, only rPLI mRNA is expressed in the primary and secondary giant cells of the placenta. At midpregnancy, rPLII transcription is activated and for a brief time both mRNAs are expressed in the same cells (25). From midpregnancy to term, only rPLII mRNA is expressed in the basal zone giant cells and in newly differentiated giant cells in the labyrinth region of the chorioallantoic placenta (27). A similar developmental switch has been shown for the mouse PLs (28, 29). Where sequence is available for comparison, the 5'-flanking sequences of the homologs of the rat and mouse PLI and PLII genes are highly related, suggesting that common factors may regulate these genes in the two species. There is little similarity, however, between the PLI and PLII 5'-flanking sequences (Ref. 30 and M. L. Duckworth, unpublished data). The expression of the PLI and PLII genes in the same giant cell type provides a useful model system for investigating whether common transcription factors or combinations of factors regulate the expression of these genes in this placental cell type or whether different regulatory factors are used that may reflect their different temporal expression patterns.
The factors important for mouse PLI (mPLI) expression have been studied by transfection assays in the rat choriocarcinoma cell line, Rcho-1 (31). This cell line and the closely related Rcho line (32) differentiate in culture to the giant cell type and are known to express several members of the rat PRL family of placental hormones specific to that cell type (25, 26). Mouse PLI expression has been shown to be dependent on both AP-1 (9) and GATA 2/3 sites (7) contained in a 274-bp 5'-flanking fragment immediately upstream of the transcription start site. Both GATA 2 and 3 are expressed in mouse placental giant cells (7). Although GATA 2 and 3 are also expressed in nonplacental cell types, coexpression of either with a mPLI -274 5'-flanking/Cat reporter construct in mouse L cells, which lack this transcription factor, is sufficient to stimulate reporter gene expression. Targeted disruption of GATA 2/3 in mice leads to a 50% loss of mPLI mRNA, suggesting that although important there are other factors also required for normal expression of this gene (33). We have found that a similar 5'-flanking region of the rat PLI gene is sufficient to regulate reporter gene expression in Rcho cells (34) and conserves the identified GATA and AP-1 sites in essentially identical locations (M. L. Duckworth, unpublished data).
GATA 2/3-deficient mouse embryos die in utero at day 10.5 to 11.5 from massive defects in several organ systems (35, 36), thereby preventing any determination of possible effects on mPLII transcription. We have recently shown that 3 kb of the rPLII gene 5'-flanking region proximal to the transcription start site are sufficient to direct luciferase reporter gene expression in Rcho cells and in the placentas of transgenic mice (30). In this current study, we identify sequences within this rPLII region that act as a placental cell enhancer. Our results suggest that placental-specific expression of PLI and PLII are regulated differently in the same giant cell type.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. Sequences within both the -1729
and -1435 5'-flanking fragments produced significantly higher levels
of luciferase expression (P < 0.05) than the
promoterless control; these levels were, however, significantly lower
than the -3031 fragment itself (P < 0.05), suggesting
the presence of enhancing sequences in the region between -3031 and
-1729. The -1435 fragment showed a slight but significantly higher
activity than the -1729 fragment, suggesting the presence of
inhibitory sequences between -1729 and -1435. A construct in which an
EcoRI fragment (-2838 to -1729) was deleted from the
-3031 5'-fragment produced a further reduction in luciferase
expression, resulting in luciferase levels that were no longer
significantly different from the promoterless vector control. These
results indicated that enhancer sequences important for rPLII
expression were located within this -2838/-1729 EcoRI
fragment and that there were further negative regulatory elements
between -3031 and -2838.
|
Functional Analysis of Sequences between -2838 and -1729
Identifies a Placental-Specific Enhancer Region
To test whether sequences between -2838 and -1729 act as a
placental-specific enhancer, we cloned the EcoRI fragment in
forward and reverse orientations into the luciferase vector, pT81luc
(37), which contains a minimal herpes simplex thymidine kinase (TK)
promoter. Each construct was transfected into Rcho cells grown for
either 6 or 14 days before transfection and into rat pituitary GC
cells. Cultures were assayed for luciferase expression 48 h later.
The results are shown in Fig. 2. Both the
5'
3' and 3'5' orientations showed significant (P
< 0.05) increases in luciferase activity over the minimal TK promoter
in the Rcho cells, but neither orientation was active in the pituitary
GC cells. These transfection data confirm that sequences within the
-2838 to -1729 fragment are important for placental cell expression.
The involvement of DNA sequences in this fragment with the
temporal-specific expression of rPLII is less clear. The enhancing
effect of the -2838/-1729 fragment in the Rcho cells was seen whether
cultures were transfected after 6 days in culture, when rPLII mRNA was
present at only very low levels, or after 14 days when rPLII mRNA was
more highly expressed (25).
|
3' orientation. These constructs were transfected
into Rcho cells grown for 6 days, and cell extracts were analyzed for
luciferase activity 48 h after transfection. The constructs and
results are shown in Fig. 3. The
enhancing activity was localized within a 65-bp
HphI/EcoRI fragment at the 3'-end of the 1109-bp
fragment. There was no significant difference between the luciferase
activities of all the constructs that contained this 3'-sequence.
|
shows an
autoradiogram of a representative DNAse I protection experiment. What
appear to be two separate but adjacent regions were protected in
identical patterns by the placental labyrinth and Rcho nuclear
extracts; nuclear extracts of the GC cell line, in which the enhancing
fragment was inactive, showed distinctly different protection patterns
in these regions. The sequence of the protected areas from
approximately -1765 to -1730 is underlined in Fig. 5.
|
|
). The hypersensitive nucleotide is indicated
by an arrow on the antisense strand in Fig. 5.
Functional Significance of the Ets- and AP-1-Binding Sites
To determine whether the identified Ets- and AP-1-binding sites
were functionally important for placental giant cell expression, we
carried out site-directed mutagenesis of the core sequences of these
sites, in the context of the active 3' StuI/EcoRI
fragment (F7). The sequence of this fragment with the transcription
factor-binding sites is shown in Fig. 5. The nucleotides that were
changed are indicated by asterisks. The PCR primers used to
create the mutants are given in the Materials and Methods.
As well as the protected Ets- and AP-1-binding sites, two sequences
closely related to GATA-binding sites (AGATAT), located within the
context of a direct repeat sequence, were also mutated, even though no
DNAse I protection had been seen with placental/Rcho nuclear extracts
in that region. Our rationale for testing the functionality of these
sites was that GATA 2/3 have been implicated in the expression of
several other placental expressed genes including mPLI (8). When the
two putative GATA sites in the active 151-bp F7 fragment were separated
by an HphI digestion, the majority of the activity remained
in the 3'-F9 fragment, which already suggested that if GATA had a role
in rPLII transcription, both sites were not required. We cloned
fragments containing the individual mutations into pT81luc, transfected
these constructs into Rcho cells, and assayed for luciferase
expression. The results are shown in Fig. 6. Mutation of either the Ets or the AP-1
site significantly affected luciferase gene expression as compared with
the wild-type sequence (P < 0.001), but neither
mutation alone was sufficient to completely eliminate enhancing
activity. Mutation of the Ets core sequence reduced luciferase activity
by approximately 70%; a mutated AP-1 site reduced activity by
approximately 40%. The luciferase activity of the double Ets/AP-1
mutant, however, was reduced to levels comparable to the vector control
and was significantly different from either of the single mutants
(P < 0.001). Luciferase activity of the construct with
mutated GATA sites was not significantly different from the native
fragment (P > 0.05), supporting the DNAse I protection
data that a GATA factor is not necessary for the enhancing activity of
this fragment.
|
). Ets2 was chosen as a representative
of the Ets family of transcription factors since we had detected Ets2,
but not Ets1, mRNA in both placenta and Rcho cells (data not shown).
Ets2 has been implicated in the placental expression of other genes
(41, 42) and is known to interact with AP-1 proteins (43, 44).
Cotransfection of the c-Fos/c-Jun expression vectors with the reporter
construct produced a 2-fold increase in luciferase activity over the
luciferase vector alone, as did cotransfection of the Ets2 expression
clone. When Ets2/c-Fos/c-Jun were cotransfected, there was a further
increase in activity, which appeared to be additive.
|
. Using the labeled 65-bp F9 fragment
as a probe, a specific retarded complex was formed with a mixture of
recombinant c-Fos/c-Jun proteins, which was competed with both a
consensus AP-1-binding site oligonucleotide and Jun-specific antisera
but not by an unrelated oligonucleotide. A 41-bp oligonucleotide
containing mutations in the AP-1 and Ets sites also did not
compete. There was no shift of the fragment with recombinant
c-Jun alone, suggesting that only Fos/Jun dimers were bound.
|
When the F9 fragment was incubated with placental nuclear extracts,
several specific complexes were formed (Fig. 8B). One of these was
competed by both the 65-bp F9 fragment and the consensus AP-1
oligonucleotide, suggesting that it represented an interaction with
c-Fos/c-Jun proteins. This complex was not competed by the
oligonucleotide that contained mutations in the AP-1 and Ets consensus
binding sites. At least one other complex was specifically competed
with the F9 fragment, but not by the oligonucleotide with the mutated
binding sites, suggesting that it may represent a complex containing an
Ets family-binding protein. Neither this nor any other complex,
however, was competed by a consensus Ets2 oligonucleotide. A further
larger complex was specifically competed by both the F9 fragment and
the oligonucleotide containing the mutations, but not by the
nonspecific competitor. The composition of this complex is unclear.
The complexes formed with nuclear extracts from GC cells were markedly
different from those seen with placental extracts (Fig. 8B). Neither
the consensus AP-1 or Ets2 oligonucleotide was able to compete any of
the GC cell complexes even at 500-fold molar excess. Unlike in the
placental extracts, both the native F9 fragment and the oligonucleotide
with the mutated AP-1- and Ets-binding sites partially competed all the
complexes, suggesting that the binding to the F9 fragment in the GC
extracts does not involve the core AP-1- or Ets-binding sites.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Using deletion analysis and transfection assays with luciferase reporter constructs in Rcho cell cultures, we have identified an EcoRI fragment from -2838 to -1729, which when deleted results in a significant decrease in luciferase activity. Consistent with the presence of enhancing sequences within this fragment is its ability to activate luciferase gene expression in either a forward or reverse orientation from the herpes simplex thymidine kinase minimal promoter. Enhancement occurs in Rcho trophoblast cells but not in rat pituitary GC cells, suggesting that specific placental protein factors or complexes are binding to sequences within this fragment. The enhancement occurs whether the Rcho cultures have been grown for 6 days or 14 days before transfection. We have previously noted that our Rcho cultures express rPLI mRNA shortly after plating, essentially as soon as giant cells differentiate in these cultures, but that rPLII mRNA is expressed at only low levels until about 14 days after plating, suggesting that further developmental changes may be occurring in these cultures (25). The fact that both early and late cultures show comparable enhancement of reporter gene expression suggests that we may have identified a placental cell enhancer that is not directly involved in the temporal regulation of the rPLII gene. Nonetheless, since rPLII mRNA is still detectable in RNA blots of the earlier Rcho cultures, we cannot rule out that sufficient amounts of factors are present in these cultures to interact with the rPLII enhancer sequences and bring about increased expression. The role of these sequences for the normal developmental expression of rPLII mRNA will require further investigation.
Detailed deletion analysis of the EcoRI fragment indicates that the enhancing activity is located at -1793 to -1729 on a 3' HphI/EcoRI fragment (F9). DNAse I protection studies identify two adjacent regions between approximately -1765 and -1730 that show similar patterns of protection by nuclear extracts from late-term rat placental labyrinth and Rcho cells. FP1 appears to be protected more by the placental nuclear extracts, while footprint 2 (FP2) is protected more by the Rcho extracts at comparable protein concentrations, suggesting that the cognate factors are present in different proportions in the two nuclear extracts and that they can bind independently to the DNA. FP1 lies over a region that contains a consensus sequence for the Ets family of transcription factors (C/AGGAA/T) (38), while FP2 is related to an AP-1-binding site (39, 40). A distinctive feature of FP1, in the placental cell extracts, is the appearance of a new hypersensitive site on the noncoding strand at the nucleotide immediately 5' of the consensus Ets-binding sequence. Although nuclear extracts from GC cells show protection of the FP1 and FP2 regions, the patterns of protection are different. With GC extracts, new hypersensitive sites appear in the FP2 region and immediately 3' of the region designated FP1, neither of which are seen with the placental or Rcho cell nuclear extracts. Gel mobility shift studies also show that the complexes formed by GC extracts on the F9 fragment are markedly different from those seen with placental nuclear extract. These data strongly suggest that the complexes formed in GC cells are distinct from those in placental cells and, from our transfection studies, inactive.
Electrophoretic mobility shift studies using in vitro transcribed and translated rat c-Fos and c-Jun proteins (45) confirm that the 65-bp F9 fragment binds Fos/Jun dimers but not Jun dimers alone; a consensus AP-1 oligonucleotide-binding site is specifically able to compete this complex as well as a complex formed between the F9 fragment and placental nuclear extract. This latter complex is not competed by an oligonucleotide containing a mutant version of the F9 AP-1-binding site.
Although AP-1 transcription factors have been implicated in the placental expression of several placental genes including the mouse and rat PLI, the involvement of the Ets family of transcription factors in placental gene expression is less well documented. Ets proteins form a large family (now >33 members), which is defined by a highly conserved DNA-binding domain of about 85 amino acids (38). All family members bind DNA elements with the core consensus sequence GGAA/T (38). The specificity of these proteins appears to lie in the sequences that surround this core region, and although there is evidence of overlapping binding specificity, not all identified Ets sites bind all members of the family (46, 47). Ets1 and Ets2 have both been implicated in the placental expression of the matrix metalloprotease genes, stromelysin (MMP3) and collagenase (MMP1), in the human and have been shown to cooperate with AP-1 factors in this role (41, 43). Very recently the targeted mutation of the mouse Ets2 gene was found to have profound effects on early placental development primarily mediated through deficient expression of the MMP9 (gelatinase B) gene, which resulted in failed implantation (42). It was also observed that in the absence of Ets2, mPLI levels were elevated in the very early placenta.
Since Ets2 and Ets1 have been implicated in the expression of several of the MMP genes in placenta and we were able to detect Ets2 (although not Ets1) mRNA in developing rat placenta and Rcho cells (data not shown) we tested these in vitro translated proteins for the ability to bind the 65-bp F9 fragment in gel mobility shift studies. Neither recombinant Ets2 or Ets1 protein binds the F9 fragment; they both shift a consensus Ets2-binding site (48). This consensus Ets2-binding site does not compete any of the complexes formed between placental nuclear extract and the F9 fragment. There is, however, a specific placental complex that is competed by the native sequence but is not competed by an oligonucleotide containing a mutation of the GGA core of the Ets-binding site. These results strongly suggest that Ets2 is not the family member that binds the rPLII enhancer GGA sequence directly in vivo, and the identity of that family member remains to be determined. We cannot rule out, however, that Ets2 could still have a role through protein-protein interactions with another Ets proteins. Such interactions have been previously reported for the human stromelysin gene (43).
The physiological importance of the AP-1 and Ets-binding sites in rPLII placental expression is supported by the effect of mutations of each core sequence on the enhancing activity of the active fragment. Mutation of the Ets site alone reduces activity to approximately 30% of that of the native -1880 to -1729 F7 fragment, while mutation of the AP-1 site reduces luciferase activity to approximately 60% of the nonmutagenized fragment. The double Ets/AP-1 mutant eliminates the enhancing activity of the fragment indicating that both the Ets and AP-1 sites are important in this activity. The fact that neither mutation alone causes complete loss of enhancing activity again suggests that the binding of each protein at these adjacent sites may be independent. Cotransfection of c-Fos/c-Jun or Ets2 expression clones with the F9/luciferase construct in Rcho cells results in an increase in reporter gene expression in both cases. When c-Fos, c-Jun, and Ets2 are co-transfected together, there is a further increase in reporter gene expression that appears to be additive. Given that our gel shift data do not demonstrate binding of Ets2 to the F9 fragment, it appears that its overexpression may be acting indirectly. In this regard, it has been reported that Ets2 binds to and activates the jun-B and c-Fos promoters (49, 50).
Unlike the PLI gene, and some other placental genes, members of the GATA family of transcription factors do not have a role in the enhancing activity of this rPLII fragment. There is no protection by the placental nuclear extracts of two putative GATA sites found in this active fragment that are adjacent and occur within the context of a direct repeat (AGATATGTAGATATGT) at -1799 to -1785. Mutation of both core GATA sequences has a slight but not significant effect (P > 0.05) on luciferase activity, supporting our conclusion from the DNAse I protection data.
Very recently, work on the 5'-flanking portion of the mouse PLII gene
has also identified a region containing activating DNA sequences (51).
This region from -1471 to -1340 has a marked similarity to the
segment of the rPLII gene immediately 3' of our reported sequence and
overlaps our F9 fragment by 27 bp at its 5'-end. A comparison of the
rat and mouse PLII sequences is shown in Fig. 9. It is intriguing that although the
most 5'-region of the mPLII gene, designated region m1 and shown by
mutagenesis studies to contain an enhancing sequence, is highly related
to our FP1 region, it does not contain an Ets-binding site. Gel
mobility shift studies with Rcho extracts showed the formation of
specific complexes on the m1 sequence, but this sequence does not
represent a binding site for a known transcription factor. The more 5'
AP-1-binding site we identified in the rat sequence is not represented
in the activating mouse PLII sequence. The region of the mPLII gene
designated m7, which was also shown to contain enhancing sequences, is
less highly conserved between the rat and mouse, while the activating
m8 and m9 regions are strongly conserved, including the consensus GATA
site, which was found not to bind a GATA factor. It therefore appears
that, in spite of the overall high degree of relatedness between the
PLII genes in these species, somewhat different regulatory mechanisms
have evolved in the rat and mouse, and it will be interesting to
determine the precise identity of the factors responsible for enhancing
the transcription of these two genes.
|
A comparison of DNA sequences and transcription factors that have been reported to be important in placental-specific gene expression, now including the rPLII enhancer, suggests that there may not be a single combination of factors that marks a gene for placental expression. The sequence we have described does not contain an AP2 (TSE) site, a functional GATA site, or a basic helix-loop-helix binding site, all of which have been previously associated with placental gene expression. In particular, the rPLII enhancer is different from the sequences that were identified as important for the expression of the mPLI and the mPLII genes, suggesting that even the same placental cell type may use different combinations of transcription factors for the regulation of expression of different genes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
]ATP were from Mandel Scientific (Guelph, Ontario,
Canada); Bio-Rad protein assay reagent was from Bio-Rad Laboratories
(Mississauga, Ontario, Canada); Qiagen DNA plasmid columns were from
Qiagen Inc. (Mississauga, Ontario, Canada); Kodak XAR film was from
Intersciences Inc. (Markham, Ontario, Canada).
Cell Lines
The rat choriocarcinoma Rcho cell line (32) was kindly provided
by Drs. A. Verstuyf and M. Vandeputte (Rega Institute for Medical
Research, Catholic University of Louvain, Louvain, Belgium). The rat
pituitary GC cell line was a gift from Dr. P. A. Cattini
(Department of Physiology, University of Manitoba, Winnipeg,
Canada).
Clones and Vectors
The rPLII genomic clone, GC I, and some subclones are described
by Shah et al. (30). Other vectors and clones were
generously provided as follows: luciferase vector pXP2 (37), a
cytomegalovirus promoter/luciferase construct (CMVp.luc), and an herpes
simplex thymidine kinase promoter/luciferase vector, pT81luc (37), Dr.
R. J. Matusik, Department of Urological Surgery, Vanderbilt
University, Nashville, TN; a cytomegalovirus promoter/chloramphenicol
transacetylase construct (pcDNA3.cat), Dr. R. P. C. Shiu,
Department of Physiology, University of Manitoba, Winnipeg, Canada;
hEts1 and hEts2 cDNA clones (52), Dr. P. A. Cattini; rat c-Fos and
rat c-Jun (45), Dr. T. Curran, St Jude Children’s Research Hospital,
Memphis, TN.
rPLII 5'-Flanking Plasmid Clones
5'-Flanking plasmid subclones were constructed from an rPLII
-genomic clone, GC I (30). We have shown that a -3031 5'-flanking
fragment proximal to the transcription start site is sufficient to
direct reporter gene expression in the rat trophoblast Rcho cell line
and the placenta of transient transgenic mice. All plasmid DNAs were
prepared from overnight bacterial cultures using Qiagen DNA plasmid
columns according to the supplier’s protocol. DNA sequencing of the
clones was carried out by the dideoxy method (53) using Sequenase or a
PCR sequencing kit and [35S]dATP. All restriction enzyme
fragments were purified by agarose gel electrophoresis followed by
electroelution and ethanol precipitation.
Chimeric rPLII 5'-flanking/luciferase clones containing the native
rPLII promoter were constructed in the vector pXP2. Clones are named
according to the 5'-end of the fragment; the 3'-end of these clones
originates at position +64. The clones -765rPLIIp.luc and
-3031rPLIIp.luc have been previously described (30). Further rPLII
native promoter luciferase clones were constructed as follows: a 4.5-kb
5'-rPLII fragment cloned into the HindIII/BamHI
sites in pBspSK (Stratagene, La Jolla, CA) was digested with
EcoRI/BamHI, and the isolated fragment was
recloned into EcoRI/BamHI cut pBspSK to remove
sequences between -3031 to -1729 and conserve a 5'-HindIII
cloning site; a HindIII/BamHI fragment of this
clone was isolated and ligated into
HindIII/BglII-digested pXP2 to form
-1729rPLIIp.luc. -1729rPLIIp.luc was digested with
BamHI/BglII and religated to form
-1435rPLIIp.luc. A pBspSK clone containing the PvuII
fragment from -765 to +64 was digested with
EcoRV/BamHI and cloned into the
SmaI/BglII site of pXP2 to form
-118-rPLIIp.luc. To produce the clone, 
ErPLIIp.luc, deleted for
-2838 to -1729, the -4.5-kb 5'flanking pBspSK clone, was digested
with EcoRI and religated; this deletion clone was further
digested with SacI/BamHI and cloned into
SacI/BglII-cut pXP2.
The analysis of the -2838 to -1729 EcoRI fragment for
placental-specific enhancing activity in the context of a heterologous
promoter was carried out in the luciferase reporter vector pT81luc,
which contains a minimal (-81 to +52) herpes simplex thymidine kinase
promoter. Clones were constructed as follows: the EcoRI
fragment was cloned into EcoRI-digested pBspSK to produce
E/EpBspSK in both 5'3' and 3'5' directions as determined by
sequencing; this fragment was removed from the 5'3'
(E/EF) clone by HindIII/SmaI
digestion and cloned into HindIII/SmaI-digested
pT81luc to produce E/EFTKpluc (5'3'); the fragment from
the 3'5'pBspSK (E/ER) clone was digested with
BamHI/EcoRV and ligated into
BamHI/SmaI-cut pT81luc to give
E/ERTKpluc. Subfragments of the EcoRI fragment
were ligated into the luciferase reporter vector, pT81luc, as follows,
moving from the 5'- to 3'-end of the fragment: E/EFpBspSK
was digested with HindIII/DraI, and the
5'-HindIII/DraI fragment (F1) was isolated and
ligated to HindIII/SmaI cut vector; an internal
DraI/DraI fragment (F2) was isolated and ligated
to SmaI cut vector. E/EFpBspSK was also digested
with DraI/SacI, and the
3'-DraI/SacI fragment (F3) was isolated and
ligated to SmaI/SacI cut vector. A
HindIII/SacI fragment isolated from the F3 clone
was further digested with BsaBI. The
5'-HindIII/BsaBI fragment (F4) was cloned into
HindIII/SmaI cut vector; the
3'-BsaBI/SacI fragment (F5) was cloned into
SmaI/SacI cut vector. The F5 clone was digested
with HindIII and StuI, and the 5'
HindIII/StuI fragment (F6) was isolated and
ligated into HindIII/SmaI cut vector; the F5
clone was also digested with StuI/SacI, and the
3'-StuI/SacI fragment (F7) was isolated and
ligated into SmaI/SacI cut vector. A
BamHI fragment from F7TKpluc (polylinker sites) was digested
with HphI and blunt-ended with Klenow polymerase. This
fragment was further digested with HindIII and
SacI; the 5'-HindIII/HphI fragment
(F8) was cloned into HindIII/SmaI cut vector; the
3'-HphI/SacI fragment (F9) was cloned into
SmaI/SacI cut vector.
All constructs were sequenced across the newly created boundaries to determine correct ligations.
Cell Culture and Transient Transfection Assays
The rat choriocarcinoma Rcho cell line was grown routinely on
RPMI 1640 medium containing HEPES buffer, supplemented with
heat-inactivated 20% FBS, 1 mM sodium pyruvate, 50
µM ß-mercaptoethanol, 50 U/ml streptomycin, and 50
µg/ml penicillin. Medium was changed every other day, and cells were
split before confluency, every 3 days, using trypsin/EDTA. When grown
for transfections and nuclear extract preparations, cells were changed
after 4 days to NCTC 135 medium with 10% FCS and supplements that
stimulate giant cell differentiation (31). The rat anterior pituitary
GC cell line was grown in DMEM as described by Cattini and Eberhardt
(54).
The Rcho cells were transfected using the calcium phosphate method essentially as described by Vuille et al. (55) at the times indicated. Transfections were routinely carried out in 10-cm dishes using 10 µg of test plasmids and 1 µg of CMVp.cat for determining transfection efficiency. The rat anterior pituitary GC cells were grown to 40–50% confluency and transfected as described by Nickel et al. (56).
Cell extracts were prepared as previously described (30) except that reporter lysis buffer (Promega, Madison, WI) was used. Luciferase assays were carried out immediately after lysate preparation using a Promega Luciferase Assay kit according to supplier’s instructions. Activity was measured in relative light units using a TROPIX luminometer (Bio/Can Scientific, Mississauga, Ontario, Canada). The chloramphenicol acetyltransferase activity was measured by the two-phase fluor diffusion assay as described by Nickel et al. (56). To standardize for variations in plasmid uptake, all luciferase activities were normalized to the CAT assay data for the same sample. Protein determinations were carried out using a Bio-Rad protein assay reagent according to the supplier’s protocol.
Nuclear Extract Preparation
Nuclear extracts were prepared from Rcho and GC cell lines, and
rat placental labyrinth regions collected and dissected at days 14 to
16, according to published protocols (57) with the following
modifications. Rcho cultures were grown for 6 days as described;
remaining small undifferentiated cells were first removed by a short
0.25% trypsin pretreatment before giant cells were collected for
nuclei (15). Frozen placentas were ground to a fine powder in a mortar
and pestle placed on dry ice; placental nuclear extracts were prepared
as for the cell culture extracts with the initial volume being taken as
the volume of powdered tissue. All procedures involving animals were
carried out according to protocols approved by the University of
Manitoba Animal Care Committee.
DNAse I Protection Assays
DNAse I protection assays of the 329-bp rPLII 5'-flanking
BsaBI/EcoRI fragment (F5) described above were
carried out according to standard protocols (58). The F7TKpluc clone
was digested at an XhoI site in the pT81luc polylinker and
treated with calf intestinal phosphatase, and the antisense strand was
end-labeled with T4 polynucleotide kinase and
[32P]ATP. The fragment was released by a
HindIII digestion and purified by agarose gel
electrophoresis, followed by electroelution and ethanol precipitation.
Binding reactions were carried out in a final volume of 20 µl
containing approximately 20,000 cpm (5–10 fmol) of fragment and 0.5
µg dI:dC. Increasing amounts of placental labyrinth and Rcho nuclear
extracts (10–80 µg) or GC nuclear extracts (20 and 40 µg) were
incubated with the probe on ice for 15 min followed by digestion with
0.05 U of DNAse I for 90 sec at 26 C. After phenol/chloroform
extraction and precipitation, the digested products were fractionated
on a denaturing 6% polyacrylamide/urea gel and exposed to Kodak XAR
film for autoradiography.
Electrophoretic Mobility Shift Assays
The 65-bp F9 fragment that had been cloned into the
SmaI/EcoRI sites of pBspSK was cut out with
EcoRI, electroeluted from an agarose gel, and end-labeled
with Klenow fragment and -dATP. Recombinant c-Fos, c-Jun, Ets1, and
Ets2 proteins were synthesized using an in vitro
transcription/translation kit according to the supplier’s protocol.
Nuclear extracts were made as described. Four microliters each of one
or more of the in vitro translated mixtures, 20 µg of rat
placental nuclear extract, or 5 µg of GC extract were used in a
binding reaction. Reactions were carried out in 40 µl of binding
buffer containing 20 mM HEPES buffer, pH 7.9, 100
mM KCl, 0.2 mM EDTA, 0.5 mM
dithiothreitol, 5 mM MgCl2, 1 mM
phenylmethylsulfonylfluoride, 0.5 µg dI:dC. Specific or nonspecific
competitors were added for 15 min before the labeled probe; Jun or Ets2
antisera were added 45 min before the probe; 10 to 20 fmoles of labeled
probe (1–2 x 104 dpm) were incubated for a further
30 min. All binding reactions were at room temperature. Complexes were
separated by electrophoresis on 4% nondenaturing polyacrylamide gels.
Gels were dried and exposed for autoradiography.
Coding strands of specific competitor oligonucleotides used:
AP-1 consensus: CGCTTGATGACTCAGCCGGAA
Ets2 consensus (48): CTAGGACCGGAAGTGGGAGT
mAP-1/mEts: CCAGGGTTATTTctagAAGGGTAAACAttcAGTAGGGCTTG
Site-Directed Mutagenesis
A PCR strategy was used to mutate transcription factor-binding
sites that had been identified by DNAse I protection studies and
sequence analysis. These sites included a consensus Ets-binding site, a
putative AP-1-binding site, and a pair of putative GATA-binding sites
found within a direct repeat sequence. The
StuI/EcoRI fragment that had been shown to retain
activity in transfection assays was cloned into the
SmaI/SacI sites of pBspKS. Three mutagenic
primers were synthesized as follows:
mEts: GCGCGAGCTCGAATTCAAGCCCTACTgaaTGTTTACCCTTGAGCA
mAP-1: GCGCGAGCTCGAATTCAAGCCCTACTTCCTGTTTACCCTTctagAAATAACCCTGGAAATG
mGATA: GCGCGAGCTCGAATTCAAGCCCTACTTCCTGTTTACCCTTGAGCAAATAACCCTGGAAATGCGTAAAACACATcggTACATAcgTTACTCACC
Sequences are shown 5'3'; mEts mutates a putative Ets-binding site,
mAP-1 mutates a putative AP-1 site; and mGATA mutates two putative GATA
sites all shown in lower case. In addition to the mutated
sequences, each primer included a SacI site and a 4-bp GCGC
extension at the 3'-end of the primers (underlined) to
facilitate directional cloning. The M13 reverse primer was used with
the mutagenic primers in PCR reactions. PCR fragments were isolated by
agarose gel electrophoresis and electroelution, digested with
HindIII/SacI, and cloned into these sites in
pBspKS for sequencing to confirm that only the expected mutations had
been introduced. The mutated HindIII/SacI
fragments were cloned into HindIII/SacI-digested
pT81luc for testing in transfection assays. The double Ets/AP-1 mutant
was generated by PCR as described using the mAP-1 primer with the mEts
pBsp clone. PCR product was cloned into pBspKS as described above for
the single mutants to facilitate sequencing, followed by cloning of the
correct fragment into pT81luc.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported by grants from the Medical Research Council of Canada, the Manitoba Medical Services Foundation, and the Manitoba Health Research Council.
Received for publication March 3, 1998. Revision received October 2, 1998. Accepted for publication November 11, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-
and ß-subunit promoters by AP-2. J Biol Chem 272:15405–15412- and ß-subunit genes of human chorionic
gonadotropin by trophoblast-specific element-binding protein. Mol
Endocrinol 7:1579–1588-subunit glycoprotein hormone promoter with trophoblast specificity
and maximal activity. Mol Endocrinol 11:1669–1680-subunit gene in the placenta and
pituitary gland. Mol Cell Biol 14:5592–5602 on the
expression of placental lactogens I and II mRNAs in a rat
choriocarcinoma cell line. Endocrine J 45:297–306[CrossRef]
This article has been cited by other articles:
 |
|
 |
|
  T. Sakurai, A. Sakamoto, Y. Muroi, H. Bai, K. Nagaoka, K. Tamura, T. Takahashi, K. Hashizume, M. Sakatani, M. Takahashi, et al. Induction of Endogenous Interferon Tau Gene Transcription by CDX2 and High Acetylation in Bovine Nontrophoblast Cells Biol Reprod, June 1, 2009; 80(6): 1223 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Das, T. Ezashi, R. Gupta, and R. M. Roberts Combinatorial Roles of Protein Kinase A, Ets2, and 3',5'-Cyclic-Adenosine Monophosphate Response Element-Binding Protein-Binding Protein/p300 in the Transcriptional Control of Interferon-{tau} Expression in a Trophoblast Cell Line Mol. Endocrinol., February 1, 2008; 22(2): 331 - 343. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. A. Winger, J. Guttormsen, H. Gavin, and F. Bhushan Heat Shock Protein 1 and the Mitogen-Activated Protein Kinase 14 Pathway Are Important for Mouse Trophoblast Stem Cell Differentiation Biol Reprod, May 1, 2007; 76(5): 884 - 891. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ozturk, L. J. Donald, L. Li, H. W. Duckworth, and M. L. Duckworth Proteomic Identification of AP2{gamma} as a Rat Placental Lactogen II Trophoblast Cell-Specific Enhancer Binding Protein Endocrinology, September 1, 2006; 147(9): 4319 - 4329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Ghosh, S. Sachdev, M. Hannink, and R. M. Roberts Coordinate Regulation of Basal and Cyclic 5'-Adenosine Monophosphate (cAMP)-Activated Expression of Human Chorionic Gonadotropin-{alpha} by Ets-2 and cAMP-Responsive Element Binding Protein Mol. Endocrinol., April 1, 2005; 19(4): 1049 - 1066. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ezashi and R. M. Roberts Regulation of Interferon-{tau} (IFN-{tau}) Gene Promoters by Growth Factors that Target the Ets-2 Composite Enhancer: A Possible Model for Maternal Control of IFN-{tau} Production by the Conceptus during Early Pregnancy Endocrinology, October 1, 2004; 145(10): 4452 - 4460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ozturk, A. Fresnoza, A. Savoie, H. W. Duckworth, and M. L. Duckworth Defining Regulatory Regions in the Rat Prolactin Gene Family Locus Using a Large P1 Genomic Clone Endocrinology, November 1, 2003; 144(11): 4742 - 4754. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kamei, S. R. Jones, B. M. Chapman, K. L. MCGonigle, G. Dai, and M. J. Soares The Phosphatidylinositol 3-Kinase/Akt Signaling Pathway Modulates the Endocrine Differentiation of Trophoblast Cells Mol. Endocrinol., July 1, 2002; 16(7): 1469 - 1481. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Cross, L. Anson-Cartwright, and I. C. Scott Transcription Factors Underlying the Development and Endocrine Functions of the Placenta Recent Prog. Horm. Res., January 1, 2002; 57(1): 221 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ezashi, D. Ghosh, and R. M. Roberts Repression of Ets-2-Induced Transactivation of the Tau Interferon Promoter by Oct-4 Mol. Cell. Biol., December 1, 2001; 21(23): 7883 - 7891. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-H. Cho, H. Kimura, T. Minami, J. Ohgane, N. Hattori, S. Tanaka, and K. Shiota DNA Methylation Regulates Placental Lactogen I Gene Expression Endocrinology, August 1, 2001; 142(8): 3389 - 3396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. W. Cheng, B. K. C. Chow, and P. C. K. Leung Functional Mapping of a Placenta-Specific Upstream Promoter for Human Gonadotropin-Releasing Hormone Receptor Gene Endocrinology, April 1, 2001; 142(4): 1506 - 1516. [Abstract] [Full Text]
|
|
|
|
|
|
C. R. Wooton-Kee and B. J. Clark Steroidogenic Factor-1 Influences Protein-Deoxyribonucleic Acid Interactions within the Cyclic Adenosine 3',5'-Monophosphate-Responsive Regions of the Murine Steroidogenic Acute Regulatory Protein Gene Endocrinology, April 1, 2000; 141(4): 1345 - 1355. [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 |
