PME-1 Modulates Protein Phosphatase 2A Activity to...

Molecular and Cellular Pathobiology

PME-1 Modulates Protein Phosphatase 2A Activity toPromote the Malignant Phenotype of Endometrial CancerCells

Ewa Wandzioch1, Michelle Pusey1, Amy Werda1, Sophie Bail1, Aishwarya Bhaskar1, Mariya Nestor2,Jing-Jing Yang2, and Lyndi M. Rice1

AbstractProtein phosphatase 2A (PP2A) negatively regulates tumorigenic signaling pathways, in part, by supporting the

function of tumor suppressors like p53. The PP2A methylesterase PME-1 limits the activity of PP2A bydemethylating its catalytic subunit. Here, we report the finding that PME-1 overexpression correlates withincreased cell proliferation and invasive phenotypes in endometrial adenocarcinoma cells, where it helpsmaintain activated ERK and Akt by inhibiting PP2A. We obtained evidence that PME-1 could bind and regulateprotein phosphatase 4 (PP4), a tumor-promoting protein, but not the related protein phosphatase 6 (PP6). Whenthe PP2A, PP4, or PP6 catalytic subunits were overexpressed, inhibiting PME-1 was sufficient to limit cellproliferation. In clinical specimens of endometrial adenocarcinoma, PME-1 levels were increased and we foundthat PME-1 overexpression was sufficient to drive tumor growth in a xenograft model of the disease. Our findingsidentify PME-1 as amodifier of malignant development and suggest its candidacy as a diagnostic marker and as atherapeutic target in endometrial cancer. Cancer Res; 74(16); 4295–305. �2014 AACR.

IntroductionProtein phosphatase 2A (PP2A), a heterotrimeric serine/

threonine phosphatase, composed of a scaffolding subunit(A), a catalytic subunit (C), and a B regulatory subunit, isimplicated as a human tumor suppressor (reviewed inrefs. 1–4). The core AC dimer recruits a B regulatory subunitgiving the complex substrate specificity. There are 4 familiesof B regulatory subunits, composed of several genes thatcode for multiple isoforms (2, 3). Specific heterotrimersdephosphorylate and stabilize tumor suppressors, such asp53 (5, 6) and p107 (7); thus, PP2A activity is important incell-cycle regulation and tumor suppression. PP2A has beenshown to dephosphorylate positive regulators of cell signal-ing pathways such as ERK and Akt (8–11), inhibiting themand promoting senescence or apoptosis. Inhibition of PP2Apromotes enhanced cell proliferation, impairment of celldifferentiation, malignant cell transformation (reviewed

in refs. 3, 4) and is thought to be a key step in cellulartransformation (12).

The carboxy-terminal tail of the catalytic subunit of PP2A ishighly posttranslationally modified, affecting PP2A activitythrough mechanisms such as altering B subunit recruitmentand the physical blockade of the catalytic site (13, 14). The C-terminal tail of Ppp2ca/b, the two isoforms of its catalyticsubunit, are conserved with the C-terminal tails of Ppp4c andPpp6c, the catalytic subunits of protein phosphatases 4 and 6(PP4, PP6), respectively, suggesting they may be regulatedsimilarly to PP2A. The reversiblemethylation of the C-terminalleucine of the catalytic subunit of PP2A provides an interest-ing molecular mechanism of regulation. Methylation is cata-lyzed by leucine carboxyl methyltransferase (LCMT1; ref. 15)and has been shown to enhance both the catalytic activity ofPP2A and the recruitment of specific B subunits to the PP2Acomplex (16, 17). The removal of the methyl group is catalyzedby protein phosphatase methylesterase 1 (PME-1; refs. 14, 18)and renders PP2A inactive (13, 19, 20). The conservation ofcatalytic subunits of PP4 and PP6 with that of PP2A suggeststhat PME-1 may also modulate PP4 and PP6 activities.

PME-1–mediated PP2A inhibition causes increased prolifer-ation and activation of the ERK pathway and promotes malig-nant cell growth of human glioblastoma cells (21). PME-1expression forces progression of low-grade astrocytic gliomasto malignant glioblastomas (22). PP2A targets RalA, Ras/Raf/MEK/ERK, and PI3K/AKT pathways (18, 23–25), and it has beenshown that the simultaneous activation of ERK and PI3K/Aktpathways is highly oncogenic in breast cancer cells (26).

Although a few PME-1 tumor-promoting mechanisms havebeen reported (9, 21), the role of PME-1 in cancer inductionand progression remains to be elucidated. Since endometrial

1Oncoveda, Cancer Signaling and Cell Cycle Team, Medical DiagnosticLaboratories, LLC, a Member of the Genesis Biotechnology Group, LLC,Hamilton, New Jersey. 2Pathology Division, Medical Diagnostic Labora-tories LLC, aMember of the Genesis Biotechnology Group, LLC, Hamilton,New Jersey.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

E. Wandzioch and M. Pusey contributed equally to this work.

Corresponding Author: Lyndi M. Rice, Oncoveda, Cancer Signaling andCell Cycle, Medical Diagnostic Laboratories LLC, 1000 Waterview Drive,Room 345, Hamilton, NJ 08691. Phone: 609-786-2871; Fax: 609-570-1039; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-13-3130

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 4295

on May 31, 2018. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 13, 2014; DOI: 10.1158/0008-5472.CAN-13-3130

http://cancerres.aacrjournals.org/

cancer is the most common gynecologic cancer affectingwomen in the United States, we asked whether PME-1 playsa role in endometrial cancer progression. We show thatincreased PME-1 correlates to decreased PP2A activity andincreased proliferation andmetastatic phenotypes through themaintenance of increased ERK and Akt signaling. We demon-strate an interaction between PME-1 and PP4, suggesting thatPME-1 is not specific for PP2A. Increased levels of PME-1 weredetected in endometrial adenocarcinoma tumors, suggestingthat PME-1 may be a diagnostic marker for patients withendometrial cancer. Currently, there are no available specificbiomarkers for endometrial adenocarcinoma. Finally,increased PME-1 led to increased anchorage-independentgrowth and increased tumor burden in a xenograft model.Our results suggest a role for PME-1 in the promotion of cancerprogression in endometrial cancer and may be a valid drugtarget for endometrial cancer treatment.

Materials and MethodsDetailed experimental procedures can be found online in

Supplementary Materials.

Cell cultures and generation of stable cell linesAll cell lines were purchased from ATCC and were main-

tained according to ATCC recommendations. Analyses ofECC-1 cells (ATCC #CRL-2923) have determined ECC-1 cellspurchased from ATCC to be genetically similar and redundantto the Ishikawa endometrial cancer cell line (27). All cell lineswere authenticated in November of 2013 via STR Analysis(Genetica DNA Laboratories). Vectors were transfected intocells with Lipofectamine 2000 (Life Technologies) or usinglentivirus (System Biosciences, Inc). shRNA sequences are inthe Supplementary Table S1. Lentivirus (5.0� 106 ifu/mL) wasadded to cell cultures (Sigma Aldrich). Cells were selected withpuromycin after 48 hours (Invivogen).

Foci formation assaysTo measure cell proliferation, 1,000 cells were plated in a 6-

well tissue culture dish in selection media and were grown for10 to 14 days. Experiments were repeated at least three times.BrdUrd incorporation experiments were conducted asdescribed in Supplementary Materials.

Analysis of endometrial cancer patient samplesMatched pairs harvested from 30 patients with type I

endometrial adenocarcinoma purchased from Proteogenex(see Supplementary Table S2) were used to determine themRNA and protein levels of PME-1 in tumor versus normaladjacent tissue. Further analysis was conducted via IHC tech-niques and immunofluorescence assays (see SupplementaryMaterials).

Phosphatase activity assayWhole-cell lysates were used with the DuoSet IC PP2A

Phosphatase Assay Kit (R&D Systems). The procedure wascompleted per kit instructions with 250 mg protein. The assaywas repeated at least three times with similar results.

TaqMan RT-PCR analysisTotal RNA was used to generate cDNA for analysis of gene

expression. TaqMan probes and PCR-mix were purchasedfrom Life Technologies. Results are representative of threeindependent experiments in which genes of interest werenormalized to the housekeeping genes, 18S or GAPDH.

ERK/Akt inhibition, protein extraction, andWesternblotanalysis

RL95-2 cells were incubated with either 50 mmol/L Aktinhibitor (LY294002, Cell Signaling Technologies) or 40mmol/L ERK inhibitor (UO126, Cell Signaling Technologies)for 1 to 2 hours at 37�C. Laemmli buffer (1.5 � 0.5 mol/L Tris,pH 6.8, 100% glycerol, 10% SDS, 100mmol/L EDTA)was used toprepare cell lysates. All experiments were repeated severaltimes with similar results.

Colony formation assaysTo measure invasive growth phenotypes, the 3D On-Top

Matrigel Assay was completed in 24-well dishes as describedpreviously (28).

In vivo tumor formationA total of 1 � 106 endometrial carcinoma cells (ECC-1)

diluted in 100 mL 1� PBS expressing empty vector (Control) oroverexpressing PME-1 (þPME-1) were injected subcutaneous-ly into the flank of nude female mice (n¼ 7 per group). Tumorformation wasmeasured weekly for 7 weeks with a caliper, andtumor volume was calculated according to the formula V ¼1/2yx

2, where y¼ tumor length and x¼ tumorwidth. At 8 weekspostinjection, mice were euthanized and tumors were resectedfor analysis. All animal work was approved by and conductedaccording to the guidelines of the Genesis BiotechnologyGroup Institutional Animal Care and Use Committee.

Statistical analysisStatistical analysis was completed using GraphPad Prism

version 5.02 forWindows, GraphPad Software (www.graphpad.com). The data were analyzed with the Mann–Whitney U testfor significance (patient samples) or the Student standard ttest; SEMswere calculated for all sample batches. ROC analysisplotting the sensitivity and specificity of PME-1mRNA levels inendometrial cancer patient samples was calculated with 95%confidence interval (CI) to determine the validity of PME-1 as abiomarker for endometrial cancer using a likelihood ratio of 21.The cutoff was determined using the GraphPad Software. Thedetermined P value (P< 0.0001) and area under the curve (AUC,0.9601) suggest that PME-1 could be a valuable predictor ofendometrial cancer. Data from animal studies were analyzedusing 2-way ANOVA. �, P < 0.05; ��, P < 0.01; ��, P < 0.001.

ResultsIncreased PME-1 levels correlate with increasedcancerous phenotypes and decreased PP2A activity

Increased PME-1 has previously been noted in glioblastomas(21) and increases cell proliferation in the cervical cancer cellline, HeLa (refs. 20, 21 and data not shown), but the role(s) of

Wandzioch et al.

Cancer Res; 74(16) August 15, 2014 Cancer Research4296




PME-1 in endometrial cancer initiation and/or progression hasnot been elucidated. Endometrial cancer is the most commongynecologic cancer in the United States with no reliable

noninvasive diagnostic available for early detection of endo-metrial cancer. We examined the expression levels of PME-1 inendometrial cancer cell lines byWestern blot analysis (Fig. 1A)

Figure 1. PME-1 promotes cancer phenotypes through inhibition of PP2A activity in endometrial cancer (EC) cells. A, Western blot analysis of PME-1 levels inEnd1 (immortalized endocervical cell line) and the endometrial cancer cell lines, RL95-2, ECC-1, and Ishikawa cells. GAPDH serves as a loading control. B,Western blot analysis of PME-1 protein in RL95-2 cells expressing empty vector (Control), overexpressing PME-1 (þPME-1), or expressing two differentshRNAs (sh #1, sh #2) againstPPME1mRNA (�PME-1); GAPDH serves as a loading control. C, a total of 1,000 cells per well were plated for each RL95-2 cellline and were grown for 10 days to measure cell proliferation. Data from seven independent studies are represented as percent of control. D, BrdUrdincorporation assays were completed tomonitor changes in cell proliferation by altering PME-1 levels and were completed three times in triplicate. E, RL95-2cellswere transfectedwith nontargeting siRNAor siRNAagainst the 30-UTRofPPME1. Cellswere cotransfectedwith empty vector (Control) or inactivePME-1S156A. Data are normalized to RL95-2 cells treated with control siRNA and the empty vector. See Materials and Methods for more detail. Foci experimentswere completed as above. Data represent five independent experiments. F, a total of 1 � 105 cells were grown in Matrigel and were stained with 1% crystalviolet and counted after 14 days of growth to measure anchorage-independent growth. Data are presented as percent of control and represent fourindependent experiments. Analysis of PP2A activity in RL95-2 cells when PME-1 was depleted using shRNA (G) or PME-1 was overexpressed (H). Controlsampleswere normalized to 100%PP2Aactivity and each experimentwas conducted three times. Significancewas calculatedby the standardStudent t test,where �, P < 0.05; ��, P < 0.01; ��, P < 0.001.

PME-1 Promotes Endometrial Cancer

www.aacrjournals.org Cancer Res; 74(16) August 15, 2014 4297




and found that the endometrioid adenocarcinoma cell lines,RL95-2, Ishikawa, and ECC-1, express more PME-1 than theimmortalized endocervical cell line, End1. No immortalizedendometrial cell lines exist due to misidentification (27), thus,we used the endocervical cell line as it is derived from a similartissue.

Wenext developed RL95-2 cell lines expressing empty vector(control), overexpressing PME-1 (þPME-1), or expressingshRNA (sh#1, sh#2) against PPME1 mRNA, which codes forPME-1 protein (�PME-1) and confirmed the appropriate levelof PME-1 via real-time RT-PCR (data not shown) and Westernblot analysis (Fig. 1B). Similar cell lines were constructed forECC-1 cells aswell as theKLE endometrial cancer cell line (datanot shown). Multiple shRNAs and siRNAs were used for theseexperiments (see Supplementary Table S1), all with similarresults. To determine the role of PME-1 in cell proliferation, wecompleted foci formation assays, which demonstrated a sig-nificant 50% increase in foci when PME-1 is overexpressed anda significant 75% decrease in foci when depleted for PME-1compared with control cells (Fig. 1C). We noted a significant 2-fold increase in foci formed in KLE EC cells when PME-1 wasoverexpressed (data not shown). To confirm that altering PME-1 levels in endometrial cancer cells affects their rate of pro-liferation, we completedBrdUrd incorporation assays (Fig. 1D).We found a significant 40% increase in BrdUrd incorporationin þPME-1 cells compared with empty vector control andabout 40% decrease proliferation in -PME-1 cells comparedwith a scrambled shRNA control. Taken together, these datasuggest that PME-1 promotes cell proliferation.

We next treated RL95-2 cells with siRNA against the 30-untranslated region (UTR) of PPME1 to decrease endogenouslevels of PPME1 and overexpressed either empty vector or acatalytically inactive form of PME-1, S156A, in which thenucleophilic serine residue was mutated to an alanine (Fig.1E). The decrease in foci due to decreased endogenous PME-1was not rescued by overexpression of the S156A mutant;however, increased foci formation was rescued by the over-expression of wild-type PME-1 (data not shown), suggestingthat active PME-1 is required for increased cell proliferation inRL95-2 cells. To determine whether PME-1 promotes anchor-age-independent growth, we completed soft agar assays andcounted the colonies formed. We determined that PME-1overexpression led to a significant 50% increase in colonyformation, whereas PME-1 inhibition led to a significant50% decrease in colony formation (Fig. 1F). Thus, PME-1promotes several cancer phenotypes.

The decrease in cell proliferation due to loss of PME-1 wasnot due to apoptosis, as demonstrated by a terminal deox-ynucleotidyl transferase–mediated dUTP nick end labeling(TUNEL) assay, but was instead due to cells senescing, evi-denced by substantial increase in the senescence proteinmarker, DcR2 (data not shown). This was accompanied by a20- to 40-fold increase in PP2A activity (Fig. 1G) likely due toincreased methylation of the catalytic subunit of PP2A (13, 19).Overexpression of PME-1 inhibited PP2A activity by about 90%(Fig. 1H). Thus, PME-1 inhibition leads to decreased cellproliferation and senescence via increased PP2A activity inendometrial cancer cell lines.

Overexpression of PME-1 increases ERK/Akt activationBecause increased PME-1 levels correlate with increased

cellular proliferation and PP2A is known to negatively regulatethe ERK and Akt pathways (24, 29, 30), we asked whetherincreased PME-1 led to the activation of these signaling path-ways. Western blot analysis demonstrated that the ERK phos-phorylation was altered with manipulation of PME-1 levels inRL95-2 cells (Fig. 2A) and in the KLE endometrial cancer cellline (data not shown); whenPME-1 is overexpressed, there is anincrease in phosphorylated ERK, whereas total ERK levelsremain unchanged (Fig. 2A). Upon PME-1 depletion, there isdecreased phospho-ERK compared with control cells. Wenoted decreased levels of phosphorylated ERK, as expected,upon treatment of cells with UO126, an upstream inhibitor ofERK phosphorylation.

Similarly, phosphorylation of Akt on threonine 308 (T308)and on serine 473 (S473) increased when PME-1 was over-expressed compared with control and decreased when PME-1was depleted (Fig. 2B). Similar results were recently publishedby Jackson and Pallas (9), although they noted increasedphosphorylation only at T308. We detected decreased phos-phorylation of Akt, as expected, upon treatment of cells with

Figure 2. Overexpression of PME-1 induces increased cell proliferationthrough activation of ERK and Akt signaling pathways. Western blotanalysis of RL95-2 cells expressing control vector, vector with PME-1, orvector carrying shRNA against PME-1 examining levels of ERKphosphorylation compared to total ERK (A) and Akt phosphorylation atthreonine 308 and serine 473 compared with total Akt levels (B). GAPDHserves as a loading control. Samples treated with the inhibitors of ERK(UO126) and Akt (LY294002) serve as controls.

Wandzioch et al.





the upstream inhibitor, LY294002. Therefore, PME-1 is a pos-itive regulator of the ERK and Akt cancer signaling pathways inendometrial cancer cells.

PME-1 is overexpressed in endometrial cancer patienttumorsUterine tissue samples from 30 patients diagnosed with type

I endometrioid adenocarcinoma were purchased from Proteo-genex (Supplementary Table S2). RNA and protein wereextracted from each matched pair, tumor (T), and normaladjacent tissue (NAT, N). Note that the ratio of PPME1 mRNAexpression is increased in tumor versus NAT in 24 of 29 patientsamples (�83%, Supplementary Table S2), suggesting thatincreased PPME1 mRNA is indicative of endometrial cancer.PPME1 expression was increased about 20-fold in tumorsamples versus normal samples (Fig. 3A). ROC analysis (Fig.3B) demonstrates that detection ofPPME1mRNA levelsmay bea diagnostic marker, as the area under the curve (AUC) score is

>0.9, the sensitivity of the assay is 80.77%, and the specificity is96.15%, suggesting that PPME1 levels may be predictive ofendometrial cancer. Moreover, PPME1mRNA expression witha cutoff of 8.14 (determined using Prism software) demon-strates a 95.5% positive predictive value (PPV) and an 83.3%negative predictive value (NPV).

We next tested the patient samples for PME-1 protein andcompared it with the loading control, COX IV (Fig. 3C).Representative Western blots are displayed for FederationInternationale des Gynaecologistes et Obstetristes (FIGO)grades 1–3. Grade 1 samples (patients 06261, 06310, 06308,06276) are shown in the top left and grade 2 samples (patients06336, 06241, 06247) are shown in the top right. Grade 3samples (patients 06268, 06294) are shown in the bottom.Most tumor samples (T) exhibit increased PME-1 proteinwhencompared with normal adjacent tissue (N).

Patient samples were assessed to determine the stage andgrade according to the FIGOguidelines andwere analyzedwith

Rel

ativ

e P

ME

1 ex

pre

ssio

n

NAT Tumor

1208050

40

30

20

10

0

Sen

siti

vity

100

80

60

40

20

00 20 40 60 80 100

100-Specificity

P < 0.0001

P < 0.0001AUC = 0.9601

Grade 1 Grade 2

Grade 3

Grade 1 Grade 2 Grade 3

N N N N N N NT T T T T T T

PME-1

COX IV

N NT T

PME-1

COX IV

H&

EH

&E

/αP

ME

-1

A B

C

D

E

Figure 3. PME-1 expression isincreased in endometrial cancerpatient samples compared withNAT. A, analysis of 30 type Iendometrial cancer patientsamples by qRT-PCR examiningthe levels of PME-1mRNA in tumorversus NAT. Means are marked byblack bars, error bars representSEM. P value is calculated with thenonparametric Mann–Whitneytest. B, ROCanalysis of data in A todetermine the efficacy of PME-1expression to be predictive ofendometrial cancer. C, Westernblot analysis of representativeendometrial adenocarcinomapatient samples determining PME-1 protein levels in FIGO grade 1(top left), FIGO grade 2 (top right),and FIGO grade 3 tumor samples(bottom). COX IV serves as aloading control. D, H&E staining ofFIGO grades 1–3 type Iendometrial cancer. E, tissuesamples stained with anti-PME-1antibody are shown to illustratePME-1 staining grading scale. 0indicates a negative stain, in whichno cytoplasmic staining isobserved or faint staining isobserved in less than 50% of cells;1þ indicates a weakly positivesample with weak cytoplasmicstaining detected in more than50% of the cells or moderate/strong staining detected in lessthan 50% of cells; and 2þindicates a strongly positivesample, in which moderate/strongstaining is observed in more than50% of cells.






hematoxylin and eosin (H&E) staining and/or a-PME-1 anti-bodies (Fig. 3D and E). Grade 1 tumors are well-differentiatedcancers with clear cellular boundaries and normal cell mor-phology. Grade 2 tumors are moderately differentiated withabnormal cell morphology. Grade 3 tumors are poorly differ-entiated exhibiting loss of clearly defined boundaries andhighly abnormal cell morphology (Fig. 3D). PME-1 proteinlevels were assessed as follows: 0 indicates negative cyto-plasmic PME-1 staining or faint staining observed in <50% ofthe cells; 1þ indicates weakly positive cytoplasmic PME-1staining in >50% of tumor cells or moderate/strong stainingin <50% of tumor cells; and 2þ indicates strongly positivestaining with moderate/strong cytoplasmic staining observedin >50% of the tumor cells. Representative images for PME-1grading are shown (Fig. 3E). Of the 9 FIGO grade 1 samplesexamined, 55% were 1þ and 45% were 2þ, whereas of the 7FIGO grade 2 samples examined, 43% were 1þ and 57% were2þ. All FIGO grade 1 and 2 samples exhibited weak-to-strongstaining for PME-1. Only 3 FIGO grade 3 samples were exam-ined by IHC for PME-1 and no trend was determined (Table 1).

PME-1 and P-cadherin costaining may indicate canceraggressivity

Our data suggested that PME-1 levels correlate withincreased cancerous phenotypes (Fig. 1) and that PME-1 levelsare increased in endometrial cancer versus normal tissue (Fig.3). E-cadherin is a marker for epithelial cells and is commonlydecreased in advanced tumors, whereas increased P-cadherinis indicative of more aggressive endometrial cancer (reviewedin ref. 31). We completed immunofluorescent studies (repre-sentative images are shown) to determine whether there was acorrelation among PME-1 and E-cadherin or P-cadherin ingrade 1 and 3 endometrial cancer patient samples. We foundthat in both grade 1 (patient 06313) and grade 3 (patient 06294),PME-1 and P-cadherin expression was increased comparedwith corresponding NAT samples (Fig. 4A). In grade 1 samples,more cells expressed PME-1 and P-cadherin, but in grade 3

samples, we noted single cells staining brightly for bothproteins (arrows in merge). We then examined PME-1 andE-cadherin expression in grade 1 and 3 tumors. There is overlapin PME-1 and E-cadherin expression in the grade 1 sample (Fig.4B, arrows inmerge), but in grade 3 endometrial cancer, PME-1is not co-expressed with E-cadherin, suggesting that the cellsexpressing PME-1 in grade 3 tumors are no longer epithelialbut may be mesenchymal in nature.

PME-1 interacts with other protein phosphatases familymembers

Several reports suggest that PME-1 is specific for PP2A(13, 18, 32); however, we hypothesized that this is unlikely asthe C-terminal tail of the catalytic subunit of PP2A (TPDYFL),the substrate for PME-1 demethylation, is similar to that ofprotein phosphatases 4 (VADYFL) and 6 (TTPYFL), or PP4 andPP6, respectively. The catalytic subunit of PP2A is coded for bytwo genes, PPP2CA and PPP2CB that produce two proteins,Ppp2ca and Ppp2cb, respectively, which are 97% identical.Their C-termini are 100% identical. Ppp4c and Ppp6c share63% and 56% identity with Ppp2ca/b, respectively. To deter-mine whether PME-1 specifically targets PP2A, we completedseveral co-immunoprecipitation studies in the endometrialcancer cell line, ECC-1. Empty vector (FLAG), wild-typePME-1 (FLAG-PME-1), or the inactive mutant of PME-1(FLAG-PME-1 S156A) were transiently expressed in ECC-1cells before immunoprecipitation on FLAG resin. Sampleswere analyzed by Western blot analysis and we probed forendogenous levels of Ppp2ca (Fig. 5A) and Ppp2cb (data notshown). We found that both wild-type and the inactive form ofPME-1 (FLAG-PME-1 S156A) are capable of binding Ppp2ca(Fig. 5A), although the interaction of wild-type PME-1 withPpp2ca is markedly weaker than the interaction of the inactiveform of PME-1. This is likely due to a transient interactionbetween wild-type PME-1 and Ppp2ca, which does not occurwith the inactive form of PME-1 as it cannot hydrolyze themethyl group from Ppp2ca. We also confirmed that wild-typePME-1 and the S156A mutant are capable of interacting withendogenous Ppp2cb (data not shown).

We next asked whether PME-1 is able to associate with theconservedPP4 andPP6 catalytic subunits. The same constructswere expressed as above and we probed for endogenous levelsof Ppp4c and Ppp6c to confirm association with PME-1. PME-1does in fact bind the catalytic subunit of PP4, but we could notdetect an interactionwith PP6 (Fig. 5B). The inactivemutant ofPME-1 boundmore strongly to Ppp4c comparedwithwild-typePME-1, suggesting that PME-1 may demethylate PP4.

We next determined whether PME-1 has a preference forbinding PP2A or PP4. For this study, we co-expressed emptyFLAG vector or FLAG-tagged Ppp2ca or Ppp4c, with empty V5vector or V5-tagged wild-type PME-1 or inactive PME-1 S156Ain HEK293T cells. Figure 5C demonstrates that PME-1 andPME-1 S156A were expressed equally across all studies (seeinput) and Ppp2ca and Ppp4c were expressed to equal levels.Likewise, when the catalytic subunits were immunoprecipi-tated (see FLAG elutions), there was equal Ppp2ca and Ppp4cpulled down. Weak interactions were detected with wild-typePME-1 (V5-PME-1) with either Ppp2ca or Ppp4c. There is a

Table 1. PME-1 is increased in endometrialcancer and may correlate with grade

FIGO grade

PME-1 stain 1 2 3

0 0 0 11þ 5 3 12þ 4 4 1

NOTE: Patient tissues were sectioned and analyzed afterH&E and immunostaining with anti-PME-1 antibody. PME-1protein levels were assessed as follows: 0 indicates negativePME-1 cytoplasmic staining or faint staining observed in<50% of the cells; 1þ indicates weakly positive PME-1cytoplasmic staining in >50% of tumor cells or moderate/strong staining in <50% of the cells; and 2þ indicatesstrongly positive staining in >50% of the tumor cells.

Wandzioch et al.





stronger association between V5-PME-1 S156A and Ppp2cathan with Ppp4c, suggesting that while PME-1 can associatewith Ppp4c, it has a higher affinity for Ppp2ca.These data suggest that PME-1 can target both PP2A and

PP4. While PP2A has tumor suppressor activity (4, 33, 34), PP4is implicated in tumor-promoting pathways and is overex-pressed in certain cancers (35, 36). Thus, we were concernedthat inhibition of PME-1 could promote PP4-dependent activ-ity and counteract the reactivation of PP2A. Although our invitro data suggest that PME-1 inhibition decreases cell prolif-eration and reduces cancer phenotypes (Figs. 1 and 2), weasked whether PME-1 inhibition had deleterious effects whenPP4 is overexpressed. We transiently transfected empty vectoror overexpressed PPP2CA, PPP4C, or PPP6C in ECC-1 cellsstably expressing scrambled (black bars) or PME-1 shRNA(white bars) and completed foci-forming assays to determinethe effects of PME-1 inhibition when various phosphatases

were overexpressed (Fig. 5D). We found that overexpression ofPPP2CA decreased foci formation by about 65% (þPPP2CA,black bar), suggesting that increased PP2A alone decreases cellproliferation. Depletion of PME-1 in this background leads to afurther significant decrease in foci (þPPP2CA, white bar).Interestingly, overexpression of PPP4C led to about 35%increase in cell proliferation (þPPP4C, black bar) when com-pared with control cells (Empty, black bar), substantiatingother reports that PP4 has a tumor-promoting role. Over-expression of PPP6C led to a significant 25% decrease in fociformation (þPPP6C, black bar), suggesting that PP6 may alsohave antitumor effects. Importantly, when PME-1was depletedin all cases, cell proliferation was significantly decreased, evenwhen PPP4C and PPP6C were overexpressed (þPPP4C andþPPP6C, white bars). These data suggest that inhibition ofPME-1 is sufficient to decrease cell proliferation regardless ofits role in PP4 and/or PP6 regulation.

Figure 4. PME-1–positive cellsundergo transition to E-cadherin–negative and P-cadherin–positivemesenchymal cells.Immunostainings of tissuesections from endometrial cancerpatient matched pairs comparinggrade 1 endometrial cancer(patient 06313) and grade 3endometrial cancer (patient 06294)using a �40 objective. A, tissueswere immunostained withantibodies specific for PME-1 or P-cadherin, a marker of aggressiveendometrial cancer, and wereimaged with fluorescentmicroscopy. B, tumor sampleswere immunostained withantibodies specific for PME-1 andE-cadherin. 40,6-Diamidino-2-phenylindole (DAPI) is a nuclearstain; merged images and whitearrows demonstrate theco-expression of PME-1 andP-cadherin (A) or PME-1 andE-cadherin (B).






PME-1 promotes tumor formation in an in vivo modelSince our data suggest that PME-1 promotes more aggres-

sive endometrial cancer, we asked whether the overexpres-sion of PME-1 in endometrial cancer cells promoted theformation of tumors in a xenograft model. Instead of RL95-2cells, which require a high number of cells to induce tumorformation (37), we used ECC-1 cells, which are aggressiveendometrial cancer cells with high levels of endogenousPME-1 (Fig. 1A) that have been used previously for similarstudies (38). ECC-1 cells expressing either the empty vector(Control) or overexpressing PME-1 (þPME-1) were subcu-taneously injected into the flank of 7 female nude mice per

group and tumor size was measured weekly. Before injec-tion, we completed Western blot analysis to confirm theproper expression of PME-1 (Fig. 6A). We found that miceinjected with þPME-1 cells formed tumors with a signifi-cantly larger tumor than control mice by 7 weeks postin-jection (Fig. 6B). Quantitative RT-PCR and Western blotanalysis of tumor tissue harvested at the end of the studyconfirmed that PME-1 levels were still increased in þPME-1cells versus control cells (data not shown). These in vivodata correlate well with our in vitro data and furthersubstantiate our hypothesis that PME-1 promotes cancerprogression.

Figure 5. PME-1 interacts withthe catalytic subunits of PP2A andPP4. A and B, ECC-1 cellsoverexpressing FLAG emptyvector, FLAG-PME-1, or FLAG-PME-1 S156A were used toprepare whole-cell extracts thatwere used to immunoprecipitatewith FLAG resin to pull-down PME-1. Western blot analysis wascompleted for inputs (10%) andimmunoprecipitation (100%), andblots were probed for endogenousPpp2ca (A) or Ppp4c and Ppp6c(B). C, HEK293T cells werecotransfected with V5-taggedempty vector, V5-PME-1, or V5-PME-1 S156A and FLAG-taggedPppp2ca or Ppp4c.Immunoprecipitation wascompleted with FLAG-resin andWestern blot analysis wascompleted to confirm that allplasmids were expressed (input)and to identify which catalyticsubunits PME-1 is capable ofbinding. D, ECC-1 cells stablyexpressing scrambled shRNA(black bars) or PPME1 shRNA(white bars) were transfected withempty vector or vectorsoverexpressing the catalyticsubunit for PP2A (PPP2CA), PP4(PPP4C), or PP6 (PPP6C) and wereplated at 3,000 cells per well. Fociwere grown for 10 days beforecrystal violet staining, andquantification and data representpercent foci formation comparedwith cells expressing scrambledshRNA and empty vector. Theexperiment was repeated sixtimes. Significance was calculatedby the standard Student t test;where �, P < 0.05; ��, P < 0.01;��, P < 0.001.

Wandzioch et al.





DiscussionPME-1 regulates PP2A activity (13, 14, 21) and we have

shown that increased PME-1 correlates with increased cancerphenotypes in endometrial cancer cell lines. Exogenous expres-sion of PME-1 led to increased cell proliferation (Fig. 1C and D)via increased activity of ERK (Fig. 2A; ref. 21) and Akt pathways(Fig. 2B; ref. 9), which promote EMT when constitutivelyactivated (26). Recent work has indicated upregulation of theAkt and ERK signaling pathways in endometrial cancersthrough several mechanisms (39–41), including the accumu-lation of mutations (40, 41). Increased PME-1 activity inendometrial cancer likely contributes to this phenomenon.PP2A dephosphorylates both MEK1/2 and ERK1/2 (42),

inhibiting the MAPK pathway and its downstream targets(reviewed in ref. 43). While the specific B subunits regulatingMEK1/2 phosphorylation status have not been identified, B/56b- and B/56g-dependent PP2A are known to dephosphory-late ERK (10). We demonstrated that overexpression of PME-1correlates with an increase in the activated and phosphory-lated ERK (Fig. 2A), thereby stabilizing the pathway. Previouswork suggests PME-1 regulates the association of B/55a sub-units with the PP2A core dimer (16, 17) and that reducedmethylation of PP2A results in decreased B/55a–PP2A forma-tion (44, 45). Our data suggest that PME-1 may affect theactivity of B/56-dependent PP2A in the regulation of ERK orthat B/55a-PP2A may also regulate ERK dephosphorylation.Sustained activation of the ERK pathway leads to increasedactivity of transcription factors, such as oncogenic c-Myc andElk1 transcription factors (reviewed in ref. 43), promoting cellproliferation.PP2A has a well-defined role in other signaling pathways.

Several isoforms of PP2A are capable of dephosphorylatingAkt. Kuo and colleagues identified B/55a-dependent PP2Aas a regulator of Akt signaling via dephosphorylation ofT308 in lymphoid and NIH3T3 cell lines (46). More recently,Rodgers and colleagues showed the B/56b subunit is activatedby Cdc2-like kinase 2 (Clk2), targeting the B/56b-PP2A holo-enzyme to Akt, dephosphorylating Akt at T308 and S473

(11). We found that overexpression of PME-1 led to a drama-tic increase in phosphorylation of Akt on T308 and S473(Fig. 2B) promoting cell proliferation (Fig. 1C and D) support-ing the recent findings (9). Activation of Akt correlates withdecreased E-cadherin expression and upregulation of EMT-promoting genes, such as Twist, Snail, and Slug (47–49). Wenote similar trends in decreased E-cadherin and increasedvimentin and noggin expression upon PME-1 overexpression(Pusey and Rice, unpublished data).

PME-1 regulation of PP2A activity varies among differentcell types and conditions. Previous work demonstrated thatPME-1 regulates ERK signaling, but not Akt signaling, inhuman gliomas (21), whereas recent findings suggest thatPME-1 overexpression stabilizes Akt phosphorylation atT308 independent of S473 in HEK-TERT cells (9) expressingshRNA against B/56g . We found that in endometrial cancercells, overexpression of PME-1 induces phosphorylation ofERK and Akt, indicating inhibition of PP2A holoenzymescontaining B/55a, among other B subunits. These data suggestthat PME-1 acts as a global inhibitor of PP2A and may nottarget specific heterotrimers.

Because of the high identity among catalytic subunits ofPP2A, PP4, and PP6, we askedwhether PME-1 can interact withother protein phosphatases. We identified a novel role forPME-1 in the regulation of PP4. Our data suggest a strongaffinity for PME-1 to associate with PP2A and a lesser affinityfor PP4, whereas PME-1 does not associate with PP6 (Fig. 5Band C). Interestingly, PP2A has tumor suppressor roles(4, 34, 50) whereas PP4 has tumor-promoting roles (35, 36),suggesting that PME-1 may counterbalance its own activitytoward PP2A by also inhibiting PP4 or that PME-1 inhibitsPP2A while activating PP4 to promote cell proliferation. Thesehypotheses require more investigation; however, we haveshown that inhibition of PME-1 in the context of PP2A, PP4,or PP6 overexpression decreases cell proliferation (Fig. 5D),substantiating our findings that inhibition of PME-1 in endo-metrial cancer cells is beneficial to decrease cancer pheno-types, regardless of PP4 and PP6 expression levels.

PME-1 is increased in tumor samples versus NAT in patientsamples (Fig. 3 and Table 1; Supplementary Table S2), indi-cating PME-1 as a potential diagnosticmarker for patientswithtype I endometrial cancer. On the basis of the analysis of 30patient samples, it is difficult to determine whether there is acorrelation with PME-1 levels and cancer grade; however, wenoted increased PME-1 mRNA and protein levels in endome-trial cancer tumors compared with NAT (Fig. 2A and C) andincreased PME-1 immunopositivity in 18 of 19 samples (�95%)tested by IHC, suggesting that PME-1 strongly correlates withdisease (Fig. 3D). A slight increase in PME-1 2þ staining wasobserved in grade 2 compared with grade 1 patient samples(57% vs. 44%); however, the acquisition of more samples isnecessary to confirm thesefindings. Loss of PME-1 staining in 1of 3 FIGO grade 3 samples may be due to dedifferentiation andloss of histologic and immunohistochemical characteristics;however, 2 of 3 grade 3 samples were positive for PME-1staining. Importantly, we correlated an increase in PME-1 andP-cadherin costaining and a concomitant loss of PME-1 and E-cadherin costaining in grade 3 samples, suggesting that PME-1

Contro

l

+PME-1

GAPDH

PME-1

Time (weeks)

Tum

or v

olum

e (m

m3 )

400

300

200

100

00 1 2 3 4 5 6 7 8

Control

+ PME-1

A B

Figure 6. Overexpression of PME-1 causes increased tumor growth in axenograft model. A total of 1� 106 ECC-1 cells expressing empty vector(control, n ¼ 7) or overexpressing PME-1 (þPME-1, n ¼ 7) in PBS wereinjected subcutaneously into the flank of nude female mice. A, Westernblot analysis of ECC-1 cells before injection. B, tumors were measured(length and width) weekly with a caliper to approximate volume over 7weeks. Averages are shown with SEM. Data were analyzed by 2-wayANOVA for significance, where �, P < 0.05; ��, P < 0.01; ��, P < 0.001.






may play a role in endometrial cancer aggressivity. Whilemore patient samples are required to fully investigate a rolefor PME-1 in promoting cancer progression, we demonstrat-ed that inhibition of PME-1 decreases cancer phenotypes(Fig. 1C–F), increases PP2A activity (Fig. 1G), decreases Akt/ERK signaling (Fig. 2), and is therefore an attractive diagnosticmarker and potential target for cancer drug therapy for type Iendometrioid adenocarcinoma.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: E. Wandzioch, M. Pusey, S. Bail, L.M. RiceDevelopment of methodology: E. Wandzioch, M. Pusey, S. Bail, L.M. RiceAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): E. Wandzioch, M. Pusey, A. Werda, S. Bail, A. BhaskarAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): E. Wandzioch, M. Pusey, A. Werda, S. Bail, J.-J. Yang,L.M. Rice

Writing, review, and/or revision of the manuscript: E. Wandzioch,M. Pusey, J.-J. Yang, L.M. RiceAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): M. Nestor, J.-J. YangStudy supervision: L.M. Rice

AcknowledgmentsThe authors thank Genesis Biotechnology Group, LLC, and Dr. Eli Mordechai

for the financial support of this work and for many helpful conversations. Theyalso thank Drs. Martin Adelson, Joseph T. Nickels, Jason Trama, Siu K. Lo, andMaria Webb for their scientific support and input throughout this conceptionand development of these studies. They express gratitude for the generous accessto the cryostat and microscope in the laboratory of Dr. Kenneth Zaret at theUniversity of Pennsylvania (Philadelphia, PA). They also thank Dr. Deni S. Galileoat the University of Delaware for the use of his microscope and his advisement.They also express gratitude to Caroline Giordano, Adam Bata, and PetronioZalmea for their help with this work.

The costs of publication of this article were defrayed in part by the pay-ment of page charges. This article must therefore be hereby marked advertise-ment in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received October 30, 2013; revised April 1, 2014; accepted April 24, 2014;published OnlineFirst June 13, 2014.

References1. Janssens V, Rebollo A. The role and therapeutic potential of Ser/Thr

phosphatase PP2A in apoptotic signalling networks in human cancercells. Curr Mol Med 2012;12:268–87.

2. Virshup DM, Shenolikar S. From promiscuity to precision: proteinphosphatases get a makeover. Mol Cell 2009;33:537–45.

3. Eichhorn PJ, Creyghton MP, Bernards R. Protein phosphatase 2Aregulatory subunits and cancer. BiochimBiophys Acta 2009;1795:1–15.

4. Westermarck J, Hahn WC. Multiple pathways regulated by the tumorsuppressor PP2A in transformation. TrendsMolMed 2008;14:152–60.

5. Shouse GP, Nobumori Y, Panowicz MJ, Liu X. ATM-mediated phos-phorylation activates the tumor-suppressive function of B56gamma-PP2A. Oncogene 2011;30:3755–65.

6. Nobumori Y, Shouse GP, Fan L, Liu X. HEAT repeat 1 motif is requiredfor B56gamma-containing protein phosphatase 2A (B56gamma-PP2A) holoenzyme assembly and tumor-suppressive function. J BiolChem 2012;287:11030–6.

7. Jayadeva G, Kurimchak A, Garriga J, Sotillo E, Davis AJ, Haines DS,et al. B55alpha PP2A holoenzymes modulate the phosphorylationstatus of the retinoblastoma-related protein p107 and its activation.J Biol Chem 2010;285:29863–73.

8. Fey D, Croucher DR, Kolch W, Kholodenko BN. Crosstalk and signal-ing switches in mitogen-activated protein kinase cascades. FrontPhysiol 2012;3:355.

9. Jackson JB, Pallas DC. Circumventing cellular control of PP2A bymethylation promotes transformation in an Akt-dependent manner.Neoplasia 2012;14:585–99.

10. Letourneux C, Rocher G, Porteu F. B56-containing PP2A dephos-phorylate ERK and their activity is controlled by the early gene IEX-1and ERK. EMBO J 2006;25:727–38.

11. Rodgers JT, Vogel RO, Puigserver P. Clk2 and B56beta mediateinsulin-regulated assembly of the PP2A phosphatase holoenzymecomplex on Akt. Mol Cell 2011;41:471–9.

12. Sablina AA, Hector M, Colpaert N, Hahn WC. Identification of PP2Acomplexes and pathways involved in cell transformation. Cancer Res2010;70:10474–84.

13. Ogris E, Du X, Nelson KC, Mak EK, Yu XX, Lane WS, et al. A proteinphosphatase methylesterase (PME-1) is one of several novel proteinsstably associating with two inactive mutants of protein phosphatase2A. J Biol Chem 1999;274:14382–91.

14. XingY, Li Z, ChenY, Stock JB, Jeffrey PD, Shi Y. Structuralmechanismof demethylation and inactivation of protein phosphatase 2A. Cell2008;133:154–63.

15. Lee J, Stock J. Protein phosphatase 2A catalytic subunit is methyl-esterified at its carboxyl terminus by a novel methyltransferase. J BiolChem 1993;268:19192–5.

16. Bryant JC,Westphal RS,Wadzinski BE.Methylated C-terminal leucineresidue of PP2A catalytic subunit is important for binding of regulatoryB alpha subunit. Biochem J 1999;339:241–6.

17. Yu XX, Du X, Moreno CS, Green RE, Ogris E, Feng Q, et al.Methylation of the protein phosphatase 2A catalytic subunit isessential for association of B alpha regulatory subunit but notSG2NA, striatin, or polyomavirus middle tumor antigen. Mol BiolCell 2001;12:185–99.

18. Lee J, Chen Y, Tolstykh T, Stock J. A specific protein carboxylmethylesterase that demethylates phosphoprotein phosphatase 2Ain bovine brain. Proc Natl Acad Sci U S A 1996;93:6043–7.

19. Longin S, Jordens J, Martens E, Stevens I, Janssens V, Rondelez E,et al. An inactive protein phosphatase 2A population is associatedwithmethylesterase and can be re-activated by the phosphotyrosyl phos-phatase activator. Biochem J 2004;380:111–9.

20. Longin S, Zwaenepoel K, Martens E, Louis JV, Rondelez E, Goris J,et al. Spatial control of protein phosphatase 2A (de)methylation. ExpCell Res 2008;314:68–81.

21. Puustinen P, Junttila MR, Vanhatupa S, Sablina AA, Hector ME,Teittinen K, et al. PME-1 protects extracellular signal-regulated kinasepathway activity from protein phosphatase 2A-mediated inactivationin human malignant glioma. Cancer Res 2009;69:2870–7.

22. Sontag JM, Nunbhakdi-Craig V, Mitterhuber M, Ogris E, Sontag E.Regulation of protein phosphatase 2A methylation by LCMT1 andPME-1 plays a critical role in differentiation of neuroblastoma cells.J Neurochem 2010;115:1455–65.

23. Dung TD, Day CH, Binh TV, Lin CH, Hsu HH, Su CC, et al. PP2Amediates diosmin p53 activation to block HA22T cell proliferation andtumor growth in xenografted nude mice through PI3K-Akt-MDM2signaling suppression. Food Chem Toxicol 2012;50:1802–10.

24. McCubrey JA, Steelman LS, Chappell WH, Abrams SL, Montalto G,Cervello M, et al. Mutations and deregulation of Ras/Raf/MEK/ERKand PI3K/PTEN/Akt/mTOR cascades. Oncotarget 2012;3:954–87.

25. Zhang Q, Claret FX. Phosphatases: the new brakes for cancer devel-opment? Enzyme Res 2012;2012:659649.

26. Boehm JS, Zhao JJ, Yao J, Kim SY, Firestein R, Dunn IF, et al.Integrative genomic approaches identify IKBKE as a breast canceroncogene. Cell 2007;129:1065–79.

27. KorchC, SpillmanMA, Jackson TA, JacobsenBM,MurphySK, LesseyBA, et al. DNA profiling analysis of endometrial and ovarian cell linesreveals misidentification, redundancy and contamination. GynecolOncol 2012;127:241–8.

28. Lee GY, Kenny PA, Lee EH, Bissell MJ. Three-dimensional culturemodels of normal and malignant breast epithelial cells. Nat Methods2007;4:359–65.

Wandzioch et al.





29. Rocher G, Letourneux C, Lenormand P, Porteu F. Inhibition of B56-containing protein phosphatase 2As by the early response gene IEX-1leads to control of Akt activity. J Biol Chem 2007;282:5468–77.

30. Van KaneganMJ, Adams DG,Wadzinski BE, Strack S. Distinct proteinphosphatase 2A heterotrimers modulate growth factor signaling toextracellular signal-regulated kinases and Akt. J Biol Chem2005;280:36029–36.

31. Engelsen IB, Akslen LA, Salvesen HB. Biologic markers in endometrialcancer treatment. APMIS 2009;117:693–707.

32. Ortega-Gutierrez S, Leung D, Ficarro S, Peters EC, Cravatt BF. Tar-geteddisruptionof thePME-1genecauses loss of demethylatedPP2Aand perinatal lethality in mice. PLoS ONE 2008;3:e2486.

33. Arnold HK, Sears RC. A tumor suppressor role for PP2A-B56alphathrough negative regulation of c-Myc and other key oncoproteins.Cancer Metastasis Rev 2008;27:147–58.

34. Mumby M. PP2A: unveiling a reluctant tumor suppressor. Cell2007;130:21–4.

35. Wang B, Zhao A, Sun L, Zhong X, Zhong J, Wang H, et al. Proteinphosphatase PP4 is overexpressed in human breast and lung tumors.Cell Res 2008;18:974–7.

36. Weng S, Wang H, Chen W, Katz MH, Chatterjee D, Lee JE, et al.Overexpression of protein phosphatase 4 correlates with poor prog-nosis in patients with stage II pancreatic ductal adenocarcinoma.Cancer Epidemiol Biomarkers Prev 2012;21:1336–43.

37. Pandey V, Qian PX, Kang J, Perry JK, Mitchell MD, Yin Z, et al. Arteminstimulates oncogenicity and invasiveness of human endometrial car-cinoma cells. Endocrinology 2010;151:909–20.

38. Osipo C, Meeke K, Liu H, Cheng D, Lim S, Weichel A, et al. Trastu-zumab therapy for tamoxifen-stimulated endometrial cancer. CancerRes 2005;65:8504–13.

39. Bradford LS, Rauh-Hain A, Clark RM,Groeneweg JW, Zhang L, BorgerD, et al. Assessing the efficacy of targeting the phosphatidylinositol 3-kinase/AKT/mTOR signaling pathway in endometrial cancer. GynecolOncol 2014;133:346–52.

40. Pavlidou A, Vlahos NF. Molecular alterations of PI3K/Akt/mTOR path-way: a therapeutic target in endometrial cancer. ScientificWorldJour-nal 2014;2014:709736.

41. Wang LE,MaH, Hale KS, YinM,Meyer LA, LiuH, et al. Roles of geneticvariants in the PI3K and RAS/RAF pathways in susceptibility toendometrial cancer and clinical outcomes. J Cancer Res Clin Oncol2012;138:377–85.

42. Sontag E, Fedorov S, Kamibayashi C, Robbins D, CobbM, MumbyM.The interaction of SV40 small tumor antigen with protein phosphatase2A stimulates the map kinase pathway and induces cell proliferation.Cell 1993;75:887–97.

43. Junttila MR, Li SP, Westermarck J. Phosphatase-mediated crosstalkbetween MAPK signaling pathways in the regulation of cell survival.FASEB J 2008;22:954–65.

44. Lee JA, Pallas DC. Leucine carboxyl methyltransferase-1 is necessaryfor normal progression through mitosis in mammalian cells. J BiolChem 2007;282:30974–84.

45. Sontag JM, Nunbhakdi-Craig V, Montgomery L, Arning E, BottiglieriT, Sontag E. Folate deficiency induces in vitro and mouse brainregion-specific downregulation of leucine carboxyl methyltransfer-ase-1 and protein phosphatase 2A B(alpha) subunit expression thatcorrelate with enhanced tau phosphorylation. J Neurosci 2008;28:11477–87.

46. Kuo YC, Huang KY, Yang CH, Yang YS, Lee WY, Chiang CW.Regulation of phosphorylation of Thr-308 of Akt, cell proliferation, andsurvival by the B55alpha regulatory subunit targeting of the proteinphosphatase 2A holoenzyme to Akt. J Biol Chem 2008;283:1882–92.

47. Fenouille N, Tichet M, Dufies M, Pottier A, Mogha A, Soo JK, et al. Theepithelial-mesenchymal transition (EMT) regulatory factor SLUG(SNAI2) is a downstream target of SPARC and AKT in promotingmelanoma cell invasion. PLoS ONE 2012;7:e40378.

48. Grille SJ, Bellacosa A, Upson J, Klein-Szanto AJ, vanRoy F, Lee-KwonW, et al. The protein kinase Akt induces epithelial mesenchymaltransition and promotes enhanced motility and invasiveness of squa-mous cell carcinoma lines. Cancer Res 2013;63:2172–8.

49. Larue L, Bellacosa A. Epithelial-mesenchymal transition in develop-ment and cancer: role of phosphatidylinositol 30 kinase/AKT pathways.Oncogene 2005;24:7443–54.

50. Sablina AA, Hahn WC. The role of PP2A A subunits in tumor suppres-sion. Cell Adh Migr 2007;1:140–1.






2014;74:4295-4305. Published OnlineFirst June 13, 2014.Cancer Res Ewa Wandzioch, Michelle Pusey, Amy Werda, et al. Malignant Phenotype of Endometrial Cancer CellsPME-1 Modulates Protein Phosphatase 2A Activity to Promote the

Updated version

10.1158/0008-5472.CAN-13-3130doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC2

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC1Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/16/4295.full#ref-list-1

This article cites 50 articles, 18 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/16/4295.full#related-urls

This article has been cited by 7 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/74/16/4295To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-3130

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC1 http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC2

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC1 http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3130.DC2

http://cancerres.aacrjournals.org/content/74/16/4295.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/16/4295.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/16/4295


PME-1 Modulates Protein Phosphatase 2A Activity to...

Documents

Transcript of PME-1 Modulates Protein Phosphatase 2A Activity to...