Protein Phosphatase 2A Regulates Estrogen Receptor α (ER) Expression through Modulation of ER mRNA Stability*

Protein phosphatase 2A (PP2A) is a ubiquitously expressed member of the serine-threonine phosphatase family that is involved in regulation of many cellular processes including transcription, translation, cellular metabolism, and apoptosis. Because of a correlation between PP2A and estrogen receptor α (ER) expression in several human breast cancer cell lines, the effect of PP2A on regulation of ER expression in the human breast cancer cell line MCF-7 was studied. Inhibition of PP2A using the pharmacologic inhibitor okadaic acid at 250 nm for 16 h resulted in a 60% reduction in PP2A activity in MCF-7 cells concurrent with a 75% reduction in ER mRNA and protein expression. Similar results were obtained with a small interfering RNA probe that specifically inhibited PP2A expression. ER promoter studies showed that regulation of ER through the PP2A pathway did not occur through transcriptional activation. Rather, PP2A mediated ER expression through modulation of ER mRNA stability through degradation of ER mRNA, reversible with concomitant treatment with the proteasomal inhibitor MG 132. These data suggest a novel pathway controlling ER expression resulting from the activation of PP2A, potentially providing a novel therapeutic target.

Protein phosphatase 2A (PP2A) 1 is a ubiquitously expressed member of a large protein phosphatase family involved in the regulation of cell proliferation, cell differentiation, RNA transcription, DNA repair, and apoptosis (1)(2)(3)(4). PP2A has been shown to inhibit major signal transduction pathways, including the phosphatidylinositol 3-kinase/AKT and mitogen-activated protein kinases pathways. Decreased PP2A activity either through pharmacologic inhibition or RNA interference techniques results in increased apoptosis and inhibition of gene expression, suggesting an important role for PP2A in the regulation of cell growth (1,(5)(6)(7)(8)(9)(10)(11). PP2A also contributes to the control of epigenetic gene regulation by decreasing histone protein phosphorylation and increasing histone acetylation (12)(13)(14).
PP2A is a holoenzyme comprised of three subunits, a structural subunit (A), a regulatory subunit (B), and a catalytic subunit (C). A and C subunits are highly conserved, each consisting of two highly homologous isoforms, ␣ and ␤, whereas the B subunit is much more complex, containing three distinct families, B, BЈ, and BЉ, each with several homologous members (3). Considering all the potential interactions between the multitudes of subunits, 75 different combinations of A-B-C subunits could form, providing the potential for immense diversity and target specificity within the cell.
Dysregulation of PP2A has been shown to be a contributing factor in many cancer types. Mutations in the A subunit, resulting in decreased interaction with the B and C subunits and improper formation of the holoenzyme, have been noted in a subset of colorectal and breast cancers (15,16). In addition, overexpression of PP2A alone resulted in cellular transformation in human embryonic kidney cells (17). Aberrant subcellular localization of PP2A has been shown to alter cellular growth and apoptosis in hepatic cancer cells (18). These studies suggest that changes in PP2A can contribute directly to cancer development, increased cellular growth, and decreased apoptosis (17,18). Decreased nuclear localization of PP2A is correlated with decreased PP2A activity in estrogen receptor ␣ (ER)-negative compared with ER-positive human breast cancer cell lines, suggesting that PP2A could be a contributing factor to the hormone-independent phenotype that comprises a large subset of all human breast cancer cases (19). The mechanism of PP2A-mediated ER expression is still largely unknown.
Approximately 70 -80% of all breast tumors express ER protein. These tumors tend to grow more slowly and are associated with a slightly better prognosis (20). More importantly, the detection of ER in breast carcinoma cells is an important indicator of potential response to endocrine therapy as tumors expressing ER protein are the most likely to respond to endocrine therapy, whereas those lacking ER seldom respond (21). The molecular mechanisms controlling these effects are still being defined. Determining the molecular mechanisms underlying ER expression, therefore, is crucial to the understanding and treatment of breast cancer.
ER expression is controlled at the transcriptional level through both epigenetic mechanisms and activation of the proximal (P1) ER promoter (22)(23)(24)(25). ER expression is also regulated at the post-transcriptional level through modulation of mRNA stability via interactions with the 3Ј-UTR (26 -28). Previous data from our laboratory suggest that the B␤ subunit of PP2A is re-expressed coordinately with ER when ER-negative cells are treated with epigenetic modulators (29). These data suggest that PP2A expression is upstream of ER expression and may play an important role in the transcriptional control of ER (29). , and Treatment-MDA-MB-231, MDA-MB-468, T47D, and MCF-7 cells were cultured at 37°C, 5% CO 2 in  Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine  serum (Hyclone, Logan, UT), and 2 M L-alanyl-L-glutamine (Mediatech, Herndon, VA). For MDA-MB-231 ERϩ (231ERϩ) cells, the entire coding region of ER was PCR amplified using MCF-7 cDNA as template and cloned into the expression vector pIRESHyg2 (Clontech, Palo Alto, CA). The cloned cDNA was sequenced to verify the ER cDNA sequence. The construct was then transfected into MDA-MB-231 cells using Lipofectamine according to the manufacturer's protocol (Invitrogen). Stable colonies were selected using Hygromycin B (Calbiochem, La Jolla, CA) at a concentration of 800 g/ml in Dulbecco's modified Eagle's medium medium. Stable clones (termed 231ERϩ) were maintained in selection medium to ensure high levels of ER expression. Unless otherwise indicated, MDA-MB-231 and MCF-7 cells were plated at 300,000 cells/ 10-cm plate and 500,000 cells/10-cm plate, respectively, 24 h before treatment with okadaic acid (250 nM, OA) (Calbiochem) for 16 h.

Cell Culture, Maintenance
Reverse Transcription, PCR, and Real-time PCR-RNA was harvested using the TRIzol reagent (Invitrogen) as previously described (30). cDNA was synthesized from 3 g of total RNA using MMLV reverse transcriptase (Invitrogen) and oligo(dT) primers (Invitrogen) at 37°C for 1 h. Conventional PCR was performed in cDNA samples as previously described (30)  Expression of ER and PP2A was also quantified using real-time PCR. cDNA was amplified using SYBR green (Sigma) for ER, PP2A, and GAPDH at an annealing temperature of 60°C with 1-min annealing time for 40 cycles using the following primers: ER 5Ј-CTC TCC CAC ATC AGG CAC, A 3Ј-CTT TGG TCC GTC TCC TCC A; PP2A 5Ј-TAT CTC TAC GGT AGA ATT CAA CCA CAC GGG, 3ЈCTT ATC ACG CTC GCT CGC TGA CT; GAPDH 5Ј-GAA GGT GAA GGT CGG AGT C, 3Ј-GAA GAT GGT GAT GGG ATT TC. Real-time PCR data were acquired and analyzed using Sequence Detector v1.7 software (PerkinElmer) and normalized using GAPDH housekeeping gene detection.
PP2A Activity Assay-PP2A activity in whole cell MCF-7 lysates was determined by the PP2A serine/threonine phosphatase assay according to the manufacturer's protocol (Upstate Biotechnology, Charlottesville, VA). Briefly, PP2Ac subunit was immunoprecipitated from 200 g of nuclear extracts using 2 g of anti-PP2Ac antibody (Upstate clone 1D6) for 2 h at 4°C. Lysates were extensively washed in Tris-buffered saline and Ser-Thr phosphatase assay buffer. Phosphatase activity was determined by measuring conversion of phosphate, nPPNP, added directly to the beads for 15 min at 37°C. Accumulation of dephosphorylated substrate was measured in the supernatant using a spectrophotometer (Molecular Devices, Sunnyvale, CA) at 405-nm wavelength.
RNA Interference-MCF-7 cells were seeded at 70,000 cells/well in 6-well plates and allowed to adhere for 24 h. Cells were then transiently transfected with specific PP2Ac subunit and non-targeting DNA control probes (Dharmacon, Lafayette, CO) using Oligofectamine transfection reagent (Qiagen, Valencia, CA) according to the manufacturer's protocol. Briefly, Oligofectamine reagent (200 nM) was incubated with Opti-MEM reduced serum medium (Invitrogen) for 5 min at room tempera-ture. Specific PP2A-c siRNA probes (5 nM) or nontargeting DNA probes (5 nM) (Dharmacon) were added to the Oligofectamine mixture and incubated at room temperature for 10 min for proper transfection complex formation. Transfection mixtures were then added to cells in serum-free medium. After 4 h, serum concentrations in each well were adjusted to 5%, and cells were harvested after 72 h for RNA and protein.
Promoter Deletion and Luciferase Assay-MCF-7 cells were seeded at 2 ϫ 10 5 cells/well in 24-well plates 24 h prior to transfection. 3 l of Gene Jammer (Invitrogen) transfection reagent was used to transiently transfect 0.5 g of pERP pGL3 basic luciferase ER promoter deletion constructs (Ϫ2769, Ϫ1000, Ϫ745, or Ϫ245 to ϩ212) (kindly provided by Dr. Suzanne Fuqua, Baylor College of Medicine). 1 g of ␤-galactosidase expression vector was cotransfected with all plasmids to determine transfection efficiency. At 48 h, luciferase activity was measured on a Monolight luminometer using the BrightGlo luciferase assay kit (Promega), and ␤-galactosidase activity was determined using the ␤-galactosidase activity kit (Promega). Experiments were completed three times, and each measurement was taken in duplicate.
Electrophoretic Mobility Shift Assay-Nuclear proteins were harvested from MCF-7 cells with or without okadaic acid treatment for 16 h at 250 nM as previously described (31). Nuclear protein extracts (4 g/lane) were incubated with 8% glycerol and either poly(dI-dC) or poly(dG-dC) (0.2 g) (Pharmacia) as indicated for 20 min on ice. ␥-32 Plabeled probes (3 ϫ 10 4 counts/min/lane) corresponding to the Ϫ245 region of the ER promoter element were then added to cell extracts and incubated for 20 min at room temperature. The oligonucleotide used, 5Ј-ACCTTAGCAGATCCTCGT-3Ј (Ϫ245 to Ϫ225) included the SP1 binding site identified by deGraffenried et al. (22). Extracts were analyzed by 5% polyacrylamide gel electrophoresis run at 150 V in 0.5ϫ TBE (Tris borate-EDTA), followed by autoradiography.
Actinomycin D Assays-MCF-7 cells were seeded at 800,000 cells/ 10-cm plate for 24 h prior to treatment with or without okadaic acid (250 nM) and actinomycin D (4 g/ml; Sigma). RNA was isolated after 0, 2, and 8 h of treatment, and gene expression changes were quantified by quantitative real-time PCR.
Transient Transfection Assays-To construct the ER 3Ј-UTR expression plasmid, the 3Ј-UTR region of the ER (3.7 kb) and the ER coding sequence were PCR amplified using RPCI-1 human clone (130E4; Invitrogen). The amplified cDNA fragments and ER coding region were cloned in the NotI/XhoI sites of pBluescript II SKϩ (Stratagene, La Jolla, CA) and sequenced to verify accuracy. The ER 3Ј-UTR and coding region cDNAs were inserted into the mammalian expression vector pIREShyg2 (Clontech). The ER 3Ј-UTR and coding region construct was transiently transfected with Lipofectamine into MDA-MB-231 cells according to the manufacturer's protocol. Following transfection, cells were allowed to recover for 24 h. Cells were then treated for 4 h with OA (250 nM), the proteasomal inhibitor MG132 (10 M; Sigma), or a combination. Actinomycin D (4 g/ml) was subsequently added to block mRNA synthesis. Expression of ER mRNA was measured by real-time PCR and normalized to GAPDH expression. Experiments were performed twice, with each measurement taken in duplicate.

PP2A Subunits Are Abundantly Expressed in Human Breast
Cancer Cell Lines and Are Inhibited by Okadaic Acid Treatment-Western blotting of whole cell lysates from MCF-7, T47D, MDA-MB-231, and MDA-MB-468 cells showed that PP2A A, B, BЈ, and C subunits are abundantly expressed in several human breast cancer cells of variable phenotype (Fig.  1). OA is a widely used PP2A inhibitor, known to inhibit PP2A activity at nM concentrations through blockade of the catalytic subunit (PP2Ac). In MCF-7 cells, OA treatment had no effect on PP2A mRNA expression as determined by real-time PCR ( Fig. 2A) but significantly decreased PP2A activity (Fig. 2B) (p Ͻ 0.01).
Inhibition of PP2A Decreased ER mRNA, Protein, and Activity-Previous reports from our laboratory (29) and others (19) have suggested a role for PP2A in the regulation of ER. Therefore, the effect of OA on ER expression was examined in ERpositive MCF-7 cells. OA inhibition of PP2A resulted in an 80% reduction in ER mRNA expression as determined by real-time PCR (Fig. 3A) (p Ͻ 0.001). This decrease in ER mRNA translated into a significant decrease in ER protein (Fig. 3C). The decrease in ER was functionally significant as demonstrated by decreased mRNA and protein expression of the progesterone receptor (PR), a well recognized ER-responsive gene (Fig. 3, B and C) (p Ͻ 0.001). In contrast, OA treatment had no effect on protein expression of another member of the nuclear receptor family, the vitamin D receptor (VDR) (Fig. 3C), which unlike PR is not regulated by ER activity.
Genetic Inhibition of PP2A with Small Interfering RNAs Resulted in a Significant Decrease in ER Expression-Although the pharmacologic inhibitor OA has been shown to be 100 -1000-fold more specific for PP2A than for any other protein phosphatase family member, the effects of PP2A inhibition were confirmed using a genetic approach, small interfering RNAs (siRNA) targeting PP2Ac. Transfection of MCF-7 cells with siRNA against PP2Ac resulted in a significant decrease in PP2A mRNA and protein expression (Fig. 4, A and D) (p Ͻ 0.05). This inhibition of PP2A resulted in a significant decrease in ER mRNA and protein (Fig. 4, B and D) (p Ͻ 0.05). Inhibition also resulted in a decrease in ER activity as determined by a decrease in mRNA and protein expression of the ER-responsive gene, PR (Fig. 4, C and D) (p Ͻ 0.01). As with OA treatment, siRNA inhibition of PP2A had no effect on VDR protein expression, confirming the specificity of the effect of PP2A on the ER pathway (Fig. 4D).
PP2A Does Not Alter ER Promoter Activity-ER expression is regulated at both the transcriptional and post-transcriptional levels. To test whether PP2A exerts its effects on ER promoter activation, the effect of OA treatment on ER transcription was tested using MCF-7 cells transiently transfected with pERP pGL3 basic luciferase promoter constructs (Fig. 5A) (22). de-Graffenried et al. (22) have reported that the ER promoter region is constitutively active in MCF-7 cells and that the most proximal 245 base pairs upstream of the transcriptional start site contains a SP1 binding site that is necessary and sufficient for activation of the ER promoter. Analysis of ϳ3000 base pairs of the ER promoter (Ϫ2700 to ϩ212) in MCF-7 cells showed substantial ER promoter activity that is not altered with OA treatment (Fig. 5B). Although there appears to be a decrease in ER promoter activity with OA treatment of the Ϫ1000 construct, neither the full-length promoter Ϫ2700 construct nor the proximal Ϫ245 region that contains known transcription factor binding sites showed any difference in promoter activation between control and OA-treated MCF-7cells (Fig. 5B). Studies have shown binding of transcription factors, including SP1 and USF-1, to the Ϫ245 region of the ER promoter transactivates ER expression (22,23). Evaluation of transcription factor binding to the Ϫ245 construct via electrophoretic mobility shift assay analysis did not show changes in transcription factor binding in OA-treated MCF-7 nuclear extracts compared with control, consistent with our findings using the luciferase promoter constructs (Fig. 5C).
PP2A Activates ER through Modulation of mRNA Stability-Because OA treatment led to a decrease in ER mRNA by reverse transcription-PCR and real-time PCR without a difference in promoter activation or transcription factor binding, the role of the 3Ј-UTR in the regulation of ER expression was assessed using the 231ERϩ cell model. These cells were generated by stable transfection of an expression construct of ER cDNA cloned from MCF-7 cells under the control of a cytomegalovirus promoter. 231ERϩ cells express high levels of functional ER localized in the nucleus (Fig. 6 and data not shown). As the expression construct contains only the coding sequence of ER without the 3Ј-UTR, any mRNA-stabilizing effects of the 3Ј-UTR are absent in 231ERϩ cells. Surprisingly, treatment of 231ERϩ cells with OA resulted in a significant increase in ER protein expression (Fig. 6); similar results were obtained with MCF10A cells that were transfected with the same ER construct (data not shown). This enhanced ER expression stands in marked contrast with the profound decrease in ER expression following OA treatment of the innately ER-positive MCF-7 and T47D cells. To verify that decreases in ER expression detected in MCF-7 cells reflected changes in endogenous expression of ER, a second ER-positive cell line, T47D, was studied after treatment with OA (10 nM). As seen following PP2A inhibition in MCF-7 cells, ER protein expression in T47D cells was significantly inhibited to below the level of detection by Western blotting (Fig. 6).
PP2A Stabilizes ER mRNA through the 3Ј-UTR-Because OA treatment of MCF-7 cells did not alter promoter activity or transcription factor binding (Fig. 5, A and B), the contribution of the 3Ј-UTR to ER mRNA stabilization was evaluated. MCF-7 and 231ERϩ cells were treated with the transcriptional inhibitor actinomycin D (4 g/ml) in the presence or absence of OA (250 nM) for 0, 2, or 8 h. ER mRNA levels were then quantified by real-time PCR. The percent of mRNA remaining after actinomycin D inhibition of transcription was significantly reduced in MCF-7 cells treated with OA and actinomycin D compared with actinomycin D alone. ER mRNA half-life was significantly reduced from 4.5 to ϳ1.5 h (Fig. 7A) (p Ͻ 0.006). As expected, no change in ER mRNA decay or half-life was detected in 231ERϩ cells (that lack the ER 3Ј-UTR) in the presence or absence of OA (Fig. 7B).
We next tested whether inhibition of the proteasome using the specific inhibitor MG132 would block OA-mediated ER mRNA degradation, suggesting that PP2A-mediated control of ER expression occurs through protein interactions with the ER 3Ј-UTR. MDA-MB-231 cells were transiently transfected with the ER 3Ј-UTR construct containing 3.7 kb of the 3Ј-UTR and treated for 4 h with OA (250 nM), MG132 (10 M), or a combination of OA and MG132. Actinomycin D (4 g/ml) was then added to inhibit further mRNA synthesis. The percent of ER mRNA remaining was measured by real-time PCR and normalized to GAPDH mRNA expression. Inhibition of the proteaso-mal pathway prevented OA-mediated ER mRNA degradation (Fig. 8). Similar studies using the Hsp70 inhibitor KNK437 (Calbiochem) to ascertain whether the Hsp70 chaperone protein might play a role in OA-mediated ER mRNA degradation showed no effect (data not shown). Together these results indicate that PP2A stabilizes ER mRNA through interactions with the 3Ј-UTR and provide a link between PP2A activation and proteasomal degradation. DISCUSSION ER expression is regulated at multiple levels in human breast cancer cell lines. These include transcription factorpromoter interactions at the transcriptional level, phosphorylation at the post-transcriptional level, and degradation through altered mRNA stability and proteasomal degradation (22, 23, 26 -28). ER expression can also be silenced through epigenetic mechanisms in some ER-negative human breast cancer tumors (24,25,30,32,33). All of these mechanisms appear to be interrelated and tightly control the regulation of ER expression in the cell. MCF-7 cells are the best characterized model of ER-expressing human breast cancer cells. One mechanism controlling ER expression in these cells is through activation of the promoter region and interaction of transcription factors with the proximal promoter (22,23). Studies have also shown that ER contains a long 3Ј-UTR that contains numerous AUUA sequences and binding sites that have been shown to affect mRNA stability and contribute to regulation of ER gene expression in MCF-7 cells (27,28).
Several studies have implicated PP2A in the regulation of ER expression. Gopalakrishna et al. (19) first reported a correlation between PP2A activity and ER expression in a panel of human breast cancer cell lines. Subsequent studies showed that PP2A activation results in inhibition of the mitogen-activated protein kinase pathway (extracellular signal-regulated kinase), resulting in decreased ER phosphorylation and activity without an alteration of ER protein expression (7,8,34,35). More recently, Lu et al. (36) reported that PP2A inhibition increases ER expression in rat pulmonary vein endothelial cells infected with an adenoviral ER-green fluorescence protein (GFP) construct, Rad91 adeno-GFP-ER. Our previous data suggest a role for PP2A in mediating ER expression (29). In a microarray analysis of ER-negative MDA-MB-231 cells treated with epigenetic modulators that induce ER expression, upregulation of PP2A expression was detected in conjunction with ER, suggesting that increased PP2A expression may play a role in the regulation of ER expression (29). Based on these findings, we sought to further characterize the role of PP2A in ER expression in MCF-7 cells.
Inhibition of PP2A activity, either through pharmacologic inhibition with OA or by genetic knockdown by siRNA, resulted in a significant decrease in ER mRNA and protein expression (Figs. 3 and 4) and diminished ER activity as determined by down-regulation of an ER-responsive gene, PR. This PP2Amediated gene expression control is specific for the ER pathway in that no change in expression of another member of the nuclear steroid receptor family, VDR, was detected using either method (Figs. 3C and 4D).
Although ER expression is regulated through transcriptional mechanisms, PP2A-mediated ER expression does not involve increased promoter activation or transcription factor interactions (Fig. 5). Transcriptional regulation of ER requires binding of transcription factors including SP1 and USF1 to the proximal promoter region (37). PP2A has been shown to physically interact with SP1 in T lymphocytes, suggesting that a potential mechanism of PP2A-mediated ER expression could result from changes in binding of SP1 and activation of ER transcription (38). However, no change in ER promoter activation using There was no change in VDR expression following PP2A inhibition. Actin expression was used as a loading control. luciferase ER promoter constructs or transcription factor binding to the ER P1 promoter was detected, thereby effectively eliminating transcriptional control as a mechanism of PP2Amediated ER expression in MCF-7 cells (Fig. 5, B and C).
Unlike in MCF-7 or T47D cells, OA treatment of 231ERϩ cells gave the seemingly paradoxical result of increased ER expression (Fig. 6). Similar findings were seen after OA treatment of MCF10A cells engineered to overexpress ER (data not shown) as well as Rad91 cells overexpressing ER cDNA (36). The overexpressed ER cDNA in the 231ERϩ and MCF10AERϩ cells was cloned from MCF-7 cells and shares an identical coding sequence, but it lacks the endogenous ER 3Ј-UTR that contains numerous binding sites for AUUA-and other RNAbinding proteins. The importance of this region for ER regulation is supported by the actinomycin and proteasome inhibition studies that show a significant decrease in ER mRNA half-life in OA-treated MCF-7 cells compared with untreated MCF-7 cells and no difference in ER mRNA half-life in 231ERϩ cells that lack the 3Ј-UTR (Fig. 7). This OA-mediated ER inhibition was clearly reversed in the presence of the proteasomal inhibitor MG132 (Fig. 8) but not altered in the presence of the Hsp70 chaperone protein inhibitor KNK437 (data not shown). That this 3Ј-UTR mechanism might be more global is suggested by findings after PP2A inhibition of another breast cancer cell line, MDA-MB-468, that resulted in decreased c-fos mRNA stability through interaction with AUUA sequences in the c-fos FIG. 5. PP2A-mediated ER inhibition does not involve the ER promoter. A, schematic of the ER proximal promoter region indicating SP1 binding site and 1/2 estrogen response elements. pGL3 basic constructs containing varying lengths of the ER promoter along with ␤-galactosidase plasmids (used to control for transfection efficiency) were transiently transfected into MCF-7 cells. B, cells were then treated with 250 nM OA for 16 h, and relative luciferase expression was measured on a Moonlight luminometer. Shown are the mean and S.E. of luciferase relative light units (RLU) normalized to ␤-galactosidase expression for three experiments conducted in duplicate. No difference in overall promoter activation was detected with OA treatment. C, electrophoretic mobility shift assay using a radiolabeled oligonucleotide specific for the SP1 site (5Ј-ACCTTAGCAGATC-CTCGT-3Ј (Ϫ245 to Ϫ225) did not show any alterations in transcription factor binding following OA treatment. The free probe lane was added to control for nonspecific binding. RNA was harvested at 0, 2, and 8 h of treatment. The percent of ER mRNA remaining following the inhibition of transcription was determined by real-time PCR. OA treatment significantly reduced the ER mRNA half-life from 4.5 to 1.5 h (p Ͻ 0.006 as determined by Student's t test) and had no effect on 231ERϩ cells. Gene expression was normalized to GAPDH housekeeping gene expression. Shown is the mean Ϯ S.E. of two experiments, each conducted in duplicate.
3Ј-UTR, suggesting that PP2A may increase c-fos mRNA stability as well (38). Experiments using the specific proteasomal inhibitor MG132 in conjunction with OA treatment clearly provide a link between PP2A and the proteasome. MG132 treatment reversed the OA-mediated ER mRNA decay, indicating that PP2A regulates ER expression through modulation of the ER 3Ј-UTR, resulting in proteasomal degradation (Fig. 8), suggesting that ER mRNA decay is regulated by PP2A-mediated binding of a factor to the 3Ј-UTR. Proteasomal degradation of this protein leads to ER mRNA degradation in the absence of PP2A activity. We, therefore, propose that PP2Amediated ER mRNA stability depends on the presence of and protein binding to the ER 3Ј-UTR.
In summary, our data suggest that PP2A plays a significant role in ER gene expression by increasing ER mRNA stability and half-life. Further, the importance of the 3Ј-UTR region in regulating ER expression is highlighted. Finally, these studies that utilize cell lines that endogenously express ER as a physiologically relevant model suggest that PP2A could be an effective treatment target. Actinomycin D (4 g/ml) was then added to block nascent RNA synthesis, and the cells were incubated for 4 more hours under normal culture conditions before harvest for ER mRNA measurement. The percent of ER mRNA remaining was measured by real-time PCR and normalized to GAPDH mRNA levels. Shown is the mean Ϯ S.E. from two experiments, each conducted in duplicate. ER mRNA levels are significantly decreased by OA treatment compared with control (**, p Ͻ 0.01 as determined by Student's t test). Concomitant treatment with both OA and the proteasomal inhibitor MG132 eliminates the effect of OA on ER mRNA degradation (*, p Ͻ 0.05 compared with OA treatment alone, determined by Student's t test).