(Received for publication, January 16, 1997)
From the Department of Cell Biology, Baylor College of Medicine, Houston, Texas 77030
Many steroid receptors, including chicken
progesterone receptor, have been shown to be activated in the absence
of their cognate ligands by modulators of kinases and phosphatases. To
investigate the molecular mechanism of ligand-independent activation,
chicken progesterone receptor mutants in which either one or all four of the previously identified phosphorylation sites have been changed to
nonphosphorylatable alanine were analyzed for their ability to be
activated by progesterone, 8-bromoadenosine 3:5
-cyclic monophosphate,
or a dopamine agonist, SKF82958. Our current study shows that the
receptor is differently phosphorylated in ligand-dependent and ligand-independent activation. The transcriptional activity of the
receptor in response to 8-bromoadenosine 3
:5
-cyclic monophosphate is
affected by mutation of either Ser211 or
Ser260. In addition, our data demonstrated that none of the
four sites is absolutely required for the activation of the receptor by
either 8-bromoadenosine 3
:5
-cyclic monophosphate or the dopamine
agonist. Treatment with 8-bromoadenosine 3
:5
-cyclic monophosphate did not increase the overall level of receptor phosphorylation or cause
phosphorylation of the receptor at alternate sites. These data raise
the possibility that ligand-independent activation of the chicken
progesterone receptor may be mediated through changes in the
phosphorylation of coregulators or other protein factors interacting
with the receptors.
In contrast to peptide hormones whose signals are transduced into the nucleus through membrane receptor-activated signal transduction pathways, steroid hormones act through intracellular receptors which themselves are ligand-regulated transcription factors (1-7). The lipophilic hormones diffuse through the cell membrane, bind the receptors inside the cell, and transform them into active transcription factors. In this way, the signal carried by steroids is directly transduced into long term changes in gene expression without an absolute requirement for a phosphorylation cascade to mediate the effect.
Studies in recent years have provided increasing evidence that there is cross-talk between the signaling processes induced by growth factors and steroids, and phosphorylation plays an important role in these processes. First, most, if not all, steroid receptors are phosphoproteins (8-14). Second, functional analyses of phosphorylation site mutants have demonstrated that phosphorylation regulates the transcriptional activity of many steroid receptors (14-19). Third, modulators of various signal transduction pathways have been shown to regulate the ligand-stimulated transcriptional activity of the receptors (20-29). Most importantly, many steroid receptors have been shown to be activated in the absence of their cognate ligands by modulation of protein kinase or phosphatase activity (22, 24, 25, 30-35). Ligand-independent activation of steroid receptors may have important physiological and clinical implications for the study and treatment of the tumors of hormone-responsive organs. Very little is known about the molecular mechanism of ligand-independent activation for any of the receptors. Since steroid receptors are phosphoproteins, it is possible that alteration of receptor phosphorylation in response to different treatments mediates the ligand-independent activation.
Chicken progesterone receptor (cPR)1 is one
of the steroid receptors that has been shown to be activated in a
ligand-independent manner by modulators of kinases and phosphatases
such as 8-bromoadenosine 3:5
-cyclic monophosphate (8-bromo-cAMP),
okadaic acid, vanadate, growth factors (e.g. EGF), and
neurotransmitters (e.g. dopamine) (25, 30, 32). The
phosphorylation sites in cPR have been identified (8, 36), making it
possible to answer the question of whether the phosphorylation of the
receptor is involved in or mediates the ligand-independent activation
of the receptor.
cPR is expressed as two forms, cPRB and cPRA, which lack the amino-terminal 128 amino acids of cPRB. Like the other members of the steroid/thyroid superfamily, cPR is composed of separable domains such as domains for DNA binding, hormone binding, and transcriptional activation (37-39). Different from the DNA binding and hormone binding domains which are not further separable, two separable transcriptional activation domains have been identified in chicken progesterone receptor (37, 40). The activation function 2 (AF-2, previously named transactivation function 2 or TAF-2), within the hormone binding domain is known to be regulated by hormone. In contrast to AF-2, the activation function 1 (AF-1, previously named transcriptional activation function 1 or TAF-1) is located in the amino-terminal region (37, 40). The fact that AF-1 is located in the sequence outside the hormone binding domain raises the possibility that the activity of AF-1 might be regulated by means such as phosphorylation rather than by ligand binding. Consistent with this idea, three out of the four phosphorylation sites identified in cPR are located in the A/B region (8, 36).
Based on the primary sequence of cPRB, the four identified phosphorylation sites are Ser211, Ser260, Ser367, and Ser530 (8, 36). All four sites are in Ser-Pro motifs (8, 36), and they account for all the Ser-Pro motifs in the receptor. The same phosphorylation pattern was detected in both cPRA and cPRB and is conserved between the endogenous receptor isolated from chicken oviduct and the recombinant receptor expressed in and purified from yeast (41). Among the four sites, Ser211 and Ser260 are basally phosphorylated, but their phosphorylation is enhanced in response to hormone stimulation; Ser367 and Ser530 are phosphorylated primarily in response to progesterone treatment (8, 36).
To examine whether any of the known phosphorylation sites in cPR is involved in or mediates the ligand-independent activation of the receptor, the four phosphorylation sites were mutated either individually or simultaneously to nonphosphorylatable alanines, and the responses of these mutant receptors to 8-bromo-cAMP and a dopamine agonist were assayed and compared with that of the wild type receptor. Our data show that receptor phosphorylation is not absolutely required for ligand-independent activation, raising the possibility that the ligand-independent activation of cPR might be mediated through changes in the phosphorylation of receptor-associated proteins such as coactivators and chaperone proteins that are known to play important roles in receptor function.
All cell culture reagents were purchased from Life Technologies, Inc. [3H]Chloramphenicol and carrier-free [32P]H3PO4 were purchased from DuPont NEN. N-Butyryl-coenzyme A and Protein-A Sepharose were purchased from Pharmacia Biotech Inc. The T7-Gen in vitro mutagenesis kit and Sequenase version 2.0 sequencing kit were purchased from United States Biochemical Corp. The oligonucleotides used in the mutagenesis and sequencing were synthesized by GenoSys (The Woodlands, TX). Triethylamine, 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide, 8-bromo-cAMP, Triton X-100, 8-methoxypsoralen, sequencing grade trifluoroacetic acid, poly-L-lysine and 2,6,10,14-tetra-methylpentadecane were purchased from Sigma. Xylene was purchased from Fisher. Tosylphenylalanyl chloromethyl ketone-treated trypsin was obtained from Worthington. Phenylisothiocyanate and HPLC reagents were obtained from J. T. Baker Inc. The D1 dopamine agonist, SKF82958 (6-chloro-7,8-dihydroxy-3-allyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide) was obtained from Research Biochemicals Inc. (Natick, MA). Monoclonal antibody to the chicken progesterone receptor (PR22) was kindly provided by Dr. David Toft. All other chemicals are of reagent grade.
In Vitro Mutagenesis and Plasmid ConstructionSingle site mutations were generated using a conventional, non-polymerase chain reaction mutagenesis strategy as described previously (15). The mutant in which all four phosphorylation sites are changed to alanine was generated by exchanging the restriction fragments between the single site mutants. All mutations were confirmed by direct sequencing using a Sequenase version 2.0 sequencing kit. The subcloning of the mutant and the wild type cPRA into the expression vector, p91023B, has been fully described (15, 38). GRE2e1bCAT (provided by Dr. John Cidlowski) is a simple promoter-based reporter composed of two progesterone/glucocorticoid response elements (GREs), the TATA box from the adenovirus e1b gene, and the cDNA sequence for chloramphenicol acetyltransferase (CAT) (42).
Preparation of Inactivated Adenovirus for TransfectioncPRA was expressed in cells using a nonrecombinant adenoviral-mediated DNA transfer technique (31, 43) with a few modifications. Replication-deficient adenovirus, dl312, was grown in human 293 embryonic kidney cells. The virus was released from the cells by three successive freeze/thaws of the cell pellet in phosphate-buffered saline. Following centrifugation at 2,500 rpm in a clinical centrifuge, the resulting supernatant was layered on top of a cesium chloride step gradient (1.5, 1.35, and 1.25 g/ml CsCl). Following centrifugation of the CsCl gradient at 150,000 × g for 1 h at 10 °C, the adenovirus band (between the 1.25 and 1.35 g/ml layers) was removed and brought to a final concentration of 1.35 g/ml CsCl. The adenovirus was centrifuged again at 150,000 × g for 5 h at 10 °C, and the adenovirus band was collected and dialyzed against HEPES-buffered saline (150 mM NaCl, 20 mM HEPES, pH 7.3) for 16 h at 4 °C with 50,000 molecular weight cutoff dialysis tubing. The dialysate was exposed to short wavelength UV light (254 nm) for 3 min. 8-Methoxypsoralen was added to the supernatant at a final concentration of 0.33 mg/ml, and the sample was exposed to long wavelength UV light (366 nm) for 20 min. The preceding treatments inactivated the viral genome without significant loss of infectivity. These treatments were a modification from a previously published method designed to inactivate adenovirus particles (44, 45). 8-Methoxypsoralen was removed from the virus preparation by G25 gel filtration chromatography. Adenovirus (1.4 × 1011 particles) was mixed with 160 µl of 10 mg/ml poly-L-lysine and 6.5 µl of 80 mM 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide in a sterile polystyrene tube in a final volume of 4 ml of HEPES-buffered saline and incubated for 4 h on ice with mixing every 30 min. The coupled virus was brought to a final concentration of 1.35 g/ml CsCl. The adenovirus was centrifuged at 150,000 × g for 5 h at 10 °C, and the adenovirus band was collected and dialyzed against HEPES-buffered saline for 16 h at 4 °C. Viral DNA was quantitated by measuring the optical density at 260 nm. Briefly, 25 µl of purified virus was mixed with 465 µl of phosphate-buffered saline and 10 µl of 5% SDS. The mixture was vortexed for 2 min and then centrifuged for 2 min at room temperature in a microcentrifuge. The optical density of the resulting supernatant was measured at 260 nm (1 optical density unit equals approximately 1 × 1012 virus particles/ml).
Transfection and CAT AssaysCell lines were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with fetal bovine serum (FBS) and antibiotics (penicillin and streptomycin). Twenty-four hours before transfection, cells were plated in 6-well plates at a density of 1 × 105 cells per well in the same medium. Before transfection, cells were rinsed with Hank's balanced salt solution, and 2 ml of DMEM was added to each well. DNA and polylysine-coupled viruses were incubated for 30 min at room temperature, and additional polylysine (polylysine/DNA ratio is 1:1.3) was added. The DNA-virus-polylysine complex was incubated at room temperature for another 30 min. During the incubations, the DNA-virus complex was kept in the dark. After the incubations, the complex was added dropwise to the cells. Cell were incubated with DNA-virus complex for 2 h. Then 2 ml of DMEM supplemented with 10% stripped FBS was added to each well. On the next day, progesterone and compounds such as 8-bromo-cAMP at the indicated concentrations were added to the cells, and the cells were incubated for an additional 24 h before being harvested and assayed for CAT activity.
For CAT assays, cells were harvested by scraping and subsequently lysed by three cycles of freeze-thawing. The protein concentrations were determined using the Bio-Rad protein assay reagent according to the manufacturer's protocols. Equal amounts of protein (usually 5-10 µg) were then used for a liquid CAT assay that has been described (46) and used in our previous studies (15, 16, 32). The samples were heated at 60 °C for 8 min before the addition of the substrates, and the reactions were terminated and extracted before the substrates become a limiting factor to ensure the accurate determination of the CAT activity. Duplicate samples were assayed for each data point.
Metabolic LabelingCV-1 monkey kidney cells were plated on
150-mm dishes (2 × 106 cells/dish) and cultured at
37 °C with 5% CO2 in DMEM with 10% FBS that had been
stripped of steroid hormones by treatment with dextran-coated charcoal.
After 24 h in culture, cells were subjected to adenovirus
infection (virus to cell ratio, 750:1) with 0.2 µg of
cPRA expression vector/plate and cultured for an additional 24 h. To remove endogenous phosphate pools from the cells, the culture medium was removed, and phosphate-free DMEM was added to the
cells for 1 h at 37 °C. Subsequently, this medium was removed, and cells were cultured in phosphate-free DMEM with 1% dialyzed, stripped FBS. 6 mCi of [32P]H3PO4
(2 mCi/ml) was added to each plate of CV-1 cells, and cells were
cultured for 1 h at 37 °C. Treatment was as follows. For 10-h
treatments, either 8-bromo-cAMP (final concentration 2 mM)
or progesterone (final concentration 108 M)
was added to cell cultures 1 h after
[32P]H3PO4 addition, and cells
were cultured for 10 h prior to harvest. For 30-min treatments,
8-bromo-cAMP (final concentration 2 mM) was added to cell
cultures 10.5 h after
[32P]H3PO4 addition, and cells
were cultured for 30 min prior to harvest.
cPR protein was
purified and subjected to trypsin digestion and HPLC analysis as
described previously (8) with the following modifications. Briefly,
cells were scraped from the plates in phosphate-buffered saline and
centrifuged at 500 × g for 10 min at 4 °C. Cell
pellets were resuspended in homogenization buffer (50 mM
potassium phosphate, pH 7.4, 10 mM sodium molybdate, 50 mM sodium fluoride, 2 mM EDTA, 2 mM
EGTA, 0.4 M sodium chloride, 5 mM
-monothioglycerol) (8) with 1% Triton X-100 and vortexed for
30 s. Following centrifugation at 100,000 × g for
30 min at 4 °C, 3 volumes of homogenization buffer without Triton
X-100 was added to the supernatant to reduce the detergent
concentration. The receptor was then purified by passing the entire
supernatant over a 1-ml PR22-Protein-A-Sepharose immunoaffinity column
as described previously (8). Purified receptor was electrophoresed on a
6.5% SDS-PAGE gel, and the wet gel was exposed to X-AR film (Eastman
Kodak) for 2-4 h at 4 °C. The phosphorylated receptor band was cut
out of the gel, and the gel slice was digested with trypsin to prepare
the tryptic cPRA peptides (8). HPLC analysis of tryptic
peptides was performed as described previously (8). Briefly,
phosphopeptides were loaded onto a C-18 reverse phase HPLC column in
HPLC-grade H20 containing 0.1% trifluoroacetic acid.
Phosphopeptides were separated by elution from the C-18 column using a
0-45% gradient of acetonitrile containing 0.1% trifluoroacetic acid
over a period of 90 min.
Purified cPRA was analyzed by SDS-PAGE and transferred to a nitrocellulose membrane. The receptor was detected using the monoclonal antibody, PR22, followed by chemiluminescent detection using enhanced chemiluminescence reagent (Amersham Corp.) as described (47).
To determine the molecular mechanism of
the ligand-independent activation of cPR by 8-bromo-cAMP, we
individually mutated each of the four known phosphorylation sites to
alanine, and the transcriptional activity of each of the four
single-site cPRA mutants was analyzed and compared with the
wild type receptor after treatment with either 2 mM
8-bromo-cAMP or 10 nM progesterone. As shown in Fig.
1, the transcriptional activity of the
Ala211 and Ala260 mutants in response to either
progesterone or 8-bromo-cAMP is significantly lower than that of the
wild type. As reported previously (15), the activity of
Ala530 is not statistically different from that of the wild
type at this hormone concentration. The mutation of Ser367
to alanine appears not to affect the transcriptional activity of the
receptor. From Fig. 1, it is obvious that the transcriptional activity
of Ala211 and Ala260 in response to
8-bromo-cAMP is affected by the mutations to a magnitude similar to the
activity in response to progesterone, suggesting that the receptor
phosphorylation at these two sites can regulate the ligand-independent
activation. The fact that all four mutants can still be activated by
8-bromo-cAMP suggests that phosphorylation at any of the individual
sites is not absolutely required for the ligand-independent activation.
However, this experiment did not rule out the possibility that
phosphorylation of cPR at the remaining sites compensates for the loss
at one position in mediating the receptor activation.
Mutation of the Four Phosphorylation Sites Simultaneously to Alanine Does Not Abolish the Receptor's Ability to Be Activated by 8-Bromo-cAMP
To investigate the possibility that phosphorylations
at the four known sites compensate for one another during the
ligand-independent activation of the receptor, all four sites were
mutated simultaneously to alanines, and the activity of this quadruple
mutant (AAAA) was compared with that of the wild type after stimulation
with either 8-bromo-cAMP or 10 nM progesterone. As shown in
Fig. 2, the transcriptional activity of the AAAA mutant
in response to progesterone as well as to 8-bromo-cAMP is lower than
that of the wild type. However, it is obvious that the AAAA mutant can still be activated by 8-bromo-cAMP in the absence of hormone. Together
with the data shown in Fig. 1, this experiment suggests that
phosphorylation of the four sites modulates but is not absolutely required for the ligand-independent activation of cPR by
8-bromo-cAMP.
Phosphorylation at Any of the Four Sites Is Not Required for the Ligand-independent Activation of cPR by the D1 Dopamine Agonist, SKF82958
Dopamine is another compound that induces
ligand-independent activation of cPRA (25). Previous
studies have also indicated that the activation by dopamine is mediated
through the D1 subtype of dopamine receptor (25). To determine whether
phosphorylation at any of the four known sites is required for the
ligand-independent activation of cPRA by dopamine, cPR
mutants were analyzed for their ability to be activated by SKF82958, an
agonist for the D1 type of dopamine receptor. As shown in Fig.
3, the change in the transcriptional activity of the
mutant receptors in response to stimulation with SKF82958 is very
similar to that in response to progesterone and 8-bromo-cAMP. However,
all mutants including the AAAA mutant can still be activated by
SKF82958, suggesting that none of the phosphorylation sites is required
for the ligand-independent activation of cPRA by
SKF82958.
No Alternate Phosphorylation Was Detected in the AAAA Mutant Receptor
The identification of the four phosphorylation sites was
achieved by [32P]H3PO4 labeling
and purification of cPR protein from tissue minces of chicken oviducts.
It is conceivable that the receptor expressed in cultured cells could
be phosphorylated differently from the endogenous receptor.
Theoretically, it is also possible that mutation of the known
phosphorylation sites could cause the phosphorylation of the receptor
at alternate sites that might counteract the mutation of the known
sites and mediate the ligand-independent activation of the mutant
receptor. To address these questions, phosphopeptide mapping of the
wild type or the AAAA mutant receptors expressed in CV-1 cells was
performed. As shown in Fig. 4, panel A, three major phosphorylation peaks were detected in the wild type receptor in
response to progesterone treatment after HPLC analysis using a C-18
reverse phase column. These are indistinguishable from the
phosphorylation pattern of the endogenous cPR isolated from chicken
oviducts (8, 36). From our previous analysis (8, 36), it is known that
the first peak corresponds to the phosphorylation at
Ser260; the peak in the middle contains both
Ser211 and Ser367 sites, and the last peak is
Ser530. In contrast to the wild type receptor, no
significant phosphorylation was detected in the AAAA mutant in either
the presence (Fig. 4, panel B) or the absence (Fig. 4,
panel C) of 10 nM progesterone. The multiple
peaks of wild type cPRA that eluted at 50-60 min after
loading the sample (Fig. 4, panel A) are likely due to
incomplete digestion because they are not present in the AAAA mutant.
Similar heterogeneity around the 260 peak has been detected in
phosphopeptide mapping of the recombinant cPRA expressed in
yeast (41). These mapping analyses suggest that no additional
phosphorylation occurs in cPRA expressed in cultured cells,
and the mutation of the four known sites to alanine did not cause
alternate phosphorylation.
Ligand-independent Activation of cPRA Following Treatment with 8-Bromo-cAMP Does Not Enhance the Phosphorylation of the Receptor or Cause Alternate Phosphorylation
It is possible that
ligand-independent activation of steroid receptors could result in
early, transient alterations in phosphorylation or prolonged
alterations in phosphorylation of cPRA. To assess these
possibilities, cPRA expressed in CV-1 cells was labeled in vivo with
[32P]H3PO4 followed by both a
short treatment period (30 min) and a long treatment period (10 h) with
8-bromo-cAMP. When phosphorylation of cPRA following
different treatments was assessed in the same experiment, it was clear
that only treatment with progesterone resulted in a significant
enhancement of the receptor phosphorylation (Fig. 5,
panel A; compare lane 4 with lanes
1-3). The increase in the amount of receptor protein detected
following steroid treatment (Fig. 5, panel B, lane 4) is not
significant enough to account for the large increase in
hormone-dependent phosphorylation. The two time points with
8-bromo-cAMP as well as no treatment showed a roughly equivalent degree
of phosphorylation when normalized for protein level. To ensure that
treatment with 8-bromo-cAMP resulted in ligand-independent activation
of cPRA, CV-1 cells were assayed for activation of the
GRE2e1bCAT reporter (using normal conditions for
infection of cells) at the same time that the labeling experiment was
performed. Although there is negligible receptor activation as measured
by CAT activity following 30 min of treatment with 8-bromo-cAMP (Fig.
6, lane 2), receptor activation following
10 h of drug treatment was comparable with receptor activation
following 10 h of steroid treatment (Fig. 6, compare lane
3 and lane 5).
Although treatment with 8-bromo-cAMP did not enhance the overall level
of cPRA phosphorylation, it was possible that site-specific phosphorylation in the receptor was altered by treatment with 8-bromo-cAMP. Following in vivo labeling of cPRA
in CV-1 cells with [32P]H3PO4 and
treatment of cells with 108 M progesterone,
separation of cPRA tryptic peptides on C-18 reverse phase
HPLC columns revealed four major 32P-labeled peaks (Fig.
7, panel A). The pattern for three of the peaks was consistent with previous studies that identified
Ser260 in the first peak, Ser211 and
Ser367 in the second peak, and Ser530 in the
third peak (as labeled in Fig. 7, panel A). The fourth broad
peak on the HPLC profile (between the Ser260 peak and the
Ser211/Ser367 peak) was not identified and
could represent incomplete tryptic digestion of peptides in the other
peaks. As expected, in the absence of hormone, only the
Ser260 and the Ser211/Ser367 peaks
were present on the HPLC profile (Fig. 7, panel B). Further analysis of the Ser211/Ser367 peak by alkaline
peptide gel electrophoresis revealed that only the peptide for
Ser211 was present in this peak (data not shown). Following
treatment with 2 mM 8-bromo-cAMP, two phosphopeptide peaks
were detected (Fig. 7, panel C) that are identical to the
peaks for Ser260 and Ser211/Ser367
detected under conditions of no treatment (compare Fig. 7, panels B and C). Analysis of the
Ser211/Ser367 peak by alkaline peptide gel
electrophoresis showed that the Ser211 peptide instead of
the Ser367 peptide was present in the peak (data not
shown). Because the preceding experiments necessitated the use of so
many cells, it was not feasible to do the different treatments and
subsequent phosphopeptide analysis concurrently. Hence the magnitude of
the peaks shown in different panels of Fig. 7 can only be compared qualitatively.
To ensure that ligand-independent activation of cPRA
occurred in this experiment, CV-1 cells were assayed for activation of the GRE2e1bCAT reporter following treatment with
2 mM 8-bromo-cAMP concurrently with the
[32P]H3PO4 labeling experiment.
As shown in Fig. 8, treatment with 8-bromo-cAMP resulted
in significant activation of cPRA with CAT activity
approximately 40% of that following progesterone treatment.
Taken together, these results show that ligand-independent activation of cPRA induced by treatment of CV-1 cells with 8-bromo-cAMP does not alter the overall level or sites of receptor phosphorylation relative to conditions of no treatment.
Based on our previous studies, phosphorylation of cPR in response to progesterone stimulation can be separated into two classes, the hormone-dependent phosphorylation at Ser530 and Ser367 which occurs only in the presence of hormone and the phosphorylation at Ser211 and Ser260 which occurs before hormone treatment and whose level is enhanced by progesterone treatment (8, 36). In the present study, our phosphopeptide mapping analyses showed that cPRA is only phosphorylated at Ser211 and Ser260 during the ligand-independent activation by 8-Br-cAMP. The differential phosphorylation of cPRA in hormone-dependent and ligand-independent activation suggests that the two activation processes use distinct mechanisms.
Our previous functional analyses have suggested that Ser530 phosphorylation in cPR enhances the response of the receptor to low levels of progesterone (15), whereas Ser211 phosphorylation is required for the receptor to achieve its full transcriptional potential at all hormone concentrations (16). In the present study, we show that the mutation of Ser260 to alanine also decreased the transcriptional activity of the receptor to a similar degree as the mutation of Ser211. Mutation of Ser367 to alanine did not affect the transcriptional activity of the receptor significantly in the experiments shown in Figs. 1 and 3. In some experiments, the activity of Ala367 is significantly higher than that of the wild type (data not shown), indicating that phosphorylation at Ser367 might down-regulate the activity of the receptor. The transcriptional activity of cPRA in response to 8-bromo-cAMP or dopamine agonist is not affected by mutation of either Ser530 or Ser367 to alanine. This is consistent with our phosphopeptide mapping results which show that these two sites are not phosphorylated following ligand-independent activation by 8-Br-cAMP. The activity of the AAAA mutant in response to either progesterone or 8-Br-cAMP is not decreased to the degree of some of the single site mutations. Whether this is due to the positive effect of mutation of Ser367 to alanine remains to be determined.
The transcriptional activity of the receptor in response to 8-Br-cAMP is reduced by mutation of either Ser211 or Ser260 to alanine to a similar degree as that in the response to progesterone. This suggests that receptor phosphorylation at either Ser211 or Ser260 regulates certain aspects of the activation process common to both ligand-dependent and ligand-independent pathways. This is also consistent with the observation that Ser211 and Ser260 are the two sites phosphorylated in receptors activated by either 8-bromo-cAMP or progesterone. The localization of Ser211 and Ser260 to the A/B region suggests that they both might be required for activation function 1 to achieve its maximal transcriptional potential.
Regarding the molecular mechanism of the ligand-independent activation of cPR, our data showed that mutation of either one or all of the phosphorylation sites to alanine did not abolish the ligand-independent activation by 8-bromo-cAMP. Mutation of all four sites did not result in alternate phosphorylation of cPRA. In addition, in concurrent experiments in which ligand-independent activation of the receptor is observed, 8-bromo-cAMP treatment failed to either increase the overall phosphorylation or cause alternate phosphorylation of the receptor. Based on these results, we conclude that phosphorylation of cPR is not absolutely required for ligand-independent activation of the receptor by 8-bromo-cAMP.
Dopamine receptor activation is known to activate multiple signal transduction pathways including the protein kinase A and the protein kinase C pathways (48-51). Thus, it is possible that ligand-independent activation of cPR in response to dopamine agonists is mediated through the same molecular mechanism that 8-bromo-cAMP uses to induce receptor activation. This is consistent with our finding that receptor phosphorylation is not absolutely required for ligand-independent activation by an agonist for the D1 type of membrane dopamine receptor.
Our conclusion that the ligand-independent activation of cPR in response to 8-bromo-cAMP or dopamine is not mediated by receptor phosphorylation appears contrary to a report which showed that the ligand-independent activation of human estrogen receptor in response to EGF is mediated through receptor phosphorylation at Ser118 (52). However, another group has shown that phosphorylation at the same site by mitogen-activated protein kinase in response to EGF enhanced the estrogen-stimulated activity of the receptor but did not cause the activation of the receptor in the absence of estrogen (53). It remains to be determined whether human estrogen receptor and cPR represent two types of receptors that are ligand-independently activated through different mechanisms. An alternative explanation is that 8-bromo-cAMP and EGF are using different mechanisms to induce ligand-independent activation of steroid receptors, with EGF using direct receptor phosphorylation as part of its mechanism to induce receptor activation while 8-bromo-cAMP does not.
Because receptor phosphorylation is not essential for the activation of cPR by 8-bromo-cAMP, it is possible that phosphorylation of receptor-associated proteins may mediate the ligand-independent activation. In this regard, the phosphorylation of two classes of proteins that interact with steroid receptors is worth further investigation. The first class of proteins is heat shock proteins. Many of the heat shock proteins are known to be phosphoproteins. Studies have shown that the homologous deletion of a heat shock protein, dnaJ, from yeast resulted in the activation of both estrogen and glucocorticoid receptors in the absence of their corresponding ligands (54). It must be pointed out that the mechanism of steroid receptor activation in yeast might be different from the mechanism in higher organisms. For example, the glucocorticoid receptor has not been shown to be ligand-independently activated in mammalian cells, and the activation of human androgen receptor by androgen in yeast was shown to be down-regulated by deletion of the same dnaJ protein (55). However, it is possible that the signal pathway activated by 8-bromo-cAMP in mammalian cells results in the phosphorylation of certain heat shock proteins. This phosphorylation could cause an alteration in either the expression level of the heat shock proteins or the affinity of the heat shock proteins toward steroid receptors, leading to receptor activation. Interestingly, the phosphorylation of a heat shock protein, GroEL, in bacteria has been shown to regulate its interaction with substrates (56).
In addition to heat shock proteins, other proteins that are known to interact with and to be involved in the activation process of steroid receptors are transcriptional coactivators such as TAFII30 (57), transcriptional intermediary factor-1 (58), CREB binding protein (59), and the recently identified steroid receptor coactivator including SRC-1 (60) and p140/p160 (61-63). So far, very little is known about the phosphorylation of these cofactors. It will be interesting to determine whether alterations in phosphorylation of coactivators is induced by 8-bromo-cAMP treatment and whether these alterations may contribute to ligand-independent activation of steroid receptors.
We are grateful to Ling Duan for technical assistance and Dr. Yixian Zhang for instruction on the phosphopeptide analysis.