| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Center for Biomedical Research, Population Council and The Rockefeller University, New York, New York 10021
Address all correspondence and requests for reprints to: Daniel J. Bernard, Ph.D, Center for Biomedical Research, Population Council, 1230 York Avenue, New York, New York 10021. E-mail: dbernard{at}popcbr.rockefeller.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Although this signal transduction pathway has been described in various systems, only recently have the specific mechanisms through which activins stimulate FSH synthesis begun to come to light. The murine LßT2 cell line, derived from targeted oncogenesis in transgenic mice (16, 17), has emerged as a powerful model system in which to dissect the signaling pathways regulating gonadotropin synthesis (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33). Recently, LßT2 cells were shown to produce and secrete FSH in response to exogenous activin A (a dimer of two inhibin ßA subunits) (27, 29, 30, 31, 32). Moreover, treatment of these cells with follistatin, an activin [and bone morphogenetic protein (BMP)]-bioneutralizing protein, blocked expression of the ß subunit of FSH (FSHß), suggesting that endogenously produced activins may stimulate FSH production in these cells (32), albeit at significantly lower levels than observed in the adult pituitary. LßT2 cells express the inhibin ßB, but not ßA, subunit, suggesting that endogenous activin B (ßB-ßB) may stimulate FSHß transcription in these cells as it appears to in the pituitary gland (32, 34, 35, 36) (c.f. Ref. 30).
Exogenous activin A stimulates both ovine and rat FSHß promoter activity in LßT2 cells (30, 31, 32, 33), demonstrating that all of the signaling components required for activin-stimulated FSHß gene transcription are represented in this cell line. In fact, ActRIIA, ActRIIB, ALK4, SMAD2, SMAD3, and SMAD4 are all expressed in LßT2 cells (31, 32, 33, 37). Interestingly, another gonadotrope cell line, 
T3–1, also expresses these components but does not produce FSH in response to activins (32), suggesting that additional coactivator proteins are expressed in LßT2 cells or perhaps that repressor proteins are expressed in T3–1 cells.
The production of FSHß is the rate-limiting step in FSH synthesis. T3–1 cells produce the gonadotropin subunit but do not express the FSHß subunit, nor do they secrete FSH. LßT2 cells produce the FSHß subunit and secrete FSH at very low levels under basal conditions. Following exogenous activin A treatment, LßT2, but not T3–1, cells increase FSHß expression followed by an increase in FSH secretion (29, 31, 32). The characterization of the mechanisms through which the activins stimulate FSHß gene expression is therefore fundamental to understanding the regulation of FSH production.
Although exogenous activin A stimulates both endogenous FSHß expression and FSHß promoter- reporter activity, the signaling mechanisms mediating these effects have only recently been the subjects of investigation. One report demonstrated that transient transfection of SMAD3, but not SMAD2, stimulated rat FSHß promoter activity in LßT2 cells (33), whereas the results of a second study suggested that SMAD2 may mediate activin A’s stimulation of endogenous FSHß expression and FSH secretion (27). The present paper examined the effects of exogenous activins on SMAD2 and SMAD3 phosphorylation and nuclear translocation, as well as the role of SMADs in stimulation of FSHß expression. The data show that both SMAD2 and SMAD3 are necessary components of the signal transduction pathway through which activins stimulate FSHß transcription.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
).
|
), suggesting that FSHß may be an immediate early response gene to activins in LßT2 cells. FSHß mRNA levels continued to increase through 6 h of activin A treatment and remained elevated for at least 24 h (data not shown).
Activin A Stimulates Rapid Increases in FSHß Primary Transcript Levels
FSHß promoter-reporter studies clearly indicate that activin A can stimulate FSHß transcription in LßT2 cells (31, 32, 33); however, given the nature of these investigations, it is not clear whether or not this represents an immediate early transcriptional response. That is, activins may first stimulate the production of a protein that then acts to increase FSHß transcription. To address this issue, we first measured activin A-stimulated increases in FSHß primary transcript (PT) levels by semiquantitative RT-PCR (35, 38). (Note: Given its greater availability, activin A was used in subsequent analyses.) In response to 30 ng/ml activin A, FSHß PT levels increased within 30 min and continued to increase throughout the 120-min treatment period (Fig. 2A). These data were consistent with the hypothesis that activins stimulate an immediate (direct) increase in FSHß transcription in LßT2 cells.
|
). This antagonism did not appear to be a consequence of altered receptor binding or activity as cells treated with activin A showed potent SMAD2 phosphorylation both in the presence or absence of actinomycin-D pretreatment (Fig. 2C). Therefore, an activin-dependent signal transduction mechanism appears to directly stimulate transcription of the FSHß gene.
ALK4- and Activin-Mediated Stimulation of FSHß Expression Are Blocked by SMAD7
In several other systems, activins bind to ActRIIA or ActRIIB, which then recruit and trans-phosphorylate the activin type IB receptor, ALK4. The activated ALK4 then phosphorylates the intracellular signaling proteins SMAD2 and SMAD3. To determine whether or not ALK4 may be involved activin-stimulated FSHß expression, LßT2 cells were transfected with a constitutively active form of ALK4 and its effects on FSHß mRNA levels were examined.
A single amino acid substitution, T206D (hereafter ALK4TD), in ALK4 enables the receptor to generate intracellular signals in the absence of ligand and/or the type II receptors (39). LßT2 cells transfected with ALK4TD had higher FSHß mRNA levels than cells transfected with an empty expression vector (pcDNA3.0) (Fig. 3A). Although these data did not show definitively that activins signal through ALK4, they did indicate that signaling events downstream of ALK4 were sufficient to stimulate FSHß expression in these cells.
|
), suggesting that SMAD2 and/or SMAD3 are part of the mechanism through which ALK4 stimulates FSHß expression.
We next wished to assess whether SMAD7 might similarly antagonize activin A-stimulated FSHß expression. Unfortunately, the low transfection efficiency we observed in LßT2 cells precluded our ability to use transient transfection (i.e. 10% of cells) to examine antagonism of activin A’s effects on endogenous FSHß mRNA levels (i.e. measured from 100% of the cells). That is, even if SMAD7 antagonized activin signaling in the small proportion of transfected cells, unperturbed signaling in the majority of (nontransfected) cells might overshadow this effect. We therefore needed a way to measure FSHß expression in only those cells that were transfected. To this end, we cloned approximately 2 kb of the mouse FSHß 5' flanking region (-1990 to +1) and ligated it upstream of a luciferase reporter gene in pGL3-Basic. When transfected into LßT2 cells, -1990/+1mFSHß-luc was stimulated about 6-fold by 30 ng/ml activin A (Fig. 3B). The empty pGL3-Basic vector showed no stimulation upon activin A treatment. Similar to what has been reported previously (32), there was also cell-type specificity in responsiveness of the promoter construct in that -1990/+1mFSHß-luc was not stimulated by activin A in T3–1 cells (Fig. 3C).
We transfected LßT2 cells with -1990/+1mFSHß-luc in the presence or absence of a SMAD7 expression vector and then treated cells with 30 ng/ml activin A for 24 h. SMAD7 blocked activin-stimulated transcription but did not significantly affect basal activity of the promoter (Fig. 3D). These data further support the hypothesis that SMAD2 and/or SMAD3 are necessary for both activin- and ALK4-stimulated FSHß transcription.
SMAD2 and SMAD3 Are Rapidly Phosphorylated after Activin A Treatment
Given that FSHß expression appears to be dependent upon SMAD2 and/or SMAD3 activation, the effect of activin A on SMAD2 and SMAD3 phosphorylation was examined. LßT2 and T3–1 cells were treated with activin A (30 ng/ml) for 0, 5, 10, 15, 30, or 60 min and cytoplasmic and nuclear protein lysates prepared. In nuclear fractions, increases in phospho-SMAD2 and phospho-SMAD3 were detected 5–10 min after activin A treatment in both cell lines (Fig. 4, A and B). Levels of both phospho-proteins continued to increase through the 60-min treatment period. Neither phospho-SMAD2 nor phospho-SMAD3 was detected in cytoplasmic fractions at any time point, indicating that once phosphorylated they translocated almost instantaneously to the nucleus (data not shown). A second, higher molecular weight band was detected in the phospho-SMAD3 blots (Fig. 4B, upper of the two bands). This likely corresponded to phospho-SMAD1, phospho-SMAD5 or both (Michael Reiss, Robert Wood Johnson Medical School, New Brunswick, NJ, personal communication). Although less intense than phospho-SMAD3, this band also increased in intensity after activin A treatment in both cell lines. TGFß1 and 3 have been shown to stimulate SMAD1 and SMAD5 phosphorylation (43, 44), so this noncanonical pattern of SMAD activation (i.e. phosphorylation of receptor-regulated SMADs in addition to SMADs 2 and 3) is not without precedent.
|
, A and B). The co-SMAD, SMAD4, also had no effect on FSHß mRNA levels when transfected alone. Immunoblots of protein lysates prepared from cells in adjacent wells transfected at the same time confirmed expression of all SMADs, with the notable exception of SMAD2 (see insets in Fig. 5).
|
, inset). SMAD2 and SMAD2/3-specific antibodies similarly failed to detect high levels of Flag-SMAD2 expression (data not shown). Two additional constructs, HA- human SMAD2 (in pcDNA3.0) and Flag-human SMAD2 (in pCMV5B), were similarly expressed at low levels relative to the other SMADs (data not shown). To determine whether this was a cell-specific phenomenon, exogenous SMAD2 expression was examined in transfected Chinese hamster ovary cells (CHO). Flag-SMAD2 and -SMAD3 were expressed at equivalent levels in this cell line; however, Flag-SMAD2 levels were again significantly lower than those of Flag-SMAD3 in LßT2 cells transfected at the same time (Fig. 6A).
|
and 4A, and data not shown. See also Ref 46). In addition, we observed that treatment of cells with the proteasome inhibitor MG132 (10 µM) had no effect on exogenous SMAD2 or SMAD3 protein levels (data not shown), indicating that the overexpressed SMAD2 was not degraded through the ubiquitin-proteasome pathway.
We next examined whether or not the mRNA for Flag-SMAD2 was detectably expressed in transfected cells. Whereas Northern blot analysis using an oligonucleotide probe directed against the N-terminal Flag tag readily detected mRNA for Flag-SMAD2 and Flag-SMAD3 in CHO cells, only full-length Flag-SMAD3 was expressed at significant levels in transfected LßT2 cells (Fig. 6B, arrow). For both SMAD2 and SMAD3 transfected LßT2 cells, we observed a high abundance, low molecular weight "smear," which was absent in empty vector transfected cells (Fig. 6B, arrowhead). These data indicated that the mRNAs derived from the expression constructs were somehow degraded. When the blot was stripped and reprobed with a cDNA corresponding to ribosomal protein L19 (RPL19), a single transcript of 0.7 kb was detected in all lanes, indicating equal loading of RNAs in all lanes and more importantly that degradation of the RNAs used in these analyses was not ubiquitous (Fig. 6B). Rather, the degradation appeared to be specific to the products of the expression constructs.
The Flag directed antisense oligonucleotide recognized sequence at the 5' end of the transcripts. We therefore asked whether the low molecular weight smear might correspond to RNAs of varying lengths that were terminated prematurely during transcription before completion of the full-length mRNA. Using an oligonucleotide probe directed against the 3' end of human SMAD2, we detected a pattern of results identical to that seen with the Flag probe (data not shown). In addition, a SMAD2 cDNA probe produced the same pattern of results (data not shown), indicating that the smearing was not an artifact of using an oligonucleotide probe. Collectively, these data suggest that full-length transcripts were generated in these cells but were then degraded. This degradation might have limited the abundance of full-length transcripts that could be translated into full-length protein products. This process appeared to be more robust in the case of SMAD2 than SMAD3 in LßT2 cells, and might have therefore contributed to the lower exogenous SMAD2 than SMAD3 protein levels in these transfected cells. The mechanisms mediating this degradation are not yet known.
Because transfected SMAD2 was expressed at significantly lower levels than was SMAD3 in our system, we could not conclude definitively that there was a qualitative difference between these two SMADs and their roles in FSHß trans-activation. That is, our results might simply reflect a quantitative difference in the overall levels of transfected SMAD2 and SMAD3. We have observed that increasing levels of SMAD3 dose-dependently stimulated FSHß mRNA levels (data not shown; see also Ref. 33). Therefore, it is possible that if SMAD2 were expressed at higher levels, then we would have observed stimulation of FSHß expression. Unfortunately, transfecting larger quantities of SMAD2 expression vector did not appreciably increase exogenous protein levels (data not shown).
A SMAD2 Splice Variant Is Expressed in Mouse Pituitary and Gonadotrope Cells and Can Stimulate FSHß Expression
SMAD2 and SMAD3 are highly conserved, sharing 91% amino acid sequence identity. These proteins appear to play redundant functions in a variety of systems. However, there are also several cases where their functions can be dissociated (47). A major structural difference between the two proteins exists within their amino terminal MAD homology 1 (MH1) domains, where SMAD2 has an additional 30 amino acids that are absent in SMAD3. These 30 amino acids are encoded entirely by a single 90-bp exon (exon 3 of 11 exon gene) and their inclusion disrupts the DNA binding capacity of SMAD2. That is, SMAD3 binds DNA through a ß-hairpin structure within its MH1 domain (48, 49). The 30-mino acid insertion in SMAD2 is just N-terminal to the ß-hairpin and appears to sterically hinder DNA binding. A naturally occurring splice variant of SMAD2 has been described in which exon 3 is skipped [SMAD2
exon3 (Ref. 50)]. This in-frame splice variant can bind DNA, like SMAD3 (50, 51). In fact, SMAD2exon3 appears to be functionally equivalent to SMAD3 in vitro. We questioned whether exon 3 contributed to the apparent functional difference between SMADs 2 and 3 described above and/or affected our ability to overexpress SMAD2.
RT-PCR analysis indicated that both SMAD2 and SMAD2exon3 were expressed in LßT2 and T3–1 cells, as well as in adult mouse pituitary (Fig. 7A). Full-length SMAD2 mRNA was estimated to be approximately 20-fold more abundant than SMAD2exon3 based on densitometry measures and cloning frequency. Thus far, we have been unable to confirm endogenous SMAD2exon3 protein expression in these cells by Western blot, although the antibody used (anti-SMAD2/3) could detect SMAD2exon3 produced recombinantly and in transfected cells (data not shown). Transient transfection of Flag-human SMAD2exon3 in LßT2 cells stimulated FSHß mRNA levels but was about half as potent as SMAD3 (Fig. 7B). Unlike the case for full-length SMAD2, transfected SMAD2exon3 was expressed at equivalent levels to SMAD3 (Fig. 7B, inset).
|
, SMAD3 potently stimulated FSHß expression, whereas SMAD + TID had no effect. Interestingly, like full-length SMAD2, SMAD3 + TID was expressed at very low levels relative to SMAD3 (Fig. 7, C and D, inset). These data were consistent with previous observations that the SMAD2 exon 3 sequence somehow hindered overexpression by transient transfection in LßT2 cells. Also like SMAD2, SMAD3 + TID could be expressed at high levels in CHO, but not in LßT2 cells (Fig. 7C), indicating that the expression construct was functional.
Suppression of SMAD2 or SMAD3 Protein Expression Attenuates FSHß Transcription in LßT2 Cells
The data presented thus far indicate that SMAD2 and SMAD3 are phosphorylated rapidly after activin stimulation and that SMAD3 overexpression stimulates FSHß expression in LßT2 cells in a ligand-independent fashion. However, these data alone do not indicate whether or not SMAD3 is necessary for activin-stimulated FSHß expression, nor do they rule out SMAD2’s involvement. Therefore, we first examined the effects of activin A on endogenous FSHß expression in the face of decreased SMAD3 protein levels in LßT2 cells. We tried several means to decrease endogenous SMAD3 protein levels, including SMAD3 antisense expression constructs, morpholino antisense oligonucleotides and a vector-based RNA interference (RNAi) method (52). Upon transfection, none of these reagents produced a detectable reduction in overall SMAD3 protein levels in LßT2 cells (data not shown). We believe this stemmed at least in part from the low transfection efficiency we have experienced with this cell line (
10%). That is, these reagents may have affected SMAD3 protein levels in transfected cells, but the persistent expression of SMAD3 in nontransfected cells might have obscured our ability to observe this effect. This contention was supported to some extent by the observed decline in endogenous SMAD3 protein levels in NIH3T3 mouse fibroblast cells transfected with the SMAD3 RNAi vector relative to cells transfected with the empty parent vector (pBS/U6) (Fig. 8A). Endogenous SMAD2 protein levels were unaffected by this vector in these cells, showing its specificity for SMAD3. NIH3T3 cells were transfected with much higher efficiency than were LßT2 cells and the decreased protein levels observed by Western blot in the former cell line, but not in the latter, likely reflected this difference. Importantly, the data show that the designed SMAD3 RNAi vector, which expresses a short hairpin RNA (shRNA) duplex under the control of the U6 promoter, could diminish endogenous SMAD3 levels in murine cells.
|
).
To show that the SMAD3 RNAi construct could also function in LßT2 cells, we transfected cells with a combination of a Flag-SMAD3 expression construct and the SMAD3 RNAi vector. As shown in Fig. 8B, the SMAD3 shRNA duplex dramatically suppressed, but did not completely block, Flag-SMAD3 production. Note, the overexpressed SMAD3 was N-terminally Flag-tagged and therefore migrated more slowly than endogenous SMAD3. In the Western blot using a SMAD3 antibody, we observed persistent endogenous SMAD3 expression in the face of diminished exogenous SMAD3 expression (lower band in Fig. 8B).
Based on these results and the ability of the SMAD3 shRNA duplex to suppress endogenous SMAD3 production in another mouse cell line (i.e. NIH3T3), we believed that this RNAi construct could be used to diminish endogenous SMAD3 in LßT2 cells. Cells were transfected with -1990/+1mFSHß-luc alone and in combination with the SMAD3 RNAi vector. We reproducibly observed a 50–60% decline in activin A-stimulated reporter gene activity in the presence of the SMAD3 RNAi construct (Fig. 8D). Under control conditions (in the absence of exogenous activin A), we observed a 25–40% reduction in basal FSHß expression by the SMAD3 shRNA duplex; however, the diminished activin response was not solely attributable to this decreased basal transcription. Activin A-stimulated FSHß transcription decreased from 4.6-fold in the absence of the SMAD3 RNAi vector to 3.2-fold in its presence. In addition, the effects of the RNAi vector were specific to the FSHß promoter, as transcription from the empty pGL3-Basic vector was unaffected by SMAD3 RNAi vector in the absence or presence of activin A (data not shown). These data indicated that SMAD3 is necessary for a maximal transcriptional response of the FSHß promoter to activin A.
A similar approach was used to assess a role for SMAD2 in activin-stimulated FSHß transcription. This approach was particularly important in light of the failure to overexpress SMAD2 in LßT2 cells (Figs. 5 and 6). An RNAi vector was designed to target a sequence within exon 3, and therefore was predicted to affect SMAD2, but not SMAD3 or SMAD2exon3, protein levels. Because of the difficulties in overexpressing SMAD2 in LßT2 cells, we determined the efficacy and efficiency of the construct in CHO cells. The SMAD2 shRNA duplex significantly diminished Flag-SMAD2 protein levels in transfected cells (Fig. 8C). When used in the promoter-reporter assay, the SMAD2 RNAi vector significantly attenuated FSHß transcription in the presence or absence of activin A (Fig. 8D).
Unlike the case with the SMAD3 RNAi experiments, the fold-induction by activin A was not greatly different in the presence or absence of the SMAD2 RNAi vector (4.0 vs. 4.6, respectively). These data suggested that SMAD2 might play a role in basal, but not necessarily, stimulated FSHß promoter activity. To examine this possibility in greater detail, cells were transfected with the -1990/+1mFSHß-luc reporter and ALK4TD plus or minus the SMAD2 or SMAD3 RNAi plasmids. Here, both RNAi constructs produced a nearly 30% decline in basal promoter activity and a 50–70% attenuation of ALK4TD-stimulated expression (relative to pBS/U6 transfected cells) (Fig. 9). The fold induction by ALK4TD was attenuated 58 and 34% by the SMAD2 and SMAD3 RNAi constructs, respectively. Why the SMAD2 RNAi construct was more inhibitory than SMAD3 RNAi construct in the case of ALK4TD-stimulated relative to activin A-stimulated reporter activity is not yet clear. Nonetheless, these data suggest that SMAD2, like SMAD3, is involved in both basal and stimulated FSHß transcription.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
As in other cell systems, activins rapidly (within 5–10 min) stimulate SMAD2 and SMAD3 phosphorylation in both T3–1 and LßT2 cells. Moreover, once phosphorylated, both SMADs translocate immediately to the nucleus. To assess their relative roles in FSHß transcription, SMAD2 and SMAD3 were transiently transfected into LßT2 cells. SMAD3, but not SMAD2, stimulated endogenous FSHß mRNA levels in a ligand-independent manner. These data indicate that SMAD3 is sufficient for stimulation of FSHß expression and are consistent with a previous report showing that SMAD3 overexpression stimulates rat FSHß promoter-reporter activity in LßT2 cells (33). Here, we further show that SMAD3 is necessary for activin A’s stimulation of mouse FSHß transcription. Using a vector-based RNAi approach (52), we knocked down endogenous SMAD3 protein expression in LßT2 cells. Cells cotransfected with a murine FSHß promoter fragment (-1990 to +1) upstream of a luciferase reporter show a significant reduction in activin A-stimulated reporter activity relative to cells transfected with the empty RNAi plasmid. In addition, under control conditions (i.e. in the absence of activin A), the SMAD3 RNAi vector decreases reporter activity, suggesting that basal FSHß transcription may also be dependent upon SMAD3. In the latter case, it is possible that endogenous activin B produced by LßT2 cells stimulates FSHß expression and does so, at least in part, via SMAD3. Consistent with this hypothesis, follistatin treatment completely blocks basal endogenous FSHß mRNA expression in these cells (32).
Although activin A-stimulated gene transcription is attenuated in cells cotransfected with the SMAD3 RNAi vector, it is not blocked completely. This may reflect the fact that the vector we used significantly diminishes SMAD3 protein expression but does not completely block it. In addition, it may indicate that signaling proteins in addition to SMAD3 are part of the signal transduction mechanism through which activins stimulate FSHß transcription. SMAD2 is an obvious candidate in this regard; however, our failure to overexpress this protein in transient transfection assays (see Discussion below) led us initially to examine alternative proteins. We observed that a splice variant of SMAD2, which lacks exon 3 of the 11 exon gene (50), stimulates endogenous FSHß expression when transfected into LßT2 cells. Importantly, SMAD2exon3 mRNA is expressed in LßT2 cells and in the adult pituitary gland, albeit at approximately 20-fold lower concentrations than full-length SMAD2. Collectively, these data suggest that this SMAD2 splice variant may be a component of endogenous activin-stimulated transcriptional complexes regulating FSHß transcription. However, given the relatively low abundance of SMAD2exon3 mRNA and our inability to detect the protein in LßT2 cells, it is unclear to what extent this variant is involved in endogenous FSHß regulation.
Because SMAD3 and SMAD2exon3 can bind DNA directly, whereas SMAD2 cannot, we originally hypothesized that direct binding of SMADs to DNA may be critical for trans-activation of the FSHß gene. However, given our inability to overexpress full-length SMAD2 or SMAD3 + TID in LßT2 cells, no definitive conclusions about the necessity for direct DNA binding can be drawn from these data. Nonetheless, other data do indicate a role for SMAD2 in FSHß regulation. First, transfection of a dominant-negative form of SMAD2 into LßT2 cells was shown to suppress activin A-stimulated increases in endogenous FSHß mRNA levels and to block completely activin-induced increases in FSH secretion (27). Although these data are suggestive, they do not conclusively indicate a role for SMAD2. That is, the SMAD2 dominant-negative construct may have nonspecifically perturbed SMAD3 signaling as well (12, 53), and this could account for the observed abrogation of activin A signaling. Second, the SMAD2 RNAi experiments presented here show decreases in basal and stimulated FSHß transcription in the face of diminished endogenous SMAD2 levels in LßT2 cells. Because the SMAD2 RNAi construct does not significantly affect SMAD3 expression (data not shown), it is clear that its effects are not the result of nonspecific (unintended) abrogation of SMAD3 signaling.
A previous report indicated that transfection of SMAD2 alone or in combination with SMAD4 failed to stimulate rat FSHß promoter activity in LßT2 cells but did not indicate whether or not exogenous SMAD2 was actually produced in transfected cells (33). The only control for expression of the SMAD2 construct reported in that study involved transfection into TSA cells and stimulation of the SMAD-responsive promoter construct, p3TP-luc. As shown here, exogenous SMAD2 expression can vary dramatically between different cell lines, making this a less than optimal control experiment. In addition, there was no assessment by immunoblot (or other means) of SMAD2 protein expression relative to SMAD3. We believe that, as observed here, SMAD2 may not have been overexpressed in LßT2 cells in the previous study.
Our results indicate that the mRNAs produced by the SMAD expression vectors are degraded rapidly in LßT2 cells, and this appears to be more extreme for SMAD2 than for SMAD3. The mechanism mediating this degradation process is currently unknown, but the degradation itself appears to contribute to abrogated SMAD2 overexpression. Even if other mechanisms are also involved, it seems clear that the SMAD2 exon 3 sequence plays some role in low level SMAD2 overexpression. When exon 3 is removed from SMAD2, the protein can be expressed. When it is introduced into SMAD3, protein production is dramatically reduced. Paradoxically, we readily detect endogenous SMAD2 mRNA and protein (total and phosphorylated) in these cells, so the inhibitory mechanism appears to be specific to the exogenous full-length SMAD2. How the cells discriminate between the endogenous and exogenous mRNAs is unknown. Nonetheless, whereas the transient transfection paradigm does not permit an assessment of SMAD2’s involvement in FSHß transcription, the RNAi approach does indicate a role for the protein in this process, and these latter data are consistent with those of a previous report (27).
Finally, LßT2 and T3–1 cell lines were both generated by targeted oncogenesis in transgenic mice (16, 17). The two cell lines are similar in expressing markers of differentiated gonadotropes, including the - gonadotropin subunit, GnRH receptor and SF-1; however, LßT2 cells also express both the LHß and FSHß subunits, but T3–1 cells do not. As such, LßT2 cells appear to be more highly differentiated and hence more gonadotrope-like in phenotype. Originally, LßT2 cells were not thought to produce FSHß (17, 54); however, activin A treatment was shown to stimulate FSHß transcription and FSH secretion (29). T3–1 cells treated with activin A do not produce endogenous FSHß nor do FSHß promoter-reporter constructs respond to activin A in this cell line (Fig. 3C) (32).
Here, activin A stimulated SMAD2 and SMAD3 phosphorylation in both cell lines. Therefore, the failure of T3–1 cells to produce FSHß in response to activins does not appear to reflect a deficit in SMAD proteins, their phosphorylation by activin receptors or their nuclear translocation. SMADs form higher order complexes with cofactor proteins in the nucleus. Some, like forkhead activin signal transducer 1 (forkhead/winged-helix transcription factor H1), are DNA binding proteins, which anchor SMAD complexes to specific DNA binding elements within target genes (14). We (and others) hypothesize that LßT2 and T3–1 cells express a different complement of nuclear cofactors and these different proteins contribute to the selective ability of LßT2 cells (and by extension gonadotropes) to produce FSHß basally and in response to activins. A direct comparison of differential gene expression in these two cell lines may help uncover cofactors proteins essential for FSHß transcription (55).
In summary, activin A stimulates phosphorylation and nuclear translocation of SMAD2 and SMAD3 in LßT2 cells within 10 min. After 30 min, activin A stimulates an increase in FSHß primary transcript levels followed by an increase in FSHß mRNA levels after 60 min. Transient transfection of SMAD3, but not SMAD2, mimics the effects of exogenous activin treatment. SMAD2exon3, which is expressed in both LßT2 cells and adult mouse pituitary, lacks a 30-amino acid domain that interferes with DNA binding. Overexpression of SMAD2exon3 stimulates FSHß expression in a manner similar to SMAD3. Reduction of endogenous SMAD2 and SMAD3 protein levels by RNAi attenuates activin A- and ALK4TD-stimulated increases in FSHß transcription. Collectively, these data suggest that activins directly stimulate FSHß transcription via both SMAD2 and SMAD3-dependent mechanisms in mouse gonadotrope cells. The data from T3–1 cells, however, demonstrate that SMAD2 and SMAD3 activation alone cannot account for gonadotrope-specific and activin-stimulated expression of the FSHß subunit. The challenge for the future will be to identify gonadotrope-restricted factors necessary for these responses and how they interface with activated SMADs at specific cis-acting elements in the FSHß promoter to regulate transcription.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Constructs
Flag-human SMAD1, Flag-human SMAD2, HA-human SMAD2, Flag-mouse SMAD4, and rat ALK4-HA (all in pcDNA3.0) were provided by Dr. Teresa Woodruff. Flag-human SMAD2 (in pCMV5B) was a gift from Dr. Liliana Attisano (University of Toronto, Toronto, Ontario, Canada). Flag-human SMAD3 (in pCMV5) and myc-rat SMAD8 (in pcDNA3.0) were obtained from Dr. Yan Chen (Indiana University, Indianapolis, IN). Flag-human SMAD2exon3 and Flag-human SMAD5 (both in pcdef3) were gifts from Drs. M. Kato (The Cancer Institute of the Japanese Foundation for Cancer Research, Tokyo, Japan) and T. Watabe (University of Tokyo, Tokyo, Japan). Untagged human SMAD3 and SMAD3 + TID (in pcDNA3.1) were gifts from Dr. Stephane Huet (GlaxoSmithKline, Les Ulis, France). Flag-mouse SMAD7 (in pcDNA3.0) was provided by Dr. Carl-Henrik Heldin (Ludwig Institute for Cancer Research, Uppsala, Sweden). The untagged mouse SMAD2 expression construct was generated by RT-PCR amplifying the coding sequence and ligating it into pcDNA3.0.
Rat ALK4-T206D-HA was produced by PCR-based site-directed mutagenesis of the parent rat ALK4-HA construct. The mouse FSHß promoter-reporter construct (-1990FSHß-luc) was made by PCR amplifying -1990 to +1 (+1 is the start of transcription as determined by RNA ligase mediated RACE, Ambion, Austin, TX) of the FSHß 5' flanking region from T3–1 cell genomic DNA using the following primer set (IDT, Coralville, IA) and Platinum Taq: forward 5'-CCTGTTCATTAACCACTGAGCT and reverse 5'-CACTGAGTCAAGTTACACCTCA. The PCR product was gel purified, kinased and then ligated into the dephosphorylated SmaI site of pGL3-Basic (Promega) using standard techniques. Correct 5'-3' orientation of the insert was determined by restriction digest and DNA sequencing.
The SMAD2 and SMAD3 RNAi vectors were created after guidelines established by Shi and colleagues (52). For the SMAD3 construct, we aligned the mouse SMAD2 and SMAD3 open reading frames to identify stretches of low sequence conservation. Within the SMAD3 MH1 domain, we identified a 22-bp region (bp 89–110 of the SMAD3 open reading frame) in which 8-bp mismatch with the corresponding SMAD2 sequence and which fulfills the other criteria for use in the RNAi expression vector. A total of four oligos were synthesized commercially. Bold sequences represent adapter sequences. mSMAD3–1a (5'-GGTGCGAGAAGGCGGTCAAGAGA-3') was annealed to mSMAD3–1b (5'-AGCTTCTCTTGACCGCCTTCTCGCACC-3') and ligated into pBS/U6 digested with ApaI and blunted with Klenow followed by digestion with HindIII. mSMAD3–2a (5'-AGCTTCTCTTGACCGCCTTCTCGCACCCTTTTTG-3') was annealed to mSMAD3–2b (5'-AATTCAAAAAGGGTGCGAGAAGGCGGTCAAGAGA-3') and ligated into the EcoRI and HindIII sites of pBS/U6 from the first ligation. The plasmid was sequenced to confirm its proper construction. The SMAD2 RNAi construct was directed against the following sequence from mouse SMAD2 exon 3: 5'-GGACTGAGTACAGCAAATACGG-3'. Primers were designed and ligated into pBS/U6 as described for the SMAD3 construct.
Cell Culture, Transfections, and Reporter Assays
LßT2 and T3–1 cells were gifts from Dr. Pamela Mellon (University of California, San Diego, San Diego, CA). CHO and NIH3T3 mouse fibroblast cells were obtained from the Cell and Tissue Culture Core Facility at the Population Council. LßT2, T3–1, and NIH3T3 cells were cultured in DMEM supplemented with 10% fetal bovine serum and 4 µg/ml gentamycin at 37 C/5% CO2. CHO cells were cultured in DMEM:F12 supplemented with 5% horse serum, 2.5% fetal calf serum, and 4 µg/ml gentamycin at 37 C/5% CO2. For transfections, cells were washed with Optimem I or serum-free media and transfected with the indicated plasmids using Lipofectamine/Plus following the manufacturer’s instructions. Transfections were balanced so that in each experiment all wells were treated with equivalent amounts of DNA. After 6 h, transfection solution was replaced with growth media and cells were allowed to recover for 18–24 h. Cells were then washed and incubated in serum-free DMEM (or DMEM:F12) for 18–24 h before collection of protein or RNA for subsequent analyses. In activin-stimulated cells, growth media was replaced with serum-free DMEM for 12–16 h. Ligand was then added to the wells at the indicated concentrations for the indicated times. For SMAD phosphorylation and nuclear translocation experiments, cells were cultured in 10-cm plates and treated once the cells were 80–90% confluent.
For luciferase reporter assays, cells were seeded at 0.5–1 x 106 cells per well in six-well dishes. After approximately 36 h, cells (LßT2 or T3–1) were transfected with 1 µg empty pGL3-Basic or -1990/+1mFSHß-luc and 100 ng of pRL-CMV (Promega) using Lipofectamine/Plus. After 6 h, transfection media was replaced with growth media or growth media plus 30 ng/ml activin A. Approximately 24 h later, cells were washed with PBS and protein extracted using 500 µl/well of 1x Passive Lysis Buffer (PLB, Promega). Twenty microliters of each sample were assayed using the Dual Luciferase Reporter System on a Luminoskan Ascent luminometer (Thermo Labsystems, Franklin, MA). For RNAi experiments, cells were treated as described except 2 µg of empty pBS/U6, SMAD3 RNAi, or SMAD2 RNAi vectors were included in the transfections and cells were allowed to recover for an additional 24 h before hormone treatment. All treatment conditions were run in triplicate and each experiment performed a minimum of two times.
Protein and RNA Extractions
After washing the cells with ice-cold PBS (pH. 7.4), protein lysates were prepared using RIPA buffer including protease inhibitors. In other experiments, cytoplasmic and nuclear proteins were purified using NE-PER reagents (with protease inhibitors) following the manufacturer’s instructions. Protein concentrations were determined by bicinchoninic acid assay.
Total RNA was extracted after the PBS wash using TRI-reagent or Trizol following the manufacturer’s instructions. For analysis of primary transcript levels, RNA samples were DNased using RQ1 DNase. RNA concentration was determined by spectrophotometry. For real-time RT-PCR analyses (see Real-Time RT-PCR), samples were DNased after determination of RNA concentration.
Western Blot Analyses
Protein lysates were diluted with 2x sodium dodecyl sulfate (SDS) gel loading buffer containing 200 mM dithiothreitol or ß-mercaptoethanol, denatured at 95 C for 5 min, and run on 8% Tris-glycine polyacrylamide gels. Proteins were transferred to Protran nitrocellulose filters (Schleicher & Schuell, Keene, NH) using 0.5x Towbin buffer (12.5 mM Tris, 96 mM glycine, 20% methanol). Blots were blocked in 5% nonfat milk in Tris-buffered saline [10 mM Tris (pH 8.0), 150 mM NaCl] containing 0.05% Tween 20 (TBST) for 1 h at room temperature. Blots were then incubated overnight at 4 C in the indicated primary antibody diluted in 2.5% nonfat milk in TBST. After extensive washes in TBST, blots were incubated for 1 h at room temperature in goat antirabbit IgG horseradish peroxidase or goat antimouse IgG horseradish peroxidase (Bio-Rad, Hercules, CA) diluted 1:3000 in nonfat milk in TBST. Blots were washed in TBST, incubated with the ECL reagent (Amersham Pharmacia, Piscataway, NJ) and exposed to x-ray film (Kodak, Rochester, NY, or Denville, South Plainfield, NJ).
RT-PCR
For some analyses, FSHß mRNA and primary transcript levels were measured by semiquantitative RT-PCR. Five micrograms of total RNA were reverse transcribed into cDNA using 100 ng random hexamer primers and 100 U Moloney murine leukemia virus reverse transcriptase. One tenth of each reverse transcriptase reaction was used as template in PCR for FSHß mRNA, FSHß PT, and RPL19 (see Table 1). PCR was run using the following conditions for 27 cycles (RPL19) or 35–38 cycles (FSHß mRNA and PT): 94 C for 30 sec, 55 C for 30 sec, and 72 C for 30 sec. Reactions contained 0.4 pmol of each primer, 200 µM deoxynucleotide triphosphates, 1.5 mM MgCl2, 1x PCR buffer, and 2.5 U Taq polymerase. Following a final 7-min extension step at 72 C, one fifth of each reaction was resolved on a 1.2% agarose gel containing ethidium bromide. Gels were photodocumented and analyzed using a digital camera interfaced with an IBM ThinkPad computer running the Kodak Digital Science 1D software (version 2.0.2) software. Each treatment was performed in duplicate and all experiments were repeated up to three times. Preliminary analyses established that all data were collected during the exponential phase of amplification.
|
T3–1 cells and female wild-type 129/SvEv mouse pituitary. The PCR primers were directed against sequences in exons 2 and 4 of mouse SMAD2 (Table 1). Amplified products of 542 and 452 bp were expected from transcripts containing (SMAD2) and skipping exon 3 (SMAD2exon3), respectively. Amplified products were gel purified and sequenced by DyeTerminator cycle sequencing (Applied Biosystems, Foster City, CA) to confirm their identities. In addition, the full-length SMAD2 and SMAD2exon3 were amplified using the following primer set: forward 5'-CGGAATTCCGATGTCGTCCATCTTGCCATTCACT and reverse 5'-GCTCTAGAGCTTACGACATGCTTGAGCATCG. The cloned full-length mouse SMAD2exon3 sequence was submitted to GenBank (accession no. AY334552). For all RT-PCR analyses, reactions with no template (H2O only) were used to confirm the absence of contaminating DNA in the reagents. In addition, the majority of the PCR primers were designed to span intronic sequences so that amplified mRNA (cDNA) could be distinguished from amplified contaminating genomic DNA in the RNA samples. No contamination was detected in any of the analyses.
Real-Time RT-PCR
Endogenous FSHß mRNA levels were measured by real-time RT-PCR using the relative standard curve method in separate tubes (56). One microgram of each RNA sample was incubated with DNase I for 15 min at room temperature, followed by incubation at 65 C for 10 min in the presence of 2.5 mM EDTA. Five hundred nanograms of each sample were reverse transcribed into cDNA using SuperScript First-Strand Synthesis System and following the manufacturer’s instructions. Next, 50 ng of each sample were subjected to PCR amplification in duplicate in 50-µl reactions containing 1x uracil DNA glycosylase mix, 1x 6-carboxy-X-rhodamine (ROX) reference dye, and 0.2 nM each of 6-carboxyfluorescein (FAM)-labeled FSHß forward LUX primer (Invitrogen) and unlabeled reverse primer (see Table 1). Samples were run in MicroAmp Optical 96-well reaction plates covered with optical caps (Applied Biosystems) in an ABI 7700 Sequence Detection System at 50 C for 2 min, 95 min for 2 min, followed by 40 cycles at 95 C for 15 sec, 55 C for 30 sec, and 72 C for 30 sec. After the last extension step, a melting curve analysis was run to confirm that single amplicons were generated in each reaction. A second set of PCRs was run as described, except that primers for mouse TATA box binding protein (TBP) were used in place of the FSHß primers. The sequences for the 6-carboxy-4',5'-dichloro-2',7'-dimethoxyfluorescein (JOE)-labeled TBP forward LUX primer and unlabeled reverse primer are indicated in Table 1.
In each experiment, a relative standard curve was generated by including dilutions (0.39, 1.56, 6.25, 25, and 100 ng) of cDNA from LßT2 cells treated with 100 ng/ml activin A for 3 h. The threshold cycle value obtained for each sample was compared with the standard curve generated in the same run and an input amount (in nanograms) was derived. Replicates were averaged. The mean value for FSHß was then divided by the mean value of TBP for the same sample. The control condition in each experiment (e.g. untreated or empty vector transfected) was set to 1 and all treatments expressed as a fold-change relative to the control.
None of the treatments were observed to affect TBP expression. In addition, melting curve and gel electrophoresis confirmed that both primer sets generated a single amplicon in each reaction. No template controls (H2O only) confirmed the absence of contaminants in the reagents.
Northern Blot Analysis
Equal amounts of total RNA derived from LßT2 and CHO cells transfected with empty vector, Flag-hSMAD2, and Flag-hSMAD3 were run on a 3[N-morpholino]propanesulfonic acid/formaldehyde gel and transferred to a positively charged nylon membrane (Nytran, Schleicher & Schuell) by capillary action using 20x saline sodium citrate (SSC). The membrane was then hybridized with an oligonucleotide probe corresponding to the Flag tag (5'-CTTGTCATCGTCGTCCTTGTAGTCCAT). The probe was labeled with [
-32P]ATP to a specific activity of 2.5 x 106 cpm/pmol and was incubated with the blot overnight in ULTRAhyb-Oligo (Ambion, Austin, TX) at 42 C using a modified sandwich method (57). After two 30 min washes at 42 C in 20x saline sodium phosphate EDTA buffer, 0.5% SDS, the blot was exposed to x-ray film overnight. The membrane was then stripped in 0.5% boiling SDS before reprobing with a [-32P] deoxy-CTP labeled cDNA probe corresponding to bp 401–595 of the rat ribosomal protein L19 (GenBank accession no. NM_031103) in ULTRAhyb overnight at 42 C. The membrane was then washed two times each in 2x SSC, 0.1% SDS, and 0.1x SSC, 0.1% SDS at 42 C before being exposed to x-ray film.
Statistical Analyses
Differences in means between treatments were assessed by analysis of variance, followed by post hoc Fisher’s least significant difference analyses. Statistical significance was set at P < 0.05.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication July 7, 2003. Accepted for publication December 19, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Wang, P. K Nicholls, P. G Stanton, C. A Harrison, M. Sarraj, R. B Gilchrist, J. K Findlay, and P. G Farnworth Extra-ovarian expression and activity of growth differentiation factor 9 J. Endocrinol., September 1, 2009; 202(3): 419 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Xia and A. L Schneyer The biology of activin: recent advances in structure, regulation and function J. Endocrinol., July 1, 2009; 202(1): 1 - 12. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Lamba, J. Fortin, S. Tran, Y. Wang, and D. J. Bernard A Novel Role for the Forkhead Transcription Factor FOXL2 in Activin A-Regulated Follicle-Stimulating Hormone {beta} Subunit Transcription Mol. Endocrinol., July 1, 2009; 23(7): 1001 - 1013. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Ho and D. J. Bernard Bone Morphogenetic Protein 2 Signals via BMPR1A to Regulate Murine Follicle-Stimulating Hormone Beta Subunit Transcription Biol Reprod, July 1, 2009; 81(1): 133 - 141. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Bak, L. Carpio, J. L Kipp, P. Lamba, Y. Wang, R.-S. Ge, M. P Hardy, K. E Mayo, and D. J Bernard Activins regulate 17{beta}-hydroxysteroid dehydrogenase type I transcription in murine gonadotrope cells J. Endocrinol., April 1, 2009; 201(1): 89 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fortin, P. Lamba, Y. Wang, and D. J. Bernard Conservation of mechanisms mediating gonadotrophin-releasing hormone 1 stimulation of human luteinizing hormone {beta} subunit transcription Mol. Hum. Reprod., February 1, 2009; 15(2): 77 - 87. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, J. Fortin, P. Lamba, M. Bonomi, L. Persani, M. S. Roberson, and D. J. Bernard Activator Protein-1 and Smad Proteins Synergistically Regulate Human Follicle-Stimulating Hormone {beta}-Promoter Activity Endocrinology, November 1, 2008; 149(11): 5577 - 5591. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Feng, M. A. Lawson, and P. Melamed A Proteomic Comparison of Immature and Mature Mouse Gonadotrophs Reveals Novel Differentially Expressed Nuclear Proteins that Regulate Gonadotropin Gene Transcription and RNA Splicing Biol Reprod, September 1, 2008; 79(3): 546 - 561. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Makanji, K. L. Walton, M. C. Wilce, K. L. Chan, D. M. Robertson, and C. A. Harrison Suppression of Inhibin A Biological Activity by Alterations in the Binding Site for Betaglycan J. Biol. Chem., June 13, 2008; 283(24): 16743 - 16751. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lamba, V. Khivansara, A. C. D'Alessio, M. M. Santos, and D. J. Bernard Paired-Like Homeodomain Transcription Factors 1 and 2 Regulate Follicle-Stimulating Hormone {beta}-Subunit Transcription through a Conserved cis-Element Endocrinology, June 1, 2008; 149(6): 3095 - 3108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Blount, J. M. Vaughan, W. W. Vale, and L. M. Bilezikjian A Smad-binding Element in Intron 1 Participates in Activin-dependent Regulation of the Follistatin Gene J. Biol. Chem., March 14, 2008; 283(11): 7016 - 7026. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L Nicol, M-O Faure, J R McNeilly, J Fontaine, C Taragnat, and A S McNeilly Bone morphogenetic protein-4 interacts with activin and GnRH to modulate gonadotrophin secretion in L{beta}T2 gonadotrophs J. Endocrinol., March 1, 2008; 196(3): 497 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. G. Thackray and P. L. Mellon Synergistic Induction of Follicle-Stimulating Hormone {beta}-Subunit Gene Expression by Gonadal Steroid Hormone Receptors and Smad Proteins Endocrinology, March 1, 2008; 149(3): 1091 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Stegmuller, M. A. Huynh, Z. Yuan, Y. Konishi, and A. Bonni TGF{beta}-Smad2 Signaling Regulates the Cdh1-APC/SnoN Pathway of Axonal Morphogenesis J. Neurosci., February 20, 2008; 28(8): 1961 - 1969. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Coss, C. M. Hand, K. K. J. Yaphockun, H. A. Ely, and P. L. Mellon p38 Mitogen-Activated Protein Kinase Is Critical for Synergistic Induction of the FSH{beta} Gene by Gonadotropin-Releasing Hormone and Activin through Augmentation of c-Fos Induction and Smad Phosphorylation Mol. Endocrinol., December 1, 2007; 21(12): 3071 - 3086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Li, J. M. Graff, A. E. O'Connor, K. L. Loveland, and M. M. Matzuk SMAD3 Regulates Gonadal Tumorigenesis Mol. Endocrinol., October 1, 2007; 21(10): 2472 - 2486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. D. Looyenga and G. D. Hammer Genetic Removal of Smad3 from Inhibin-Null Mice Attenuates Tumor Progression by Uncoupling Extracellular Mitogenic Signals from the Cell Cycle Machinery Mol. Endocrinol., October 1, 2007; 21(10): 2440 - 2457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zhu, A. S. Gleiberman, and M. G. Rosenfeld Molecular Physiology of Pituitary Development: Signaling and Transcriptional Networks Physiol Rev, July 1, 2007; 87(3): 933 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. L. Burger, D. J. Haisenleder, G. M. Wotton, K. W. Aylor, A. C. Dalkin, and J. C. Marshall The regulation of FSHbeta transcription by gonadal steroids: testosterone and estradiol modulation of the activin intracellular signaling pathway Am J Physiol Endocrinol Metab, July 1, 2007; 293(1): E277 - E285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Lawson, R. Tsutsumi, H. Zhang, I. Talukdar, B. K. Butler, S. J. Santos, P. L. Mellon, and N. J. G. Webster Pulse Sensitivity of the Luteinizing Hormone {beta} Promoter Is Determined by a Negative Feedback Loop Involving Early Growth Response-1 and Ngfi-A Binding Protein 1 and 2 Mol. Endocrinol., May 1, 2007; 21(5): 1175 - 1191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. B Lee, V. Khivansara, M. M Santos, P. Lamba, T. Yuen, S. C Sealfon, and D. J Bernard Bone morphogenetic protein 2 and activin A synergistically stimulate follicle-stimulating hormone {beta} subunit transcription J. Mol. Endocrinol., February 1, 2007; 38(2): 315 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. McGillivray, V. G. Thackray, D. Coss, and P. L. Mellon Activin and Glucocorticoids Synergistically Activate Follicle-Stimulating Hormone {beta}-Subunit Gene Expression in the Immortalized L{beta}T2 Gonadotrope Cell Line Endocrinology, February 1, 2007; 148(2): 762 - 773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhang, J. S. Bailey, D. Coss, B. Lin, R. Tsutsumi, M. A. Lawson, P. L. Mellon, and N. J. G. Webster Activin Modulates the Transcriptional Response of LssT2 Cells to Gonadotropin-Releasing Hormone and Alters Cellular Proliferation Mol. Endocrinol., November 1, 2006; 20(11): 2909 - 2930. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ao, K. Williams, N. A. Bhowmick, and S. W. Hayward Transforming Growth Factor-{beta} Promotes Invasion in Tumorigenic but not in Nontumorigenic Human Prostatic Epithelial Cells Cancer Res., August 15, 2006; 66(16): 8007 - 8016. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M Bilezikjian, A. L Blount, C. J Donaldson, and W. W Vale Pituitary actions of ligands of the TGF-{beta} family: activins and inhibins. Reproduction, August 1, 2006; 132(2): 207 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Lamba, M. M Santos, D. P Philips, and D. J Bernard Acute regulation of murine follicle-stimulating hormone {beta} subunit transcription by activin A J. Mol. Endocrinol., February 1, 2006; 36(1): 201 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. I. Suszko, D. M. Balkin, Y. Chen, and T. K. Woodruff Smad3 Mediates Activin-Induced Transcription of Follicle-Stimulating Hormone {beta}-Subunit Gene Mol. Endocrinol., July 1, 2005; 19(7): 1849 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Gregory, C. T. Lacza, A. A. Detz, S. Xu, L. A. Petrillo, and U. B. Kaiser Synergy between Activin A and Gonadotropin-Releasing Hormone in Transcriptional Activation of the Rat Follicle-Stimulating Hormone-{beta} Gene Mol. Endocrinol., January 1, 2005; 19(1): 237 - 254. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L L Burger, D J Haisenleder, A C Dalkin, and J C Marshall Regulation of gonadotropin subunit gene transcription J. Mol. Endocrinol., December 1, 2004; 33(3): 559 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tomic, K. P. Miller, H. A. Kenny, T. K. Woodruff, P. Hoyer, and J. A. Flaws Ovarian Follicle Development Requires Smad3 Mol. Endocrinol., September 1, 2004; 18(9): 2224 - 2240. [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 |
