Protein Phosphatase Types 1 and/or 2A Regulate Nucleocytoplasmic Shuttling of Glucocorticoid Receptors
Donald B. DeFranco, Ming Qi, Kristina C. Borror, Michael J. Garabedian, and David L. Brautigan Department of Biological Sciences University of Pittsburgh (D.B.D., M.Q., K.C.B.) Pittsburgh, Pennsylvania 15260 Department of Biochemistry and Biophysics University of California (M.J.G.) San Francisco, California 94143 Section of Biochemistry Brown University (D.L.B.) Providence, Rhode Island 02912
We have used okadaic acid (OA), a cell-permeable inhibitor of serine/threonine protein phosphatase types 1 (PP-1) and 2A (PP-2A), to demonstrate that the subcellular distribution of glucocorticoid receptor (GR) in rat fibroblasts is influenced by its phosphorylation state. Nuclear GRs in OA-treated cells retain transcriptional enhancement activity. Nuclear import or export of hormone agonist-bound GRs is not affected by OA. However, a dose of OA that fully inhibits PP-2A and partially inhibits PP-1, but not a lower dose that only partially inhibits PP-2A, leads to inefficient nuclear retention of agonist-bound GRs, and their redistribution into the cytoplasm. These receptors appear to be trapped in the cytoplasmic compartment and are unable to recycle (i.e. reenter the nucleus). Addition of OA during different steps of GR recycling demonstrates that OA must be present during nuclear export of GRs to block GR recycling. A direct role for PP-1 and/or PP-2A in GR recycling is suggested by site-specific hyperphosphorylation of GRs in vivo during OA inhibition of recycling. These are the same sites that undergo in vitro site-specific dephosphorylation by PP-1 and PP-2A. The block in GR recycling that results from inhibition of PP-1 and/or PP-2A resembles effects previously observed in v-mos-transformed rat fibroblasts. Interestingly, OA inhibition of PP-2A in v-mostransformed cells leads to the reversal of oncoprotein effects on GR recycling and retention of receptors within the nuclear compartment. We propose that GR recycling is influenced by the activities of distinct protein phosphatases (PP-1 and/or PP-2A), and that the interference of this pathway observed
in v-mos-transformed cells may be the result of effects of the oncoprotein on the phosphatases or a specific subset of their targets. (Molecular Endocrinology 5: 1215-1228, 1991)
INTRODUCTION The modulation of gene activity in response to steroid hormones is mediated through the action of soluble intracellular receptor proteins (1). Unoccupied glucocorticoid receptor (GR) proteins reside predominantly within the cytoplasmic compartment, but after ligand binding, there is rapid and efficient translocation of receptors to the nucleus (2). It has been postulated that nuclear GRs are not necessarily confined to that compartment, but are capable of recycling between nuclear and cytoplasmic compartments (3,4). Although various approaches have been used to demonstrate the reversible shuttling of proteins between nuclear and cytoplasmic compartments (5), nuclear export of steroid receptors remains poorly characterized. In cells that have been briefly exposed to glucocorticoids and then subjected to an extended period of hormone withdrawal, hormone-binding activity is recovered in cytosolic fractions in correspondence with a decay in nuclear hormone-binding activity (6, 7). The redistribution of GRs from nuclear to cytoplasmic compartments has also been observed in v-mos-transformed cells chronically exposed to hormone (8). In this case, the regenerated cytoplasmic receptors cannot be reused and may represent trapped recycling intermediates (8). Thus, although results from various systems provide support for nucleocytoplasmic shuttling of GRs (3, 4, 7, 8), the mechanisms involved in the nuclear import
0888-8809/91 /1215-1228$03.00/0 Molecular Endocrinology Copyright© 1991 by The Endocrine Society
1215
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1216
and export of these signal transduction proteins remain enigmatic. Since steroid receptor proteins are known to be phosphoproteins, numerous studies have been directed toward uncovering a physiological role for this modification (9). GRs are phosphorylated predominantly on serine residues (10, 11) and become hyperphosphorylated in vivo in response to hormone agonist treatment (12,13). The effects of hormone-induced phosphorylation on GR function have not been clearly established. Hormone-induced phosphorylation of GR may be reversible, as Orti and co-workers (14) have postulated that dephosphorylation of GR accompanies its recycling. However, the protein kinases and phosphatases that act on steroid receptor proteins in vivo have not been identified, nor has it been established how their actions on GR are influenced by hormone or other physiological signals. Protein phosphorylation regulates enzymatic activities involved in various cellular processes ranging from glycogen metabolism (15-17) to cell cycle (18, 19). Ultimately, it is the balance between protein kinase and phosphatase activities that determines the phosphorylation state of any given protein. We have set out to examine the role of phosphorylation on GR function in vivo through the use of okadaic acid (OA), a specific protein phosphatase type 1 (PP-1) and 2A (PP-2A) inhibitor (20-23). In previous studies, a temperaturesensitive v-mos-transformed rat fibroblast cell line (6m2 cells) (24) was used to reveal v-mos effects on GR recycling (8). An 85-kDa gag-mos fusion protein, p85ga9"mos, which is a serine/threonine protein kinase, is expressed in 6m2 cells from a spliced mRNA species at temperatures permissive for transformation (28-33 C) (25, 26). At the nonpermissive temperature (39 C), the 4.0-kilobase genomic viral transcript is not efficiently spliced, leaving an internal translational stop codon in frame and thereby generating a 58-kDa gag protein that lacks transformation activity (25, 26). We report here that as observed in v-mos-transformed rat NRK fibroblasts (8), GR recycling is impaired in nontransformed cells upon OA inhibition of PP-1 and PP-2A. A direct role for PP-1 and PP-2A in GR recycling is suggested by site-specific hyperphosphorylation of GRs in OA-treated cells that coincides with its inability to recycle and by site-specific dephosphorylation of GR by PP-1 and PP-2A in vitro. Finally, the analysis of OA effects on GR recycling in v-mos-transformed cells suggests an involvement of PP-1 and PP-2A in v-mosmediated transformation.
RESULTS OA Blocks Hormone Agonist-Dependent Recycling of GR in Nontransformed 6m2 Rat Fibroblasts We examined the effects of OA, a potent PP-1 and PP2A inhibitor (20), on the subcellular distribution of GR. The specific activity of protein phosphatase (measured
Vol 5 No. 9
with phosphorylase as substrate) was the same in extracts of nontransformed and p85gag"mos-transformed 6m2 cells. As expected, the PP-2A activity (resistant to 12, the PP-1-specific protein inhibitor) (16) was highly sensitive to OA, being partially inhibited at 10 nM OA. At 25 and 100 nM OA, the PP-2A activity was fully inhibited, and there was increasing, but partial, inhibition of PP-1 (one quarter to one third PP-1 remaining at 100 nM). In our experiments, 25 nM OA was used to fully inhibit PP-2A, with minimal effect on PP-1, and 100 nM OA was used to fully inhibit PP-2A and inhibit more than half the PP-1. Higher doses of OA, to more fully inhibit PP-1, proved to be toxic to 6m2 cells. Nontransformed 6m2 cells were treated with OA, and the subcellular distribution of GR was visualized by indirect immunofluorescence analysis. In the absence of hormone, GRs are predominantly localized in the cytoplasm of nontransformed 6m2 cells (8) (Fig. 1) and remain within that compartment during 12- to 16-h treatments with 25 nM and 100 nM OA (Fig. 1). Within 1 h after the addition of 1 ^ M dexamethasone to nontransformed 6m2 cells, the majority of GRs translocate to the nucleus in either the presence (Fig. 1) or absence (8) of 25 and 100 nM OA. Thus, hormone-dependent nuclear import of GRs in vivo is not altered upon OA inhibition of PP-2A activity. GRs appear in nuclei of nontransformed cells for up to 12-16 h in the continued presence of dexamethasone and 25 nM OA (Fig. 1). However, treatment of nontransformed 6m2 cells with dexamethasone and a higher dose of 100 nM OA leads to redistribution of GRs from nuclei to the cytoplasmic compartment within 12 h (Fig. 1). Identical results were obtained when cycloheximide was added in conjunction with dexamethasone and OA (data not shown), indicating that GRs which accumulate in the cytoplasm after prolonged dexamethasone and OA treatment derive from preexistent nuclear forms. The redistribution of GRs from nuclear to cytoplasmic compartment demonstrates that nuclear export of GRs occurs despite inhibition of protein phosphatases PP-1 and PP-2A. The exclusion of GRs from the nucleus of cells treated with dexamethasone and 100 nM OA could result from abnormal export of normally static nuclear receptors or a block in the reentry of receptors into the nucleus that are actively recycling between nuclear and cytoplasmic compartments. Previously, we had used the glucocorticoid antagonist RU486 to demonstrate that displacement of nuclear GRs in p85gagmos-transformed 6m2 cells most likely results from an inhibition of GR recycling (27). GRs that translocate to the nucleus of p85gag"mos-transformed 6m2 cells in response to RU486 treatment do not redistribute to the cytoplasm and are retained in the nucleus (27). Similarly, GRs that translocate to the nucleus of nontransformed 6m2 cells in response to a combined treatment with RU486 and 100 nM OA are not depleted from that compartment (Fig. 1). Thus, both OA and \/-mos oncoprotein appear to exert analogous effects in blocking the reutilization (i.e. nuclear reentry) of GRs. The results of our experiments permit two distinct modes of nuclear transport
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
PP-1, PP-2A, and GR Recycling
1217
Time(hrs) treatment
lOOnMOA
Dex/25nMOA
Dex/100nMOA
RU486/100nMOA
Fig. 1. Subcellular Distribution of GR in OA-Treated Nontransformed 6m2 Cells: Indirect Immunofluorescence Staining Nontransformed 6m2 cells were treated with 100 nM OA (top panels), 100 nM OA and 1 ^M dexamethasone (Dex; middle panels), or 100 nM OA and 0.1 MM RU486 (bottom panels) for the lengths of time indicated. Cells were fixed and processed for anti-GR immunofluorescence as described in Materials and Methods. GRs are localized in the nucleus of cells treated for 1 h with dexamethasone and OA, and treated for 1 to 12 h with RU486 and OA. GRs are only detected in the cytoplasm after a 12-h treatment with dexamethasone and OA.
to be distinguished, which differ in their sensitivity to OA. Nuclear import of GRs is apparently insensitive to OA, while GRs are unable to reenter the nucleus in the presence of 100 nM OA. Inhibition of GR Recycling by OA Is Exerted during Nuclear Export of GRs Hormone binding assays in conjunction with hormone withdrawal regimens have been used to demonstrate the reversibility of hormone-dependent nuclear translocation of GRs (6, 7). We have used indirect immunofluorescence to visualize the subcellular redistribution of GRs upon hormone withdrawal (Fig. 2). Corticosteronedependent translocation of GRs into the nucleus of nontransformed 6m2 cells is reversed upon hormone withdrawal, as GRs redistribute to the cytoplasmic compartment (Fig. 2). Since cycloheximide was added to cells during the hormone withdrawal period to inhibit de novo synthesis of GRs (8), receptors that appear in the cytoplasm after hormone withdrawal derive from preexistent nuclear GRs. Recycled cytoplasmic GRs reenter the nucleus upon readdition of hormone to withdrawn cells (Fig. 2). However, in the continuous presence of 100 nM OA, GRs that redistribute to the cytoplasmic compartment upon hormone withdrawal
are unable to reenter the nucleus upon readdition of hormone (Fig. 2). In vivo hormone binding assays demonstrated that OA did not significantly affect the hormone-binding activity of cytoplasmic GRs in hormonewithdrawn cells (Table 1). Thus, blocked reentry of GRs into the nucleus cannot be explained by the effects of OA on the ability of GR to bind hormone. The reversible shuttling of GRs between nuclear and cytoplasmic compartments provided a useful paradigm to assess OA effects on three distinct GR translocation steps: 1) import into the nucleus, 2) export from the nucleus, and 3) reentry into the nucleus. With addition of OA to nontransformed 6m2 cells only during the initial hormone treatment, effects on import of GRs to the nucleus from the cytoplasm could be monitored. Addition of OA during either the hormone withdrawal period or second hormone treatment that followed withdrawal allowed the monitoring of OA effects on GR export from or reentry into the nucleus, respectively. As shown by the indirect immunofluorescence micrographs in Fig. 3, GRs that exit the nucleus upon hormone withdrawal are still capable of reentering the nucleus if 100 nM OA is present during either the initial or secondary hormone treatments (Fig. 3, top and bottom panels, respectively). In contrast, if nontransformed 6m2 cells are treated with 100 nM OA during
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
Vol 5 No. 9
MOL ENDO-1991 1218
treatment
- cort.
cort.l hr
cort.l hr withdrawal o/n
cort.l hr withdrawal o/n readd cort.l hr lOOnM OA
-
+
Fig. 2. Subcellular Distribution of GR in OA-Treated Nontransformed 6m2 Cells: Indirect Immunofluorescence Staining Where indicated, nontransformed 6m2 cells were treated with 1 HM corticosterone (cort.) for 1 h, then subjected to an overnight (o/n) withdrawal period in hormone-free medium containing 10 ^g/ml cycloheximide. In one set of experiments (bottom panels), 1 fiM corticosterone was readded to withdrawn cultures for 1 h. OA (100 nM) was either absent (left) or present (right) throughout the duration of the experiment. Cells were fixed and processed for anti-GR immunofluorescence as described in Materials and Methods. GRs are localized in the nucleus of cells treated for 1 h with corticosterone and OA, and redistribute to the cytoplasmic compartment after an overnight withdrawal period. These replenished cytoplasmic GRs are only able to reenter the nucleus after readdition of corticosterone in cells that have not been exposed to OA (bottom left). They remain cytoplasmic in OA-treated cells (bottom right).
the hormone withdrawal period, GRs do not appear to reenter the nucleus in response to a second hormone treatment (Fig. 3, middle panel). GRs that remain in the cytoplasm under these conditions can still bind hormone effectively, as revealed by in vivo hormone binding assays (Table 1). The fact that a 12- to 16-h withdrawal period is required to displace GRs from the nucleus necessitates a prolonged exposure of cells to OA. GRs can still translocate to the nucleus in response to hormone treatment if nontransformed 6m2 cells were pretreated with 100 nM OA for 12 h before the initial hormonal stimulus (data not shown). Thus, the block of GR recycling that ensues upon exposure of cells to OA during export of receptors from the nucleus is not simply the result of spurious effects of prolonged OA
treatment. It seems likely that reentry of GRs into the nucleus requires a dephosphorylation event(s) that is catalyzed by PP-2A alone or in conjunction with PP-1. This probably occurs while GRs are being exported from the nucleus. Site-Specific Hyperphosphorylation of GR in Vivo upon OA Treatment OA leads to hyperphosphorylation of numerous proteins, some of which may be direct substrates of PP-1 and PP-2A (23, 28). Hormone-induced phosphorylation of GR had previously been observed in mouse WEHI-7 T-lymphoma (12) and NIH 3T3 cell lines (13). In addition, GR phosphorylation and dephosphorylation have been proposed to be critical events for GR recycling (14). If
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
1219
PP-1, PP-2A, and GR Recycling
Table 1. Whole Cell Binding of [3H]Dexamethasone to Nontransformed 6m2 Cells after Hormone Withdrawal in the Presence or Absence of OA T oatmont i reaimeni
Specific Binding ( C p m / m g T o t a | p ro tein)
Average Specific Binding (cpm/mg Total Protein)
-OA
Exp 1:37,200 Exp 2: 23,400 Exp 3: 24,000
28,200 ± 6,400
Exp 1: 35,500 Exp 2: 24,000 Exp 3: 21,700
27,100 ±6,000
+OA
Whole cell binding was performed using [3H]dexamethasone, and specific binding was determined as described previously (8). Nontransformed 6m2 cells (three separate experiments) were treated with 1 ^ M corticosterone for 1 h and then withdrawn from hormone for 16 h in the presence or absence of 100 niwi OA, as described in Materials and Methods. Specific binding is expressed per mg total protein. Nonspecific binding (typically
Donald B. DeFranco, Ming Qi, Kristina C. Borror, Michael J. Garabedian, and David L. Brautigan Department of Biological Sciences University of Pittsburgh (D.B.D., M.Q., K.C.B.) Pittsburgh, Pennsylvania 15260 Department of Biochemistry and Biophysics University of California (M.J.G.) San Francisco, California 94143 Section of Biochemistry Brown University (D.L.B.) Providence, Rhode Island 02912
We have used okadaic acid (OA), a cell-permeable inhibitor of serine/threonine protein phosphatase types 1 (PP-1) and 2A (PP-2A), to demonstrate that the subcellular distribution of glucocorticoid receptor (GR) in rat fibroblasts is influenced by its phosphorylation state. Nuclear GRs in OA-treated cells retain transcriptional enhancement activity. Nuclear import or export of hormone agonist-bound GRs is not affected by OA. However, a dose of OA that fully inhibits PP-2A and partially inhibits PP-1, but not a lower dose that only partially inhibits PP-2A, leads to inefficient nuclear retention of agonist-bound GRs, and their redistribution into the cytoplasm. These receptors appear to be trapped in the cytoplasmic compartment and are unable to recycle (i.e. reenter the nucleus). Addition of OA during different steps of GR recycling demonstrates that OA must be present during nuclear export of GRs to block GR recycling. A direct role for PP-1 and/or PP-2A in GR recycling is suggested by site-specific hyperphosphorylation of GRs in vivo during OA inhibition of recycling. These are the same sites that undergo in vitro site-specific dephosphorylation by PP-1 and PP-2A. The block in GR recycling that results from inhibition of PP-1 and/or PP-2A resembles effects previously observed in v-mos-transformed rat fibroblasts. Interestingly, OA inhibition of PP-2A in v-mostransformed cells leads to the reversal of oncoprotein effects on GR recycling and retention of receptors within the nuclear compartment. We propose that GR recycling is influenced by the activities of distinct protein phosphatases (PP-1 and/or PP-2A), and that the interference of this pathway observed
in v-mos-transformed cells may be the result of effects of the oncoprotein on the phosphatases or a specific subset of their targets. (Molecular Endocrinology 5: 1215-1228, 1991)
INTRODUCTION The modulation of gene activity in response to steroid hormones is mediated through the action of soluble intracellular receptor proteins (1). Unoccupied glucocorticoid receptor (GR) proteins reside predominantly within the cytoplasmic compartment, but after ligand binding, there is rapid and efficient translocation of receptors to the nucleus (2). It has been postulated that nuclear GRs are not necessarily confined to that compartment, but are capable of recycling between nuclear and cytoplasmic compartments (3,4). Although various approaches have been used to demonstrate the reversible shuttling of proteins between nuclear and cytoplasmic compartments (5), nuclear export of steroid receptors remains poorly characterized. In cells that have been briefly exposed to glucocorticoids and then subjected to an extended period of hormone withdrawal, hormone-binding activity is recovered in cytosolic fractions in correspondence with a decay in nuclear hormone-binding activity (6, 7). The redistribution of GRs from nuclear to cytoplasmic compartments has also been observed in v-mos-transformed cells chronically exposed to hormone (8). In this case, the regenerated cytoplasmic receptors cannot be reused and may represent trapped recycling intermediates (8). Thus, although results from various systems provide support for nucleocytoplasmic shuttling of GRs (3, 4, 7, 8), the mechanisms involved in the nuclear import
0888-8809/91 /1215-1228$03.00/0 Molecular Endocrinology Copyright© 1991 by The Endocrine Society
1215
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1216
and export of these signal transduction proteins remain enigmatic. Since steroid receptor proteins are known to be phosphoproteins, numerous studies have been directed toward uncovering a physiological role for this modification (9). GRs are phosphorylated predominantly on serine residues (10, 11) and become hyperphosphorylated in vivo in response to hormone agonist treatment (12,13). The effects of hormone-induced phosphorylation on GR function have not been clearly established. Hormone-induced phosphorylation of GR may be reversible, as Orti and co-workers (14) have postulated that dephosphorylation of GR accompanies its recycling. However, the protein kinases and phosphatases that act on steroid receptor proteins in vivo have not been identified, nor has it been established how their actions on GR are influenced by hormone or other physiological signals. Protein phosphorylation regulates enzymatic activities involved in various cellular processes ranging from glycogen metabolism (15-17) to cell cycle (18, 19). Ultimately, it is the balance between protein kinase and phosphatase activities that determines the phosphorylation state of any given protein. We have set out to examine the role of phosphorylation on GR function in vivo through the use of okadaic acid (OA), a specific protein phosphatase type 1 (PP-1) and 2A (PP-2A) inhibitor (20-23). In previous studies, a temperaturesensitive v-mos-transformed rat fibroblast cell line (6m2 cells) (24) was used to reveal v-mos effects on GR recycling (8). An 85-kDa gag-mos fusion protein, p85ga9"mos, which is a serine/threonine protein kinase, is expressed in 6m2 cells from a spliced mRNA species at temperatures permissive for transformation (28-33 C) (25, 26). At the nonpermissive temperature (39 C), the 4.0-kilobase genomic viral transcript is not efficiently spliced, leaving an internal translational stop codon in frame and thereby generating a 58-kDa gag protein that lacks transformation activity (25, 26). We report here that as observed in v-mos-transformed rat NRK fibroblasts (8), GR recycling is impaired in nontransformed cells upon OA inhibition of PP-1 and PP-2A. A direct role for PP-1 and PP-2A in GR recycling is suggested by site-specific hyperphosphorylation of GRs in OA-treated cells that coincides with its inability to recycle and by site-specific dephosphorylation of GR by PP-1 and PP-2A in vitro. Finally, the analysis of OA effects on GR recycling in v-mos-transformed cells suggests an involvement of PP-1 and PP-2A in v-mosmediated transformation.
RESULTS OA Blocks Hormone Agonist-Dependent Recycling of GR in Nontransformed 6m2 Rat Fibroblasts We examined the effects of OA, a potent PP-1 and PP2A inhibitor (20), on the subcellular distribution of GR. The specific activity of protein phosphatase (measured
Vol 5 No. 9
with phosphorylase as substrate) was the same in extracts of nontransformed and p85gag"mos-transformed 6m2 cells. As expected, the PP-2A activity (resistant to 12, the PP-1-specific protein inhibitor) (16) was highly sensitive to OA, being partially inhibited at 10 nM OA. At 25 and 100 nM OA, the PP-2A activity was fully inhibited, and there was increasing, but partial, inhibition of PP-1 (one quarter to one third PP-1 remaining at 100 nM). In our experiments, 25 nM OA was used to fully inhibit PP-2A, with minimal effect on PP-1, and 100 nM OA was used to fully inhibit PP-2A and inhibit more than half the PP-1. Higher doses of OA, to more fully inhibit PP-1, proved to be toxic to 6m2 cells. Nontransformed 6m2 cells were treated with OA, and the subcellular distribution of GR was visualized by indirect immunofluorescence analysis. In the absence of hormone, GRs are predominantly localized in the cytoplasm of nontransformed 6m2 cells (8) (Fig. 1) and remain within that compartment during 12- to 16-h treatments with 25 nM and 100 nM OA (Fig. 1). Within 1 h after the addition of 1 ^ M dexamethasone to nontransformed 6m2 cells, the majority of GRs translocate to the nucleus in either the presence (Fig. 1) or absence (8) of 25 and 100 nM OA. Thus, hormone-dependent nuclear import of GRs in vivo is not altered upon OA inhibition of PP-2A activity. GRs appear in nuclei of nontransformed cells for up to 12-16 h in the continued presence of dexamethasone and 25 nM OA (Fig. 1). However, treatment of nontransformed 6m2 cells with dexamethasone and a higher dose of 100 nM OA leads to redistribution of GRs from nuclei to the cytoplasmic compartment within 12 h (Fig. 1). Identical results were obtained when cycloheximide was added in conjunction with dexamethasone and OA (data not shown), indicating that GRs which accumulate in the cytoplasm after prolonged dexamethasone and OA treatment derive from preexistent nuclear forms. The redistribution of GRs from nuclear to cytoplasmic compartment demonstrates that nuclear export of GRs occurs despite inhibition of protein phosphatases PP-1 and PP-2A. The exclusion of GRs from the nucleus of cells treated with dexamethasone and 100 nM OA could result from abnormal export of normally static nuclear receptors or a block in the reentry of receptors into the nucleus that are actively recycling between nuclear and cytoplasmic compartments. Previously, we had used the glucocorticoid antagonist RU486 to demonstrate that displacement of nuclear GRs in p85gagmos-transformed 6m2 cells most likely results from an inhibition of GR recycling (27). GRs that translocate to the nucleus of p85gag"mos-transformed 6m2 cells in response to RU486 treatment do not redistribute to the cytoplasm and are retained in the nucleus (27). Similarly, GRs that translocate to the nucleus of nontransformed 6m2 cells in response to a combined treatment with RU486 and 100 nM OA are not depleted from that compartment (Fig. 1). Thus, both OA and \/-mos oncoprotein appear to exert analogous effects in blocking the reutilization (i.e. nuclear reentry) of GRs. The results of our experiments permit two distinct modes of nuclear transport
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
PP-1, PP-2A, and GR Recycling
1217
Time(hrs) treatment
lOOnMOA
Dex/25nMOA
Dex/100nMOA
RU486/100nMOA
Fig. 1. Subcellular Distribution of GR in OA-Treated Nontransformed 6m2 Cells: Indirect Immunofluorescence Staining Nontransformed 6m2 cells were treated with 100 nM OA (top panels), 100 nM OA and 1 ^M dexamethasone (Dex; middle panels), or 100 nM OA and 0.1 MM RU486 (bottom panels) for the lengths of time indicated. Cells were fixed and processed for anti-GR immunofluorescence as described in Materials and Methods. GRs are localized in the nucleus of cells treated for 1 h with dexamethasone and OA, and treated for 1 to 12 h with RU486 and OA. GRs are only detected in the cytoplasm after a 12-h treatment with dexamethasone and OA.
to be distinguished, which differ in their sensitivity to OA. Nuclear import of GRs is apparently insensitive to OA, while GRs are unable to reenter the nucleus in the presence of 100 nM OA. Inhibition of GR Recycling by OA Is Exerted during Nuclear Export of GRs Hormone binding assays in conjunction with hormone withdrawal regimens have been used to demonstrate the reversibility of hormone-dependent nuclear translocation of GRs (6, 7). We have used indirect immunofluorescence to visualize the subcellular redistribution of GRs upon hormone withdrawal (Fig. 2). Corticosteronedependent translocation of GRs into the nucleus of nontransformed 6m2 cells is reversed upon hormone withdrawal, as GRs redistribute to the cytoplasmic compartment (Fig. 2). Since cycloheximide was added to cells during the hormone withdrawal period to inhibit de novo synthesis of GRs (8), receptors that appear in the cytoplasm after hormone withdrawal derive from preexistent nuclear GRs. Recycled cytoplasmic GRs reenter the nucleus upon readdition of hormone to withdrawn cells (Fig. 2). However, in the continuous presence of 100 nM OA, GRs that redistribute to the cytoplasmic compartment upon hormone withdrawal
are unable to reenter the nucleus upon readdition of hormone (Fig. 2). In vivo hormone binding assays demonstrated that OA did not significantly affect the hormone-binding activity of cytoplasmic GRs in hormonewithdrawn cells (Table 1). Thus, blocked reentry of GRs into the nucleus cannot be explained by the effects of OA on the ability of GR to bind hormone. The reversible shuttling of GRs between nuclear and cytoplasmic compartments provided a useful paradigm to assess OA effects on three distinct GR translocation steps: 1) import into the nucleus, 2) export from the nucleus, and 3) reentry into the nucleus. With addition of OA to nontransformed 6m2 cells only during the initial hormone treatment, effects on import of GRs to the nucleus from the cytoplasm could be monitored. Addition of OA during either the hormone withdrawal period or second hormone treatment that followed withdrawal allowed the monitoring of OA effects on GR export from or reentry into the nucleus, respectively. As shown by the indirect immunofluorescence micrographs in Fig. 3, GRs that exit the nucleus upon hormone withdrawal are still capable of reentering the nucleus if 100 nM OA is present during either the initial or secondary hormone treatments (Fig. 3, top and bottom panels, respectively). In contrast, if nontransformed 6m2 cells are treated with 100 nM OA during
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
Vol 5 No. 9
MOL ENDO-1991 1218
treatment
- cort.
cort.l hr
cort.l hr withdrawal o/n
cort.l hr withdrawal o/n readd cort.l hr lOOnM OA
-
+
Fig. 2. Subcellular Distribution of GR in OA-Treated Nontransformed 6m2 Cells: Indirect Immunofluorescence Staining Where indicated, nontransformed 6m2 cells were treated with 1 HM corticosterone (cort.) for 1 h, then subjected to an overnight (o/n) withdrawal period in hormone-free medium containing 10 ^g/ml cycloheximide. In one set of experiments (bottom panels), 1 fiM corticosterone was readded to withdrawn cultures for 1 h. OA (100 nM) was either absent (left) or present (right) throughout the duration of the experiment. Cells were fixed and processed for anti-GR immunofluorescence as described in Materials and Methods. GRs are localized in the nucleus of cells treated for 1 h with corticosterone and OA, and redistribute to the cytoplasmic compartment after an overnight withdrawal period. These replenished cytoplasmic GRs are only able to reenter the nucleus after readdition of corticosterone in cells that have not been exposed to OA (bottom left). They remain cytoplasmic in OA-treated cells (bottom right).
the hormone withdrawal period, GRs do not appear to reenter the nucleus in response to a second hormone treatment (Fig. 3, middle panel). GRs that remain in the cytoplasm under these conditions can still bind hormone effectively, as revealed by in vivo hormone binding assays (Table 1). The fact that a 12- to 16-h withdrawal period is required to displace GRs from the nucleus necessitates a prolonged exposure of cells to OA. GRs can still translocate to the nucleus in response to hormone treatment if nontransformed 6m2 cells were pretreated with 100 nM OA for 12 h before the initial hormonal stimulus (data not shown). Thus, the block of GR recycling that ensues upon exposure of cells to OA during export of receptors from the nucleus is not simply the result of spurious effects of prolonged OA
treatment. It seems likely that reentry of GRs into the nucleus requires a dephosphorylation event(s) that is catalyzed by PP-2A alone or in conjunction with PP-1. This probably occurs while GRs are being exported from the nucleus. Site-Specific Hyperphosphorylation of GR in Vivo upon OA Treatment OA leads to hyperphosphorylation of numerous proteins, some of which may be direct substrates of PP-1 and PP-2A (23, 28). Hormone-induced phosphorylation of GR had previously been observed in mouse WEHI-7 T-lymphoma (12) and NIH 3T3 cell lines (13). In addition, GR phosphorylation and dephosphorylation have been proposed to be critical events for GR recycling (14). If
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 31 October 2015. at 04:47 For personal use only. No other uses without permission. . All rights reserved.
1219
PP-1, PP-2A, and GR Recycling
Table 1. Whole Cell Binding of [3H]Dexamethasone to Nontransformed 6m2 Cells after Hormone Withdrawal in the Presence or Absence of OA T oatmont i reaimeni
Specific Binding ( C p m / m g T o t a | p ro tein)
Average Specific Binding (cpm/mg Total Protein)
-OA
Exp 1:37,200 Exp 2: 23,400 Exp 3: 24,000
28,200 ± 6,400
Exp 1: 35,500 Exp 2: 24,000 Exp 3: 21,700
27,100 ±6,000
+OA
Whole cell binding was performed using [3H]dexamethasone, and specific binding was determined as described previously (8). Nontransformed 6m2 cells (three separate experiments) were treated with 1 ^ M corticosterone for 1 h and then withdrawn from hormone for 16 h in the presence or absence of 100 niwi OA, as described in Materials and Methods. Specific binding is expressed per mg total protein. Nonspecific binding (typically
