| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
Department of Obstetrics and Gynecology (T.G.G.), University of Wisconsin Medical School; and Wisconsin Regional Primate Research Center (J.T.S., C.M.C., M.D., J.M.F., T.G.G.), University of Wisconsin, Madison, Wisconsin 53715-1299
Address correspondence and requests for reprints to: Dr. Thaddeus G. Golos, Ph.D., Wisconsin Regional Primate Research Center, 1223 Capitol Court, University of Wisconsin, Madison, Wisconsin 53715-1299. E-mail: golos{at}primate.wisc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4 variant was cloned but coded for
a truncated and presumably nonfunctional protein. Only by using a
nested RT-PCR approach were Pit-1 and Pit-1ß variants consistently
detectable in both human and rhesus placental tissue. The Pit-1ß
variant mRNA was not detectable in JEG-3 choriocarcinoma cells unless
the cells were stimulated with 8-Br-cAMP. Immunoblot studies with
nuclear extracts from primary rhesus syncytiotrophoblast cultures or
JEG-3 choriocarcinoma cells indicated that although mRNA levels were
very low, Pit-1 protein was detectable in differentiated
cytotrophoblasts, and levels increased after treatment with 8-Br-cAMP.
Two major species of Pit-1 protein were detected that corresponded to
the two major bands in rat pituitary GH3 cell nuclear extracts. Low
levels of slightly larger bands also were seen, which may represent
Pit-1ß protein or phosphorylated species. We conclude that Pit-1
splice variants expressed in the primate pituitary gland differ from
those in the rodent gland and that the Pit-1 and Pit-1ß mRNAs
expressed in the placenta give rise to a pattern of protein expression
similar to that seen in pituitary cells, which is inducible by
treatment with 8-Br-cAMP. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The primate GH gene locus is unique among vertebrates in that it is a multigene cluster with divergent tissue-specific expression of its members (12, 13). The transcription of pituitary GH is largely restricted to the somatotrope of the anterior pituitary, whereas the chorionic somatomammotropin genes and the GH variant gene (GH-V) are expressed in the placental syncytiotrophoblast (12, 13, 14, 15). The molecular control of the placental members of this gene cluster remains incompletely understood. The expression of Pit-1 in the human (16) and the rat (17) placenta has been reported recently, and its localization to the syncytiotrophoblast in the human placenta suggests that Pit-1 may play a role in placental hormone gene transcription. Alternatively, other genes may be POU protein targets in the placenta. The potential function of Pit-1 in the trophoblast may depend on the specific variant expressed, because Pit-1 splice variants have been shown to code for proteins with differential activation of target genes and divergent interactions with other transcription factors (18, 19, 20, 21, 22, 23, 24). Neither the splice variants expressed in the placenta nor the relative levels of Pit-1 protein in trophoblasts have been reported previously. Additionally, although the messenger RNAs (mRNAs) of the GH locus expressed in the rhesus monkey have been characterized (25), there is essentially no information available on the expression of the Pit-1 gene in nonhuman primates.
Thus, we undertook the present study to define the Pit-1 variant mRNAs
expressed in the nonhuman primate pituitary gland, characterize the
mRNAs of the human and rhesus monkey placenta, and determine whether
the putative protein products of the two major splice variants can be
detected in this tissue. Our results demonstrate that Pit-1 and
Pit-1ß are abundant in the rhesus pituitary gland, whereas the Pit-1T
variant previously described in mouse thyrotrope tumor cells (23) was
not found in rhesus tissue. However, a potentially functional variant
with a novel splice pattern (Pit-1
) was identified. Whereas very low
levels of Pit-1 and Pit-1ß mRNA variants are expressed in rhesus and
human trophoblasts, Pit-1 protein levels are readily detectable and are
increased by treatment with 8-Br-cAMP, implying dynamic tropic
regulation of this gene in the placenta.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All procedures involving the use of animals were approved by the University of Wisconsin Graduate School Animal Care and Use Committee and were conducted in accordance with the NIH Guide for the Care and Use of Experimental Animals. Pituitary glands obtained at necropsy from healthy rhesus monkeys or placental tissue obtained at cesarean section was processed immediately for mRNA isolation. Placental tissue was dispersed also with trypsin and deoxyribonuclease, as previously described (26), to obtain cytotrophoblasts for cell culture. A highly purified fraction of cytotrophoblasts was obtained by Percoll gradient centrifugation, and cells were plated for RNA isolation or Western blot experiments. Typically, freshly isolated cytotrophoblasts were plated in 60-mm dishes at 2 x 106 cells/dish in DMEM supplemented with 10% FCS, as previously described (26). JEG-3 choriocarcinoma cells also were maintained in this medium. Rat GH3 cells were maintained in MEM supplemented with 15% FCS. Some trophoblast cultures were treated with 1–5 mmol/L 8-Br-cAMP for 48 h. This dose and treatment time are based on previous analysis of chorionic somatomammotropin and GH-V mRNA expression and gene transcription in our laboratory (26, 27).
Complementary DNAs (cDNAs) library screening
Rhesus monkey pituitary and placental cDNA libraries were
prepared by Stratagene (Thousand Oaks, CA) with mRNA isolated from
pituitaries of three castrated adult female rhesus or from midpregnancy
total placental mRNA and have been previously described (25).
Approximately 1,000,000 plaques were blotted onto Nytran filter circles
(Schliecher and Schuell, Keene, NH) according to the manufacturers
instructions. The pituitary cDNA library was screened by previously
described methods (25) with a 32P-labeled 170-bp cDNA for
the rhesus Pit-1 POU-specific domain. The cDNA was amplified from
rhesus pituitary RNA with primers based on the sequence of the bovine
mRNA. Sequencing reactions used supercoiled plasmid DNA as sequencing
template, incorporating deoxyadenosine 5'-
-[
35S]thiophosphate with the chain termination method, using
standard procedures, or were sequenced on an ABI Model 373 automated
sequencer (ABI, Foster City, CA). All DNA was sequenced in both
directions. DNA sequence analyses were accomplished using the
University of Wisconsin Genetics Computer Group Sequence Analysis
Software (28).
RT-PCR analysis
RNA from rhesus pituitary gland and placenta and human placenta
and JEG-3 choriocarcinoma cells were analyzed by RT-PCR with primers
based on the rhesus monkey Pit-1 mRNA sequence. The sequences of the
primers are given in Table 1
. A map of the location of
the primers is shown in Fig. 1A. The outside primers are
located in the 5' untranslated (UT) region and just upstream of the
termination codon. The two nested primer pairs flank the first/second
exon-exon junction (mPIT1/mPIT2) or lie in the third and sixth exons
(mPIT3/4), assuming a genomic structure similar to the human, which has
been reported to be identical to the mouse (9, 20, 29).
|
|
Protein electrophoresis and Western analysis
Nuclear extracts were prepared by a modification (30) of a previously described method (31) from GH3 rat somatotroph cells, JEG-3 human choriocarcinoma cells, or primary rhesus syncytiotrophoblast cultures. Rhesus cytotrophoblasts were prepared from fresh placental tissue as previously described (26) and were allowed to differentiate in culture for 48 h. Replicate cultures were untreated or treated with 1.5 mmol/L 8-bromo-cAMP for 48 h before nuclear extract preparation. For immunoblot analysis, 5–50 µg nuclear proteins were fractionated on a 12% SDS-polyacrylamide gel and electrophoretically transferred to a nitrocellulose membrane. All reagents for immunoblots were from Amersham (Arlington Heights, IL). After transfer, the membrane was blocked with 1% BSA and 5% blocking reagent for 1 h and incubated with rabbit anti-Pit-1 antibody (a kind gift of Drs. Simon Rhodes and Michael G. Rosenfeld) or an antibody to the Pit-1 ß/T splice variant (23) (Babco, Berkeley, CA) for 1 h, followed by goat antirabbit antiserum coupled to horseradish peroxidase (Amersham). The blot was developed using the ECL chemiluminescence system (Amersham).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To logically proceed with investigation of placental splice
variants in both human and nonhuman primates, we first needed to verify
the sequences of rhesus monkey Pit-1 expressed in the pituitary gland.
Screening of 106 plaques from a rhesus pituitary cDNA
library with a cDNA probe that encompassed the POU-specific domain of
Pit-1 yielded four verified clones. Two of the inserts were fully
sequenced in both directions. One clone (X8) was found to be homologous
to Pit-1, the other (Y7) represented the Pit-1ß splice variant (Fig. 2). The two clones had very long 3' UT regions of
approximately 1.1 and 1.5 kb, respectively. Except for the 78
nucleotide Pit-1 ß insertion, the sequences of the two clones were
identical. The nucleotide and predicted amino acid sequences from clone
X8 were compared with those of the bovine, rat, and human Pit-1. Table 2 indicates how the homology breaks down among the
coding sequences, the 3' UT region, and the POU domain. The human mRNA
is most highly homologous to the rhesus mRNA, with an overall homology
of 
98%. The amino acid sequence is nearly identical, with only
three differences (in the transactivation domain) out of 292 amino
acids. The nucleotide and amino acid sequences of bovine and rat Pit-1
exhibit lower homology; however, the amino acid homology in the POU
domain is higher than the overall protein homology. The nucleotide
homologies of the rat and bovine UT regions are substantially lower
than that of the regions coding for the protein.
|
|
To obtain some indication of the relative levels of the Pit-1 and
Pit-1ß splice variants, as well as to explore whether additional
variants are present in the rhesus pituitary gland, we used RT-PCR with
primer pairs (Fig. 1
) flanking the predicted exon 1/2 junction, and
primers flanking the fourth exon, which is deleted in a minor variant
in the rat pituitary gland (20, 21). One major and one very minor band
were detected in pituitary RNA with primers that hybridize with the
first and second exons, and one band was detected with the primers
flanking the fourth exon (Fig. 3A). All bands were
cloned and sequenced. With the 5' exons, the most abundant band was the
Pit-1 form initially cloned by Ingraham et al (1) and Bodner
et al. (2); a relatively low abundance Pit-1ß variant
(18, 19, 20) was also consistently detectable. However, the
thyrotrope-specific Pit-1T variant identified in mouse thyrotrope tumor
cell lines or hypothyroid pituitary glands (22) was not detected. The
rat Pit-14 variant (with the fourth exon spliced out) was not
detected in rhesus pituitary RNA, but a rare cDNA, similar (although
not identical) to the Pit-1 4 variant, was detected in rhesus
pituitary gland mRNA (one clone identified of >36 sequenced).
Additionally, the fourth exon was not precisely spliced out, as
reported for the rat; rather, nucleotides at the 3'end of exon 3 were
also deleted, which resulted in a frame shift and premature termination
of translation at the 3'-end of exon 5 (see Fig. 2). The resulting
frame-shifted amino acid sequence was unrelated to Pit-1, and the
truncated protein would lack the POU domain and be unable to bind to
DNA and transactivate gene expression.
|
demonstrates that several other
bands, as well as those corresponding with Pit-1 and Pit-1ß, were
identified with nested PCR. The indicated bands were subcloned and
sequenced, and of 14 cDNAs, 6 represented Pit-1, 6 represented
Pit-1ß, and two represented novel transcripts, neither of which were
homologous to Pit-1T. One cDNA, 1319–6, contained a novel splice
acceptor site, partway between the Pit-1ß and predicted Pit-1T splice
acceptor sites, and would code for a prematurely truncated protein
caused by a frame shift at the splice acceptor (Pit-1 stop in Fig. 3C).
The other cDNA, 1319–7, was unusual in that it used a novel splice
donor, seemingly derived from the 5'-end of the first intron and the
Pit-1 exon II splice acceptor site (Pit-1 in Fig. 3, C and D). This
conclusion was based on the near identity of the novel nucleotide
sequence with the first intron of the human Pit-1 gene (unpublished DNA
sequences: Tatsumi et al., GenBank accession number D12887;
Delhase et al., accession number X77223). This alternatively
spliced mRNA retains the Pit-1 reading frame and would code for a
protein with a novel 23-amino acid insertion unrelated to the Pit-1ß
or Pit-1T inserts (Fig. 3C). No additional exon II variants were
identified with nested PCR.
Having identified multiple Pit-1 splice variants in the primate
pituitary, we wished to determine whether the placenta may likewise
express Pit-1 mRNA variants. Screening of a rhesus placental cDNA
library, with either the initial PCR-amplified 170-bp fragment or the
full-length Pit-1 cDNA, failed to detect any positive clones. We used
the primers indicated above to screen human placental RNA, rhesus
placental RNA, and the human JEG-3 choriocarcinoma cell line for Pit-1
mRNA by a single round of RT-PCR. We were successful only sporadically
at identifying Pit-1-related cDNAs from placental cells or tissues,
indicating that the mRNA may be expressed at very low levels, which is
in agreement with our failure to isolate a cDNA clone from the
placental library. We thus turned to a nested amplification scheme, as
indicated in Fig. 1, and amplified the Pit-1 and Pit-1ß splice
variants identified in pituitary mRNA (Fig. 4A). The
Pit-14 variant was not detected in placental RNA (not shown).
Interestingly, whereas human and rhesus placentas had detectable
Pit-1ß mRNA, this band was not visible with RT-PCR analysis of RNA
from JEG-3 choriocarcinoma cells, even after a 2-week exposure of the
Southern blot (Fig. 4B and data not shown).
|
demonstrates
that Pit-1ß mRNA was readily detectable in JEG-3 choriocarcinoma
cells after treatment with 8-Br-cAMP. Given this novel result, it was
important to determine whether Pit-1 variant mRNAs are translated in
trophoblast cells. We conducted immunoblot analysis of Pit-1 in nuclear
extracts from rat GH3 cells and rhesus and human trophoblasts. Two
major Pit-1 immunoreactive bands were readily visible with 5 µg GH3
extract (Fig. 6A, lane 1). These bands, which arise from
alternative translation initiation codon usage of the Pit-1 mRNA (34),
previously have been referred to as 31 and 33 kDa (34), although their
relative mobility is slightly slower in our hands. Western blots also
clearly showed similar immunoreactive bands in nuclear extracts from
both human and rhesus trophoblasts (Fig. 6A, lanes 3–6), although the
maximal expression of Pit-1 proteins in trophoblasts (50 µg JEG-3
extract, lane 6) was approximately 10-fold lower than that in GH3 cells
(5 µg GH3 extract, lane 1). Although Pit-1 protein was not readily
detectable in some isolated trophoblast samples (lane 2), most
immunoblots clearly showed 31 and 33 kDa Pit-1 proteins in primary
trophoblasts allowed to differentiate in culture to terminally
differentiated syncytiotrophoblasts (lanes 3 and 5). Pit-1 bands were
readily detected, as well, in JEG-3 choriocarcinoma cells (lanes 4 and
6), although their relative distribution seemed to differ from that in
primary cultures, particularly in untreated cells (lanes 3 and 4).
Cells treated with 8-Br-cAMP for 48 h, in particular JEG-3 cells,
showed an increase in the amount of the 31- and 33-kDa proteins (lanes
5 and 6). Treatment with 8-Br-cAMP also seemed to result in an increase
in several minor higher molecular mass bands. It is possible that these
represent Pit-1ß variant protein. In our hands, however, an
anti-Pit-1ß/T antiserum (23) did not specifically recognize any
nuclear extract proteins (not shown). Panel B, which presents equal
amounts of each extract, silver-stained to examine the range and
distribution of nuclear proteins, demonstrates that there were no
significant qualitative or quantitative differences among the extracts.
Thus, differences in levels of immunoreactive Pit-1 between samples are
unlikely to be caused by 8-Br-cAMP-induced changes in the overall
pattern of nuclear proteins.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We examined the Pit-1 mRNA expressed in the rhesus monkey pituitary gland and found that the pattern of Pit-1 mRNA expression in the nonhuman primate seems to differ from that in the rodent. Although Pit-1 and Pit-1ß were readily detected in rhesus pituitary glands, a Pit-1T variant was not found, even with nested PCR. It may be that a Pit-1T mRNA is expressed in vanishingly small amounts in the intact gland and can be detected only in transformed cell lines representing thyrotropes (23). Alternatively, it could be that there are significant differences in the splicing patterns of the Pit-1 gene in primate thyrotropes vs. rodents. Pit-1 protein and mRNA have been identified in human pituitary adenomas expressing TSH (35, 36, 37); however, a human Pit-1T transcript has not been identified at this time.
Other novel splice variants were found at the Exon I/II junction. One
of the variant mRNAs detected in rhesus pituitary, which we have
referred to as Pit-1, used a unique splice donor and the previously
described Exon II Pit-1 splice acceptor and would code for a protein
with a novel insertion in the transactivation domain. Because Pit-1,
Pit-1ß, and Pit-1T have distinct transactivation potentials for
selected DNA recognition elements in the GH, PRL, and TSHß genes (18, 20, 23, 24), it is tempting to speculate that the Pit-1 variant
likewise may have a novel biological activity. This hypothesis remains
to be directly explored.
One cDNA, with a deletion similar but not identical to the Pit-14
variant (21, 22), was cloned by RT-PCR from rhesus pituitary gland
mRNA. This mRNA would result in a predicted protein with 19 amino acids
at the carboxy-terminus that were unrelated to Pit-1, followed by a
premature stop codon. Because this would delete both the POU-specific
and POU-homeodomains, which are critical for DNA binding, the mRNA
would not be expected to code for a functional POU transcription factor
(4, 5), and its significance remains unclear.
The expression of Pit-1 has been demonstrated recently in the human placenta (16); however, it was not reported whether this tissue expresses multiple splice variants, as does the pituitary gland. Based on our experience with rhesus pituitary Pit-1, we examined Pit-1 mRNA expression in the rhesus and human placenta and noted several differences with pituitary expression. First, the relative levels of the Pit-1 mRNA in both the rhesus and human placenta are dramatically lower than those in the pituitary gland. In addition, Pit-1 and Pit-1ß mRNAs were the only variants detected in placental tissues. Pit-1 protein was detectable also in trophoblasts, although levels were relatively low compared with GH3 cells, and the levels of immunoreactive Pit-1 proteins also increased in cAMP-treated trophoblasts. Because higher molecular mass bands are among those increased by cAMP, it is possible that these represent Pit-1 phosphorylated species, based on previous observations in pituitary cell cultures (38). Although the functional role of phosphorylation of the Pit-1 protein recently has been called into question (39, 40), it has been shown clearly that Pit-1 itself and CREB are involved in the upregulation of Pit-1 gene transcription by cAMP-dependent protein kinase pathways (41, 42). The Pit-1ß variant mRNA was detectable in JEG-3 cells only after treatment with 8-Br-cAMP, a well-characterized regulator of trophoblast endocrine activity (32, 33). However, this may not necessarily represent a selective increase of this variant in response to tropic stimulation but may simply represent an increase in overall Pit-1 transcription over the low levels in unstimulated cells. This view is supported by the increase in all Pit-1 protein species in JEG-3 cells upon stimulation by 8-Br-cAMP.
Although we were able to demonstrate that several variant Pit-1 mRNAs and proteins are expressed in rhesus, as well as human, placental cells, their functional role in transcription in trophoblasts remains unclear. First, Shepard et al. (43) have reported that mutagenesis of either the proximal or distal Pit-1-like elements in the 5'-flanking DNA of the hCS-A gene had no effect on transcriptional activation or cAMP-responsiveness in JEG-3 choriocarcinoma cells. Consistent with these observations, we have not found any significant effect of mutation of the proximal or distal Pit-1 binding sites on transcriptional activation of the rhesus GH-V in primary syncytiotrophoblast cultures, nor were we able to detect Pit-1 binding to the monkey GH-V distal Pit-1 element (Schanke et al, unpublished data). However, because Pit-1 interacts with additional factors to confer developmentally regulated and tissue-specific expression of GH and PRL in the pituitary (44, 45), it may be that other elements can sustain transcription of placental hormone genes in the absence of functional Pit-1 binding sites. Alternatively, these studies presume a role in the placenta analogous to the role controlling hormone gene transcription in the pituitary gland, and it is possible that Pit-1 subserves an entirely different function in the placenta, such as a developmental role in trophoblast differentiation not related to transcription of the GH gene cluster.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Pediatrics, University of
California- San Diego School of Medicine, San Diego, California
92093-5003.
Received September 25, 1996.
Revised November 4, 1996.
Accepted November 4, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Seoane and R. Perez-Fernandez The Vitamin D Receptor Represses Transcription of the Pituitary Transcription Factor Pit-1 Gene without Involvement of the Retinoid X Receptor Mol. Endocrinol., April 1, 2006; 20(4): 735 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Gil-Puig, S Seoane, M Blanco, M Macia, T Garcia-Caballero, C Segura, and R Perez-Fernandez Pit-1 is expressed in normal and tumorous human breast and regulates GH secretion and cell proliferation Eur. J. Endocrinol., August 1, 2005; 153(2): 335 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Yusta, E. T. Alarid, D. F. Gordon, E. C. Ridgway, and P. L. Mellon The Thyrotropin {beta}-Subunit Gene Is Repressed by Thyroid Hormone in a Novel Thyrotrope Cell Line, Mouse T{alpha}T1 Cells Endocrinology, November 1, 1998; 139(11): 4476 - 4482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Schanke, M. Durning, K. J. Johnson, L. K. Bennett, and T. G. Golos SP1/SP3-Binding Sites and Adjacent Elements Contribute to Basal and Cyclic Adenosine 3',5'-Monophosphate-Stimulated Transcriptional Activation of the Rhesus Growth Hormone-Variant Gene in Trophoblasts Mol. Endocrinol., March 1, 1998; 12(3): 405 - 417. [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 |
