| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Physiology, University of Kentucky College of Medicine, 800 Rose Street, Lexington, KY 40536, USA
1 Department of Biology, Western Kentucky University, 1 Big Red Way, Bowling Green, KY 4210, USA
(Requests for offprints should be addressed to S E Diamond; Email: sediam0{at}uky.edu)
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
), which binds DNA with high affinity and specificity (Ingraham et al. 1990) and interacts with transcription cofactors (Xu et al. 1998).
|
The ß-isoform of Pit-1, Pit-1ß, differs only by the splice-insertion of the 26 amino acid ß-domain at the beginning of exon 2 in the TAD (Fig. 1
), yet it has diametrically opposite effects with regard to hormone gene expression in anterior pituitary cells. While Pit-1 enhances PRL and GH gene expression, Pit-1ß blocks their gene expression and decreases the histone acetylation of their promoters in a manner dependent upon five hydrophobic ß-domain residues (Fig. 1) (Diamond & Gutierrez-Hartmann 1996, 2000, Ferry et al. 2005). Yet, in HeLa non-pituitary cells, Pit-1ß activates hormone gene transcription. The reason for this cell-type specificity has remained unknown to date, but it seems that a difference between the pituitary and non-pituitary cell environments modulates Pit-1ß transcriptional activity.
An important feature of POU factors is their ability to form cooperative homodimers and even heterodimers with other POU factors, both in the absence of DNA and even more so when bound to an appropriate DNA site. Consequently, diverse configurations of homo-dimeric and heterodimeric interactions emerge when different members of the POU domain gene family are coexpressed in the same cells or tissues. On some sites, Pit-1 can bind as a heterodimer with the POU domain factor Oct-1, which results in synergistic transactivation of the PRL promoter. However, on most sites, Pit-1 appears to bind as a homodimer (Elsholtz et al. 1990, Ingraham et al. 1990, Voss et al. 1991, 1993). The specific domain of the POU domain (POU-S) is essential for Pit-1–Pit-1 interactions and the result is efficient transactivation. The importance of this homodimerization is underscored by the fact that dominant-negative Pit-1 mutants, such as those that occur naturally in combined pituitary hormone deficiency (Arg271 Trp, Pro14 Leu and Pro24 Leu), impede the function of the wild-type Pit-1 transcribed from the other, wild type, allele. This phenomenon occurs despite their ability to bind DNA (Cohen et al. 1999, Parks et al. 1999, Pfaffle et al. 1999).
Similarly, Pit-1ß blocks hormone gene expression even though it effectively binds DNA (Konzak & Moore 1992, Morris et al. 1992, Theill et al. 1992). Some studies showed that the Pit-1ß might induce repression in pituitary cells by altering the balance between competing coactivator and corepressor interaction with Pit-1 (Xu et al. 1998). We recently showed that Pit-1ß reduces CBP recruitment to the PRL and GH promoters (Ferry et al. 2005). However, it is not clear whether the Pit-1ß directly inhibits coactivator recruitment, or whether the repression is simply an effect of decreased Pit-1 or Pit-1ß availability due to heterodimerization. In order to distinguish between these possibilities, we tested the constraints for: (1) reconstitution of repression in non-pituitary cells; (2) the further constraints on repression added by CBP coactivator and PKA; (3) the effect of the ß-domain on physical interaction with CBP; and (4) ß-domain sidechain chemistry. We demonstrate that the presence of Pit-1, CBP and PKA allow reconstitution of repression by Pit-1ß in non-pituitary cells, and that the ß-domain blocks physical interaction between Pit-1 and a Pit-1-binding domain of CBP. In addition, we showed that specific chemical properties beyond hydrophobicity are dispensable. We propose a model in which Pit-1ß heterodimerizes with Pit-1 and thus decreasing their efficiency as transcriptional activators. Furthermore, this heterodimerization may account for the limited access of Pit-1 to the CBP zinc fingers and therefore prevent the formation of the Pit-1–CBP complex required for PRL and GH promoter activity in pituitary cells.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
We maintained monolayer cultures of HeLa non-pituitary and GH3 rat pituitary tumor cells (Tashjian et al. 1968) in Dulbecco’s Modified Eagle Medium (DMEM), 20% fetal bovine serum (FBS) and 50 µg/ml penicillin and streptomycin, at 37 °C in 5% CO2 and changed the medium 16–18 h before each transfection. We harvested cells for transfection at approximately 80% confluence, using 0.05% trypsin and 0.5 mM EDTA.
Plasmids
Plasmid pA3 PRL luc, firefly luciferase driven by the proximal (-425) rat (r) PRL promoter, was described previously (Conrad & Gutierrez-Hartmann 1992, Diamond & Gutierrez-Hartmann 1996, 2000, Diamond et al. 1999). Plasmids pRSV HA Pit-1, pRSV HA Pit-1ß, pRSV PKAß and pRSV ß-globin, respectively driven by the RSV promoter, were described previously (Diamond & Gutierrez-Hartmann 1996, 2000, Diamond et al. 1999). Plasmid RSV CBP was the generous gift of Dr Chee-Gun Lee. Plasmid pCMX NCoR was the generous gift of Dr Kate Horwitz. Plasmid pGEX-Pit-1 was the generous gift of Dr Gutierrez-Hartmann.
Mutant Pit-1ß constructs in mammalian expression vectors were constructed by nested PCR mutagenesis of the Pit-1 transactivation domain as described previously (Diamond & Gutierrez-Hartmann 1996, 2000). The pRSV HA Pit-1 plasmid was used as a substrate for PCR mutagenesis, in which the 26 amino acid ß-domain was substituted with altered sequences, and a HA epitope tag was retained at the amino terminus of all of the Pit-1 constructs. Amplified DNA was initially subcloned into pCR 2.1 (Invitrogen, Carlsbad, CA, USA), and the presence of each introduced mutation and the integrity of its TAD region verified by dideoxy sequencing. HA-tagged Pit-1 TAD sequences were then excised from pCR2.1 by digestion with HinD III and PpuM I and were ligated to the unique HinD III and PpuM I sites of pRSV-Pit-1.
Expression vectors for labeled Pit-1 and Pit-1ß, T7 Pit-1 and T7 Pit-1ß, were constructed by PCR amplification of full length rat Pit-1 and Pit-1ß, followed by ligation to pCR4 Blunt TOPO (Invitrogen). The GST-fusion expression vector CBP AA 312–440 was constructed by PCR amplification of mouse CBP nucleotides 936–1320 with primers encoding BamH I and Sma I sites, followed by ligation to pCR4 Blunt TOPO. The CBP insert was then subcloned to the unique BamH I and Sma I sites of pGEX 4T-1.
Transfection
We introduced DNA into GH3 pituitary or HeLa non-pituitary cells by electroporation as follows. Approximately 2–3 x 106 enzymatically dispersed cells were mixed with plasmid DNA in a sterile gene-pulse chamber and exposed to a controlled electrical field of 525 µFarads at 220 V, as described previously (Diamond & Gutierrez-Hartmann 1996, Ferry et al. 2005). We performed transient transfections in triplicate in at least three separate experiments. Cells from individual transfections then were maintained in DMEM, 20% FBS, 50 µg/ml penicillin and streptomycin, at 37 °C in 5% CO2.
A 1:3 Pit-1:Pit-1ß plasmid DNA ratio that results in equal levels of protein expression was used (Diamond & Gutierrez-Hartmann 1996, 2000, Diamond et al. 1999). We controlled the nonspecific effect of the RSV promoters upon transcription factor availability by including amounts of pRSV ß-globin plasmid DNA in all assays to render the total pRSV DNA concentration constant. Plasmid pRLC Renilla (15 ng) was included as an internal control for all transfections.
Luciferase assays
After incubation for 24 h, cells were harvested and assayed using the Dual Luciferase Reporter Assay System (Promega) and a Monolight 3010 Luminometer (Analytical Luminescence Laboratories, San Diego, CA, USA). Firefly luciferase light units for each transfection were normalized for Renilla luciferase light units. We expressed results as mean fold activation of the PRL promoter ± S.E.M. and used Fisher’s protected test of least significant differences to calculate the significance of differences between multiple conditions. In determining the ratios of two independent variables, the relative errors were added in quadrature to determine the S.E.M. of the ratio.
Western Blot analysis of HA-tagged Pit-1 proteins
Transient transfections were performed as above. Three aliquots of cells were pooled and harvested with PBS containing 3 mM EDTA, pelleted and resuspended in 2X Laemmli sample buffer, and passed through a 25 G needle seven times. Equal volumes of each extract were separated on 15% SDS polyacrylamide gels and transferred to Immobilon-P PVDF membrane (Millipore, Bedford, MA, USA). The HA-tagged Pit-1 proteins were detected with a mouse monoclonal anti-HA HRP-conjugated antibody (Roche Diagnostics, Indianapolis, IN, USA) and ECL Advance media (Amersham Biosciences, Piscataway, NJ, USA). Dilutions of 1:1000 of the anti-HA monoclonal antibody were used.
GST fusion protein synthesis
Recombinant fusion proteins GST only, GST-Pit-1, and GST-CBP AA 312–440 were prepared from bacterial extracts (Bradford et al. 1995, 2000). Overnight cultures of Escherichia Coli BL-21 (DE3) pLysS (Stratagene, La Jolla, CA, USA), transformed with plasmids pGEX 4T-1, pGEX Pit-1, or pGEX CBP 312–440 were diluted 1:10 in fresh Luria broth supplemented with ampicillin (50 µg/ml) and grown at 30 °C to an optical density at 600 nm of 0.4. Cultures were induced by addition of 1 mM isopropyl-ß-D-thiogalactopyranoside for 2 h at 30 °C. Bacterial cells were harvested by centrifugation and resuspended and lysed in 1/10 volume of B-Per (Promega, Madison, WI, USA), with the recommended concentration of complete protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN, USA). Cellular debris was removed by centrifugation. Supernatants were bound to glutathione–sepharose (Amersham-Pharmacia, Piscataway, NJ, USA) for 1 h at room temperature and washed extensively in PBS supplemented with protease inhibitors. Bound protein was analyzed by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and western blot with antibodies against GST. Protein concentration was measured by the Bradford assay (Bradford 1976).
In vitro protein interaction assays
Flourotect-Green (Promega, Madison, WI, USA) lysine-labeled Pit-1 and Pit-1ß proteins were synthesized in vitro with the TNT coupled transcription-translation reticulocyte lysate system (Promega), according to the manufacturer’s protocol. Equal amounts (2 µg) of GST fusion proteins were bound to glutathione–agarose beads and suspended in binding buffer (40 mM HEPES, 100 mM NaCl, 5 mM MgCl2, 0.5 mM EDTA, 0.5 mM PMSF, 0.05% Nonidet P-40, 1 mM dithiothreitol, pH 7.5) supplemented with protease inhibitors. In vitro-translated and labeled protein was incubated with immobilized GST, GST-Pit-1, or GST-CBP AA 312–440 in a final volume of 0.5 ml binding buffer containing 50 µg/ml ethidium bromide and mixed by rocking for 1 h at room temperature. Beads were collected by a rapid, 30 s centrifugation at 1000g, and then washed five times for 5 min each in 0.5 ml binding buffer containing 0.1% Triton X-100. Bound labeled proteins were eluted by heating to 80 °C in SDS sample buffer and analyzed by SDS-polyacrylamide gel electrophoresis and autoradiography (Bradford et al. 1995, 2000). Bands were quantitated using a Molecular Dynamics Typhoon laser-scanning densitometer with Imagequant software (Amersham Biosciences, NJ, USA).
Statistics
The data were analyzed using GraphPad Prism software (San Diego, CA, USA). Differences between groups were considered significant for P value <0.05 as analyzed by Fishers least significant differences multiple comparison procedure.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Pit-1 and Pit-1ß have a very limited expression pattern and are mainly found in the anterior pituitary gland. Our previous studies showed that Pit-1ß functioned as a repressor of the PRL promoter only in pituitary cells (Diamond & Gutierrez-Hartmann 1996, 2000, Diamond et al. 1999, Ferry et al. 2005). We speculated that this pituitary-specific repression might be the result of the heterodimerization between Pit-1 and Pit-1ß. To test this hypothesis, we measured the effect of increasing doses of Pit-1ß on activation of the proximal PRL promoter in HeLa non-pituitary cells in the presence of Pit-1. When transfected alone, both Pit-1 and Pit-1ß significantly increased PRL promoter activity by 12-fold over the vector only control (Fig. 2) Combination of equal amounts (10 µg) of each isoform resulted in a significant drop in the PRL promoter activity compared with each isoform alone. However, increasing the amount of Pit-1ß to 15, 20, 25 and 30 µg increased PRL promoter activity by 20-fold, 27-fold, 80-fold and 100-fold respectively. These results suggest that repression by the ß-domain requires the presence of both isoforms and that the interaction or heterodimerization between Pit-1 and Pit-1ß leads to decreased intracellular availability and consequently to reduced efficiency as transcriptional activators in HeLa cells.
|
, there was no effect on PRL promoter activity when HeLa cells were transfected with CBP alone. Addition of 10 µg of either Pit-1ß or Pit-1 significantly increased PRL promoter activity 9-fold compared with vector only. Again, the further addition of increasing amounts of 15, 20, 25, and 30 µg Pit-1ß resulted in augmented PRL promoter activity by 12, 19, 25, 33-fold respectively. Interestingly, addition of equal inputs of Pit-1 and Pit-1ß did not repress PRL promoter activity in the presence of CBP. It is possible that CBP binds Pit-1 before Pit-1ß does and thus prevents the heterodimerization between Pit-1 and Pit-1ß. As well, interactions between all three players may take place simultaneously and produce a sufficient amount of Pit-1–CBP complex to activate the PRL promoter.
|
). Either Pit-1 or Pit-1ß in combination with CBP and PKAß activated the PRL promoter approximately 7-fold compared with the vector only control. Addition of both isoforms in combination with CBP and PKAß resulted in a significant drop in the PRL promoter activity. However, addition of 15, 20, 25, and 30 µg Pit-1ß increased PRL promoter activity in a dose-dependent manner between 7 and 10-fold. Pit-1ß repression was more pronounced in the absence of CBP. Interestingly, activation of the PKC signaling pathway also allows reconstitution of repression by Pit-1ß in the presence of Pit-1 and CBP, while expression of N-CoR does not allow reconstitution of repression in the presence of any combination of Pit-1ß, Pit-1 and CBP (data not shown). These experiments reconstitute, for the first time, repression by Pit-1ß in non-pituitary cells, and they indicate that the ß-domain blocks the ability of Pit-1 and CBP to transduce cAMP/PKA activation of the PRL promoter.
|
We next sought to address the question whether Pit-1ß might repress PRL promoter activity by blocking Pit-1 dimerization itself, or by blocking a downstream action of Pit-1 dimers such as recruitment of phosphorylated CBP. In order to test this possibility, we examined the ability of Pit-1 vs Pit-1ß to physically interact with Pit-1 and CBP in vitro through a GST pull-down assay. We chose the region of CBP containing AA 312–440 because it contains a CBP zinc-finger domain sufficient to bind Pit-1 in vitro, and functionally interact with Pit-1 in vivo (Tolon et al. 1998, Zanger et al. 1999).
Fusion proteins containing GST-linked Pit-1 and CBP AA 312–440 were constructed, immobilized on glutathione agarose, and used in pull-down assays with fluorescently labeled Pit-1 or Pit-1ß (see Materials and methods). Incubations were performed with equivalent amounts of beads, bound GST protein, bound GST-Pit-1 fusion protein, and bound GST-CBP AA 312–440 fusion protein in the presence of ethidium bromide to block nonspecific protein–DNA interactions (Lai & Herr 1992, Bradford et al. 2000). As shown in a representative experiment in Figure 5A, and quantified from seven such experiments in Figure 5B, Pit-1 bound both GST-Pit-1 and GST-CBP AA 312–440 at levels 4-fold above background, while Pit-1ß bound GST-Pit-1 at levels nearly 3-fold above background but failed to bind GST-CBP AA 312–440 at levels significantly above background. These results provide physical evidence that Pit-1ß interacts with Pit-1 but not with the AA 312–440 binding region of CBP, and that the ß-domain presence may render the heterodimer incapable of interacting
|
In order to address how the ß-domain insertion might affect Pit-1–CBP interaction, we analyzed the ß-domain sequence constraints on repression. Changes in AA sidechain length can eliminate binding between polypeptides by steric interference or by removing interacting sidechain elements from the same vicinity (Vu et al. 2001). In order to test the constraints on ß-domain sidechain length, we substituted the five residues previously shown to be required for repression (Leucine 7, Isoleucine 8, Tyrosine 17, Phenylalanine 18 and Methionine 20) with AAs of similar chemistry but different sidechain length (Fig. 6). For Leucine 7 and Isoleucine 8, we substituted Valine and Methionine; for Tyrosine 17, we substituted Tryptophan and Phenylalanine; for Phenylalanine 18, we substituted Tryptophan and Tyrosine; for Methionine 20, we substituted Valine and Isoleucine.
|
) (Camper et al. 1985, Gutierrez-Hartmann et al. 1987, Elsholtz et al. 1991) (See Materials and methods). Pit-1ß increased transcription of the PRL promoter 17-fold compared with the vector-only control, as seen previously (Diamond & Gutierrez-Hartmann 1996, Ferry et al. 2005). All of the sidechain length substitution mutants increased PRL promoter activity (from 3- to 21-fold). The variation in transcriptional potency in these experiments was similar to that among single amino acid substitutions in our previous substitution mutagenesis of the ß-domain (Diamond & Gutierrez-Hartmann 2000), and most of the mutants had similar transcription potency to wild-type Pit-1ß.
In order to determine the tolerance of repression for these alterations in sidechain-length, we introduced wild type and mutant Pit-1ßs into GH3 pituitary cells by electroporation with pA3 –425 PRL luc (See Materials and methods) (Fig. 6B). Pit-1ß reduced PRL promoter activity by 60% compared with the vector only control, and all of the sidechain-length mutants repressed PRL promoter activity by 41 – 85%. These data demonstrate that the sidechain lengths of these hydrophobic ß-domain residues are not necessary for repression.
The aliphatic/aromatic character of hydrophobic residues is not necessary for repression
Hydrophobic amino acids fall into two broad categories with different structures and chemical properties: aliphatic vs aromatic. Aliphatic sidechains are unsubstituted hydrocarbon chains subject to nucleophilic substitution, while aromatic sidechains incorporate closed planar rings and are subject to electrophilic substitution. These differences affect protein conformation and intermolecular interactions with ligands and other proteins (Heyl et al. 1994, Deardorff & Sachs 1997, Murray et al. 1998, Nakagawa et al. 2000).
For the aliphatic residues Leucine 7, Isoleucine 8, and Methionine 20, we substituted the aromatic residues Tyrosine and Phenylalanine. For the aromatic residues Tyrosine 17 and Phenylalanine 18, we substituted the aliphatic residues Leucine and Isoleucine. In addition, we constructed two multiple substitution constructs, where multiple hydrophobic residues were similarly substituted: 4 substitutions (Leu 7 Tyr, Ile 8 Phe, Tyr 17 Leu, Phe 18 Ile) and 5 substitutions (Leu 7 Tyr, Ile 8 Phe, Tyr 17 Leu, Phe 18 Ile, Met 20 Phe).
When expressed alone in HeLa non-pituitary cells (Fig. 7A), the sidechain-length substitution mutants increased rPRL promoter activity compared with the vector-only control from 9- to 104-fold versus 54-fold for Pit-1ß. The differences in transcriptional potency in these experiments were smaller than those in our previous substitution mutagenesis of the ß-domain (Diamond & Gutierrez-Hartmann 2000), but most of the mutants had a somewhat reduced transcription potency compared with wild-type Pit-1ß. Of note, the four- and five-substitution mutants were among the least transcriptionally active of the constructs. This is reminiscent of the more severe effects on transcription potency among larger epitope substitutions (Diamond & Gutierrez-Hartmann 2000). These data demonstrate that the ß-domain mutants retain transcriptional potency.
|
) compared with the vector-only control. Surprisingly, the ß-domain tolerated substitution of four or five hydrophobic residues. These data demonstrate that the aliphatic vs aromatic chemistry of Leucine 7, Isoleucine 8, Tyrosine 17, Phenylalanine 18, and Methionine 20 is not required for repression. ß-domain amino acid order is not required for repression
The lack of constraint on sidechain chemistry for these five hydrophobic amino acids suggested that the ß-domain might function solely through its hydrophobicity. A prediction of such a model would be that sequence order would not be necessary for repression, so long as hydrophobicity was preserved. We used a random number generator to scramble the ß-domain sequence in three different ways. We then substituted a scrambled domain for the wild-type ß-domain in mutant constructs (Table 1).
|
). They increased PRL promoter activity 7- to 15-fold compared with the vector-only control, vs 18-fold for wild-type Pit-1ß. In GH3 pituitary cells, the scrambled ß-domain-containing proteins reduced PRL promoter activity (Fig. 8B) between 69 and 77% compared with the vector only control. These data demonstrate that ß-domain amino acid order is not required for repression.
|
If the ß-domain functions solely through its hydrophobicity, other hydrophobic ß-domain sidechains should be required for repression. We sequentially substituted alanines for five untested hydrophobic ß-domain residues –valine 1, isoleucine 4, leucine 5, leucine 14, and methionine 23. Alanine substitution has been used in several systems to probe structure and function by eliminating most sidechain chemistry (Wertman et al. 1992, Diamond & Gutierrez-Hartmann 2000) reviewed earlier by Diamond & Kirkegaard 1994. Alanine is common (Klapper 1977), uncharged, and has minimal effects on protein secondary structure. While routinely categorized as hydrophobic because of its aliphatic methyl sidechain, it is actually no more hydrophobic than Threonine (Karplus 1997), and is found as often at solvent-exposed as at internal positions (Chothia 1976, Rose et al. 1985). Moreover, since there are no alanines in the ß-domain (Fig. 1), each substitution is distinct from the endogenous sequence.
All of the alanine-substituted Pit-1ß proteins retained basal transcriptional potency in isoform-insensitive HeLa non-pituitary cells, and activated the PRL promoter when compared with the vector only control (Fig. 9A). Pit-1ß activated the PRL promoter 18-fold, and the five alanine-scanning mutants activated the promoter between 21- and 98-fold. In contrast to the sidechain-length and aliphatic vs aromatic mutations, the five alanine-substitution mutants activated the PRL promoter when compared with the vector only control even more than did wild type Pit-1ß.
|
). Two of the alanine-substitution proteins, Val 1 Ala and Met 23 Ala, failed to reduce PRL promoter activity, and three of the alanine-substitution proteins actually increased PRL promoter activity between 2- and 3-fold. In our previous alanine-substitution mutagenesis studies, we also observed varying degrees of reversal of repression, from loss of repression to activation (Diamond & Gutierrez-Hartmann 2000).
The lack of effect of the Val 1 Ala and Met 23 Ala constructs on rPRL promoter activity in GH3 pituitary cells raised the concern that the mutant proteins may not be expressed, though their retention of transcription potency in HeLa non-pituitary cells argues against considerably altered protein expression. We performed western blot analysis of protein expression for these constructs in GH3 cells (Fig. 9C). The protein expression levels of the mutant constructs were roughly equivalent to that of wild-type Pit-1ß. We detected small differences in apparent mobility among the Pit-1 mutant constructs, likely due to sequence-specific effects on gel mobility of the mutant proteins, similar to those that we have seen previously with Pit-1 (Diamond & Gutierrez-Hartmann 1996, 2000) and the poliovirus RNA polymerase (Diamond & Kirkegaard 1994).
Nonhydrophobic residues are not required for repression
A further prediction of ß-domain function solely through hydrophobicity would be that nonhydrophobic side-chains, unlike the hydrophobic sidechains, would be dispensable for repression. To test this hypothesis, we replaced nonhydrophobic ß-domain amino acids with alanine. We chose five nonhydrophobic AAs (serine 6, glutamine 9, lysine 12, cysteine 13, and threonine 21) whose physical properties (Kawashima et al. 1999, Kawashima & Kanehisa 2000) had been changed in previous mutageneses only by multiple AA substitution mutations that eliminated repression (Diamond & Gutierrez-Hartmann 1996, 2000).
All of the alanine substitutions retained transcriptional potency in HeLa non-pituitary cells (Fig. 10A). Pit-1ß activated the PRL promoter 7-fold compared with the vector-only control, and the five alanine-scanning mutants activated 2- to 10-fold. For three of the mutants (Ser 6 Ala, Cys 13 Ala and Thr 21 Ala) transcriptional potency was less than that of the wild-type Pit-1ß, and for two of the mutants (Gln 9 Ala and Lys 12 Ala) transcriptional potency was greater than for that of the wild type.
|
). These data demonstrate that the non-hydrophobic ß-domain amino acids serine 6, glutamine 9, lysine 12, cysteine 13, and threonine 21 are not required for repression. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The results of the reconstitution experiments (Figs 2–4) indicate that the decrease in CBP recruitment to the PRL promoter seen previously (Ferry et al. 2005) are due to interference by Pit-1ß with the function of the Pit-1–CBP complex to transduce cAMP/PKA signaling. It has been shown that the POU domain of Pit-1 may interact with CBP through two cysteine–histidine rich domains (Xu et al. 1998, Zanger et al. 1999) and that POU factors can be modified by phosphorylation. For example, Pit-1 is phosphorylated in the POU domain (Thr-220). This phosphorylation results in decreased DNA binding to GH sites and increased DNA binding to TSHß and PRL promoter sites. The failure of Pit-1 alone, CBP or even PKA alone seen previously (Diamond & Gutierrez-Hartmann 1996, Diamond et al. 1999) to allow Pit-1ß to reconstitute repression suggest that this repression is an emergent property of Pit-1ß’s interaction with the Pit-1–CBP complex.
The correlation between the dosage of Pit-1ß and the degree of activation of the PRL promoter by Pit-1, PKA and CBP (Figs 2–4) is typical of the dosage effects seen with dominant-negative poison products, which typically interact with one partner of a complex but fail to interact with others (Mains et al. 1990, Mounkes & Fuller 1999). It has been shown that Pit-1 and non-liganded nuclear receptors (NR) can repress each other’s signaling (Gonzalez & Carlberg 2002). In this study, we propose a mechanism by which Pit-1 interacts with Pit-1ß resulting in cross-repression. This phenomenon may take place whenever Pit-1 and Pit-1ß are co-expressed. Our finding that the ß-domain blocks physical interaction between Pit-1 and CBP AA 312–440 (Fig. 5) provides a physical basis for this model. Thus, the cell-type specificity of repression by Pit-1ß would be due to the unique environment of anterior pituitary lactotroph and somatotrophs, which display tonic activation of the cAMP/PKA signaling cascade, and high levels of both Pit-1 and CBP.
The negative effect of the ß-domain on both Ets-1 and Ras signaling transduction (Diamond & Gutierrez-Hartmann 1996) as well as the differential interaction of Pit-1 and Pit-1ß with Ets-1 (Bradford et al. 2000) may also be a product of the insertion of the hydrophobic ß-domain into the Pit-1 TAD. The ß-domain insertion-point lies at the boundary between the two TAD exons, R1 and R2, which regulate Pit-1–ETS-1 synergy and Ras response respectively (Duval et al. 2003). The lack of requirement for physical interaction between Ets-1 and Pit-1 for either Ras responsiveness or Pit-1–ETS-1 synergy suggests that the ß-domain may interfere with coactivator recruitment by the Pit-1 and Ets-1 complex. Indeed, both Pit-1 and Ets-1 can interact with the AA 312–440 region of CBP.
The AA sequence requirements for ß-domain repression are most consistent with a model of ß-domain function solely through hydrophobicity. Current models of the function of transcriptional activator require specific charge and hydrophobic interactions to strengthen binding between conformationally flexible TADs and partner proteins (Hermann et al. 2001, De Guzman et al. 2004, Razeto et al. 2004, Zor et al. 2004). The lack of constraint for sidechain length (Fig. 6), sidechain chemistry (Fig. 7) or even AA order (Fig. 8) argues against the ß-domain forming one side of a specific interface that involves surface complementarity. The requirement for only the ß-domain’s hydrophobic (Figs 9 and 10) residues argues against a requirement for charge interactions, and suggests a threshold effect: the ß-domain may be hydrophobic enough to block interaction between Pit-1ß and CBP, but losses of hydrophilic properties do not increase the repression.
The effect of the ß-domain mutations on transcription potency when expressed alone in HeLa cells correlates with their effect on repression in pituitary cells. This suggests that changes in ß-domain sidechain structure have similar effects on both basal transcription potency in nonpituitary cells and the ability of Pit-1ß to interact with CBP. The mechanism by which either Pit-1 or Pit-1ß transduce PKA signaling in HeLa cells in the absence of CBP is clearly different from the mechanism in pituitary cells or in our current reconstitution experiments with CBP, and underlines the importance of characterizing mutant transcription factors in their native environment.
How might the hydrophobic ß-domain actually block interaction with and recruitment of CBP? The hydrophobicity of the ß-domain is reminiscent of that of the hydrophobic rtARNTa repression domain (Necela & Pollenz 1999), which can repress promoter activation by either ARNT or the yeast transcription factor Gal 4 through protein misfolding, possibly due to hydrophobic sidechain burial in the protein’s interior. Similarly, the ß-domain might alter the conformation of the Pit-1, perhaps directly affecting the POU-homeodomain, which has been shown to directly interact with CBP (Xu et al. 1998, Zanger et al. 1999). Alternatively, the ß-domain might interfere with the conformational flexibility of the TAD, and prevent necessary interactions between Pit-1 and CBP. Two of CBP’s zinc-finger domains in the AA 312–440 region form stable, well-ordered complexes with transcription factor TADs, which appear to involve the induced fit of an otherwise disordered TAD with a CBP zinc finger (Dial et al. 2003, Freedman et al. 2003).
Support for a model in which the ß-domain alters the Pit-1 TAD to interfere with recruitment of CBP comes from the analysis of a mutation in Pit-1 that lead to familial combined pituitary hormone deficiency (CPHD), a genetic disorder of anterior pituitary function in which multiple anterior pituitary hormone genes are under-expressed. Twenty mutations in the Pit-1 locus have been identified that lead to CPHD (Li et al. 1990, Ohta et al. 1992, Pfaffle et al. 1992, Radovick et al. 1992, Tatsumi et al. 1992, Cohen et al. 1995, 1999, Irie et al. 1995, Pellegrini-Bouiller et al. 1996, Brown et al. 1998, Fofanova et al. 1998, Pernasetti et al. 1998, Pernasetti et al. 1999, Blankenstein et al. 2001, Hendriks-Stegeman et al. 2001, Vallette-Kasic et al. 2001). Of particular interest is a mutation that converts the conserved Proline at position 24 of the TAD to Leucine, and interferes with interaction between Pit-1 and CBP (Ohta et al. 1992, Cohen et al. 1996). Interestingly, the dominant-negative Pro 24 Leu Pit-1, like Pit-1ß, fails to block either PRL or GH promoter activity in non-pituitary cells and like Pit-1ß, actually activates those promoters (Kishimoto et al. 2002).
Future experiments will investigate the mechanism(s) that underlie the ability of the ß-domain to block recruitment of CBP by Pit-1. Owing to its dominant-negative function with regard to the functional recruitment of CBP to the PRL promoter, Pit-1ß also may provide a probe to identify additional downstream targets of Pit-1 and CBP.
|
Acknowledgements |
|---|
This work was supported by the American Cancer Society (RSG TBE-105036), the National Institutes of Health (NIDDK K01 DK02752 and NCRR P20 RR15592), and the Kentucky Research Challenge Trust Fund. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Bodner M & Karin M 1987 A pituitary-specific trans-acting factor can stimulate transcription from the growth hormone promoter in extracts of nonexpressing cells. Cell 50 267–275.[CrossRef][ISI][Medline]
Bradford AP, Conrad KE, Wasylyk C, Wasylyk B & Gutierrez-Hartmann A 1995 Functional interaction of c-ets-1 and ghf-1/Pit-1 mediates Ras activation of pituitary-specific gene expression: mapping of the essential c-ets-1 domain. Molecular and Cellular Biology 15 2849–2857.[Abstract]
Bradford A, Brodsky KS, Diamond S, Kuhn L, Liu Y & Gutierrez-Hartmann A 2000 The Pit-1 homeodomain and beta-domain interact with ets-1 and modulate synergistic activation of the rat prolactin promoter. Journal Biological Chemistry 275 3100–3106.
Bradford M 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72 248–254.[CrossRef][ISI][Medline]
Brown M, Parks J, Adess M, Rich B, Rosenthal I, Voss T, Vanderheyden TC & Hurley DL 1998 Central hypothyroidism reveals compound heterozygous mutations in the Pit-1 gene. Hormone Research 49 98–102.[CrossRef][ISI][Medline]
Camper SA, Yao YAS & Rottman FM 1985 Hormonal regulation of the bovine prolactin promoter in rat pituitary tumor cells. Journal of Biological Chemistry 260 12246–12251.
Chothia C 1976 The nature of the accessible and buried surfaces in proteins. Journal of Molecular Biology 105 1–12.[CrossRef][ISI][Medline]
Cohen L, Wondisford F, Salvatoni A, Maghnie M, Brucker-Davis F, Weintraub B & Radovick S 1995 A "hot spot" in the Pit-1 gene responsible for combined pituitary hormone deficiency: clinical and molecular correlates. Journal of Clinical Endocrinology and Metabolism 80 679–684.[Abstract]
Cohen L, Wondisford F & Radovick S 1996 Role of Pit-1 in the gene expression of growth hormone, prolactin, and thyrotropin. Endocrinology and Metabolism Clinics of North America 25 523–540.[CrossRef][ISI][Medline]
Cohen LE, Hashimoto Y, Zanger K, Wondisford F & Radovick S 1999 Creb-independent regulation by CBP is a novel mechanism of human growth hormone gene expression. Journal of Clinical Investigation 104 1123–1130.[ISI][Medline]
Conrad KE & Gutierrez-Hartmann A 1992 The Ras and protein kinase A pathways are mutually antagonistic in regulating rat prolactin promoter activity. Oncogene 7 1279–1286.[ISI][Medline]
Day RN, Walder JA & Maurer RA 1989 A protein kinase inhibitor gene reduces both basal and multihormone-stimulated prolactin gene transcription. Journal of Biological Chemistry 264 431–436.
De Guzman RN, Martinez-Yamout MA, Dyson H & Wright P 2004 Interaction of the TAZ1 domain of the CREB-binding protein with the activation domain of CITED2: regulation by competition between intrinsically unstructured ligands for non-identical binding sites. Journal of Biological Chemistry 279 3042–3049.
Deardorff J & Sachs A 1997 Differential effects of aromatic and charged residue substitutions in the RNA binding domains of the yeast poly(A)-binding protein. Journal of Molecular Biology 269 67–81.[CrossRef][ISI][Medline]
Dial R, Sun Z & Freedman S 2003 Three conformational states of the p300 CH1 domain define its functional properties. Biochemistry 42 9937–9945.[CrossRef][Medline]
Diamond S & Gutierrez-Hartmann A 1996 A 26-amino acid insertion domain defines a functional transcription switch motif in Pit-1 beta. Journal of Biological Chemistry 271 28925–28932.
Diamond SE & Gutierrez-Hartmann A 2000 The Pit-1 beta domain dictates active repression and alteration of histone acetylation of the proximal prolactin promoter. Journal of Biological Chemistry 275 30977–30986.
Diamond S & Kirkegaard K 1994 Clustered charged-to-alanine mutagenesis of poliovirus RNA-dependent RNA polymerase yields multiple temperature-sensitive mutants defective in RNA synthesis. Journal of Virology 68 863–876.
Diamond SE, Chiono M & Gutierrez-Hartmann A 1999 Reconstitution of the protein kinase A response of the rat prolactin promoter: differential effects of distinct Pit-1 isoforms and functional interaction with Oct-1. Molecular Endocrinology 13 228–238.
Duval D, Jean A & Gutierrez-Hartmann A 2003 Ras signaling and transcriptional synergy at a flexible ets-1/Pit-1 composite DNA element is defined by the assembly of selective activation domains. Journal of Biological Chemistry 278 39684–39696.
Elsholtz H, Albert V, Treacy M & Rosenfeld M 1990 A two-base change in a POU factor-binding site switches pituitary-specific to lymphoid-specific gene expression. Genes and Development 4 43–51.
Elsholtz HP, Lew AM, Albert PR & Sundmark VC 1991 Inhibitory control of prolactin and Pit-1 gene promoters by dopamine. Journal of Biological Chemistry 266 22919–22925.
Ferry A, Jonsen M, Locasto DM, Brodsky K, Gutierrez-Hartmann A & Diamond SE 2005 Pit-1 beta blocks transcription in a DNA context-dependent manner. Journal of Endocrinology 185 173–185.
Fofanova O, Takamura N, Kinoshita E, Yoshimoto M, Tsuji Y, Peterkova V, Evgrafov O, Dedov I, Goncharov N & Yamashita S 1998 Rarity of Pit1 involvement in children from russia with combined pituitary hormone deficiency. American Journal of Medical Genetics 77 360–365.[CrossRef][ISI][Medline]
Freedman S, Sun Z, Kung A, France D, Wagner G & Eck M 2003 Structural basis for negative regulation of hypoxia-inducible factor-1 alpha by CITED2. Nature Structural Biology 10 504–512.[CrossRef][ISI][Medline]
Gonzalez M & Carlberg C 2002 Cross-repression, a functional consequence of the physical interaction of non-liganded nuclear receptors and POU domain transcription factors. Journal Biological Chemistry 277 18501–18509.
Gutierrez-Hartmann A, Siddiqui S & Loukin S 1987 Selective transcription and DNase I protection of the rat prolactin gene by GH3 pituitary cell-free extracts. PNAS 84 5211–5215.
Hashimoto K, Zanger K, Hollenberg A, Cohen L, Radovick S & Wondisford F 2000 Camp response element-binding protein-binding protein mediates thyrotropin-releasing hormone signaling on thyrotropin subunit genes. Journal of Biological Chemistry 275 33365–33372.
Hendriks-Stegeman BI, Augustijn K, Bakker B, Holthuizen P, van der Vliet PC & Jansen M 2001 Combined pituitary hormone deficiency caused by compound heterozygosity for two novel mutations in the POU domain of the Pit1/POU1F1 gene. Journal of Clinical Endocrinology and Metabolism 86 1545–1550.
Hermann S, Berndt K & Wright A 2001 How transcriptional activators bind target proteins. Journal of Biological Chemistry 276 40127–40132.
Herr W, Sturm RA, Clerc RG, Corcoran LM, Baltimore D, Sharp PA, Ingraham HA, Rosenfeld MG, Finney M, Ruvkun G & Horvitz HR 1988 The POU domain: a large conserved region in the mammalian pit-1, oct-1, oct-2, and caenorhabditis elegans unc-86 gene products. Genes and Development 2 1513–1516.
Heyl D, Schmitter S, Bouzit H, Johnson T, Hepp A, Kurtz K & Mousigian C 1994 Substitution of aromatic and nonaromatic amino acids for the Phe3 residue in the delta-selective opioid peptide deltorphin I: effects on binding affinity and selectivity. International Journal of Peptides and Protein Research 44 420–426.
Ingraham HA, Chen RP, Mangalam HJ, Elsholtz HP, Flynn SE, Lin CR, Simmons DM, Swanson L & Rosenfeld MG 1988 A tissue-specific transcription factor containing a homeodomain specifies a pituitary phenotype. Cell 55 519–529.[CrossRef][ISI][Medline]
Ingraham HA, Flynn SE, Voss JW, Albert VR, Kapiloff MS, Wilson L & Rosenfeld MG 1990 The POU-specific domain of Pit-1 is essential for sequence-specific, high affinity DNA binding and DNA-dependent Pit-1–Pit-1 interactions. Cell 61 1021–1033.[CrossRef][ISI][Medline]
Irie Y, Tatsumi K, Kusuda S, Kawawaki H, Boyages SC, Nose O, Ichiba Y, Katsumata N & Amino N 1995 Screening for PIT1 abnormality by PCR direct sequencing method. Thyroid 5 207–211.[ISI][Medline]
Iverson R, Day K, D’emden M, Day R & Maurer R 1990 Clustered point mutation analysis of the rat prolactin promoter. Molecular Endocrinology 4 1564–1571.[Abstract]
Kapiloff MS, Farkash Y, Wegner M & Rosenfeld MG 1991 Variable effects of phosphorylation of Pit-1 dictated by the DNA response elements. Science 253 786–789.
Karplus P 1997 Hydrophobicity regained. Protein Science 6 1302–1307.[Abstract]
Katz D, Reginato M & Lazar M 1995 Functional regulation of thyroid hormone receptor variant TR alpha 2 by phosphorylation. Molecular and Cellular Biology 15 2341–2348.[Abstract]
Kawashima S & Kanehisa M 2000 AAindex: amino acid index database. Nucleic Acids Research 28 374.
Kawashima S, Ogata H & Kanehisa M 1999 AAindex: amino acid index database. Nucleic Acids Research 27 368–369.
Keech CA & Gutierrez-Hartmann A 1989 Analysis of rat prolactin promoter sequences that mediate pituitary-specific and 3',5'-cyclic adenosine monophosphate-regulated gene expression in vivo. Molecular Endocrinology 3 832–839.[Abstract]
Keech C, Jackson S, Siddiqui S, Ocran K & Gutierrez-Hartmann A 1992 Cyclic adenosine 3',5'-monophosphate activation of the rat prolactin promoter is restricted to the pituitary-specific cell type. Molecular Endocrinology 6 2059–2070.[Abstract]
Kievit P & Maurer R 2005 The pituitary-specific transcription factor, Pit-1, can direct changes in the chromatin structure of the prolactin promoter. Molecular Endocrinology 19 138–147.
Kishimoto M, Okimura Y, Yagita K, Iguchi G, Fumoto M, Iida K, Kaji H, Okamura H & Chihara K 2002 Novel function of the transactivation domain of a pituitary-specific transcription factor, Pit-1. Journal of Biological Chemistry 277 45141–45148.
Klapper M 1977 The independent distribution of amino acid near neighbor pairs into polypeptides. Biochemical and Biophysical Research Communications 78 1018–1024.[ISI][Medline]
Konzak KE & Moore DD 1992 Functional isoforms of Pit-1 generated by alternative mRNA splicing. Molecular Endocrinology 6 241–247.[Abstract]
Lacerte A, Lee E, Reynaud R, Canaff L, De Guise C, Devost D, Ali S, Hendy G & Lebrun JJ 2004 Activin inhibits pituitary prolactin expression and cell growth through SMADs, Pit-1 and menin. Molecular Endocrinology 18 1558–1569.
Lai J & Herr W 1992 Ethidium bromide provides a simple tool for identifying genuine DNA-independent protein associations. PNAS 89 6958–6962.
Lew A & Elsholtz H 1995 A dopamine-responsive domain in the N-terminal sequence of Pit-1. Transcriptional inhibition in endocrine cell types. Journal of Biological Chemistry 270 7156–7160.
Li S, Crenshaw EB III, Rawson E, Simmons DM, Swanson LW & Rosenfeld M 1990 Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1. Nature 347 528–533.[CrossRef][Medline]
Mains P, Sulston I & Wood W 1990 Dominant maternal-effect mutations causing embryonic lethality in caenorhabditis elegans. Genetics 125 351–369.[Abstract]
-helices are represented graphically. Lower, Pit-1ß, with the location of the 26 amino acid ß-domain and its sequence.
