Palbociclib Promotes Dephosphorylation of NPM/B23 at Threonine 199 and Inhibits Endometrial Cancer Cell Growth

cancers

Article Palbociclib Promotes Dephosphorylation of NPM/B23 at Threonine 199 and Inhibits Endometrial Cancer Cell Growth

1,2, 3, 1,2,4 5 4 Chiao-Yun Lin †, Li-Yu Lee †, Tzu-Hao Wang , Cheng-Lung Hsu , Chia-Lung Tsai , Angel Chao 1,2,* and Chyong-Huey Lai 1,2,*

1 Department of Obstetrics and Gynecology, Chang Gung Memorial Hospital Linkou Medical Center and Chang Gung University College of Medicine, Taoyuan 333, Taiwan 2 Gynecologic Cancer Research Center, Chang Gung Memorial Hospital, Taoyuan 333, Taiwan 3 Department of Pathology, Chang Gung Memorial Hospital Linkou Medical Center, Taoyuan 333, Taiwan 4 Genomic Medicine Research Core Laboratory, Chang Gung Memorial Hospital, Taoyuan 333, Taiwan 5 Department of Medical Oncology, Chang Gung Memorial Hospital Linkou Medical Center, Taoyuan 333, Taiwan * Correspondence: [email protected] (A.C.); [email protected] (C.-H.L.) These authors contributed equally to this work. † Received: 2 July 2019; Accepted: 18 July 2019; Published: 20 July 2019

Abstract: Endometrial cancer incidence rates are growing, especially in countries with rapid socioeconomic transitions. Despite recent advances in chemotherapy, hormone therapy, and targeted therapy, advanced/recurrent disease remains a clinical challenge. Palbociclib—a selective inhibitor of cyclin-dependent kinases (CDK) 4/6—has therapeutic potential against estrogen receptor (ER)-positive and HER2-negative breast cancer. However, the question as to whether it can be clinically useful in endometrial cancer remains open. Here, we show that combined treatment with palbociclib and megesterol acetate exerts synergistic antiproliferative eﬀects against endometrial cancer cells. Treatment of cancer cells with palbociclib suppressed NPM/B23 phosphorylation at threonine 199 (Thr199). We further demonstrated that CDK6 acts as a NPM/B23 kinase. Palbociclib-induced NPM/B23 dephosphorylation sensitized endometrial cancer cells to megesterol acetate through the upregulation of ERα expression. Immunohistochemistry revealed an overexpression of phospho-NPM/B23 (Thr199) in human endometrial cancer, and phospho-NPM/B23 (Thr199) expression levels were inversely associated with Erα in clinical specimen. In a xenograft tumor model, the combination of palbociclib and megesterol acetate successfully inhibited tumor growth. Taken together, our data indicate that palbociclib promoted NPM/B23 dephosphorylation at Thr199—an eﬀect mediated by disruption of CDK6 kinase activity. We conclude that palbociclib holds promise for the treatment of endometrial cancer when used in combination with megesterol acetate.

Keywords: endometrial cancer; palbociclib; megesterol acetate; combination treatment; NPM/B23; Thr199

1. Introduction Endometrial cancer is the sixth most common gynecologic cancer worldwide and its incidence rate is growing—especially in countries with rapid socioeconomic transitions [1]. Between 2014 and 2015, Taiwan has witnessed a 7.73% increase in the incidence of this malignancy, with 20.71 new cases per 100,000 persons/year [2]. Although hysterectomy remains the mainstay of deﬁnitive therapy [3], adjuvant chemo- and/or radio-therapy may confer a survival beneﬁt in patients who are at high risk for recurrences or presenting with advanced disease. Unfortunately, the clinical management of

Cancers 2019, 11, 1025; doi:10.3390/cancers11071025 www.mdpi.com/journal/cancers Cancers 2019, 11, 1025 2 of 16 advanced/recurrent disease is still challenging and the role of chemotherapy, hormone therapy, and targeted therapy remains under scrutiny. Early endometrial cancers generally express estrogen and progesterone receptors [4]. Patients with progesterone receptor (PR)-positive as well as estrogen receptor (ER)-positive responded to medroxyprogesterone acetate, a progestin [5]. Furthermore, the single best predictor of response was ER, which was significantly associated with clinical response and survival [6]. A reduced expression of ER has been associated with poor differentiation, advanced-stage tumors, disease recurrence, and adverse outcomes [7]. Meanwhile, the disappearance of hormonal receptor expression is common in patients with recurrent estrogen-associated cancers [8]. It is critical to find new therapeutic targets that can restore functional hormones through either PR or ER, which sensitize to hormone therapy. During early mitosis, the nucleolus starts to disassemble—an event accompanied by phosphorylation of cyclin-dependent kinase (CDK) [9]. NPM/B23 (nucleophosmin, NPM, B23)—one of the most abundant nucleolar proteins—undergoes CDK-mediated phosphorylation [9]. During cell cycle, NPM/B23 can be phosphorylated at different sites [10]. Phosphorylation at the casein kinase 2 (CK2) site (Ser125) regulates NPM/B23 function in the nucleolus [11], whereas phosphorylation at CDK1 sites, including threonine (Thr) 199, 219, 234, and 237, promotes dissociation of NPM/B23 from the nucleolus [12]. NPM/B23 dissociation from the centrosome at the beginning of cell division is mediated by CDK2/cyclin E phosphorylation at Thr199 [13]. However, NPM/B23 hyperphosphorylation at Thr199 deregulates both CDK2 and CDK4, ultimately resulting in an abnormal centrosome duplication and the disruption of genome integrity [14]. Interestingly,the v-cyclin and cellular CDK6 kinase of Kaposi’s sarcoma herpesvirus is capable of phosphorylating NPM/B23 on Thr199, with phosphorylated NPM/B23 being detectable in primary Kaposi’s sarcoma [15]. Moreover, CDK1/cyclin B can induce NPM/B23 phosphorylation at Thr234 and Thr237, which promotes its re-association to the centrosome during mitosis [13]. An altered cell proliferation and the capability to escape immune-mediated clearance are common features of malignant cells [16]. At the molecular level, dysregulation of the cyclin D-CDK 4/6-inhibitor of the CDK4 (INK4)-retinoblastoma (Rb) pathway is frequently observed in cancer [17]. The cyclin D-CDK4/6-INK4-Rb pathway is the master regulator of the G1-S transition during the cell cycle. Inhibition of CDK4/6 exerts multiple biological effects, including (1) decreased levels of phosphorylated Rb, (2) increased formation of Rb-E2F complexes, and (3) inhibition of E2F transcription factor activity. CDK4/6 inhibitors (i.e., palbociclib, abemaciclib, and ribociclib) have been granted approval from the the U.S. Food and Drug Administration for the treatment of patients with previously untreated estrogen receptor (ER)-positive advanced breast malignancies [18–22]. Notably,palbociclib has shown promising therapeutic activity against endometrial cancer cells [23] and PTEN-deficient endometrial malignant cells [24]. We have previously shown that estrogens may stabilize NPM/B23 [25]. We have also demonstrated that NPM/B23 silencing via AP2γ (activator protein-2γ, TFAP2C) may activate ERα expression in endometrial cancer cells [26]. Knockdown of AP2γ in an animal model sensitized endometrial cancer cells to megesterol acetate through an upregulation of ERα expression. Unfortunately, NPM/B23 or AP2γ inhibitors are not currently available for clinical use. In this context, we reasoned that palbociclib, a CDK4/6 inhibitor that is capable of regulating the cell cycle machinery [27], could modulate ERα expression in endometrial cancer cells by affecting the NPM/B23 phosphorylation status. Here, we show that combined treatment with palbociclib and megesterol acetate exerts synergistic antiproliferative effects against endometrial cancer cells. Exposure of cancer cells to palbociclib suppressed NPM/B23 phosphorylation at Thr199. We further demonstrated that CDK6 acts as a NPM/B23 kinase. Palbociclib-induced NPM/B23 dephosphorylation sensitized endometrial cancer cells to megesterol acetate through upregulation of ERα expression.

2. Results

2.1. Exposure to Palbociclib Induced ERα Expression in Endometrial Cancer Cells Because palbociclib is a selective inhibitor of cyclin-dependent kinases CDK4 and CDK6 and NPM/B23 can be phosphorylated either by CDK4 [14] or CDK6 [15], we tested whether this drug was Cancers 2019, 11, 1025 3 of 16 able to inhibit NPM/B23 phosphorylation at several sites. Upon treatment of ERα-negative (ARK2, high-passage Ishikawa cells [26]) and ERα-positive HEC1B cells with palbociclib, cell lysates were probed with phospho specific NPM/B23 antibodies. Notably, palbociclib enhanced ERα expression and promoted NPM/B23 dephosphorylation at threonine 199, 234, and 237, whereas NPM/B23 dephosphorylation at serine 125 and PR expression was not significantly altered (Figure1 and Supplementary Figures S1 and S2). Exposure to palbociclib also induced the expression of several ERα-regulated genes, including cathepsin D, EBAG9, and TFF1/pS2 (Figure1c). Taken together, these findings indicate that palbociclib is able to specifically inhibit NPM/B23 phosphorylation at Thr199,

234, and 237. Cancers 2019, 11, x 4 of 17

Figure 1. PalbociclibFigure 1. Palbociclib induces induces ERα expression ERα expression and and promotes promotes NPM/B NPM23/B23 dephosphorylation dephosphorylation at multiple at multiple sites in endometrialsites in endometrial cancer cells. cancer (a, bcells.) Palbociclib (a,b) Palbociclib (2 µ M)(2 µM) was was used used to to treat treat ARK2ARK2 (left (left panel) panel) and and HEC1B cells (right panel)HEC1B for cells 24 (right h. Cell panel) lysates for 24 wereh. Cell subsequentlylysates were subsequently resolved resolved on SDS-PAGE on SDS-PAGE and and subjected to subjected to immunoblotting with antibodies raised against ERα, phospho-NPM/B23 (Ser125), immunoblottingphospho-NPM/B23 with antibodies (Thr199), raised phospho-NPM/B23 against ERα, phospho-NPM (Thr234/237), /B23NPM/B23, (Ser125), and phospho-NPMβ-actin. /B23 (Thr199), phospho-NPMDensitometry-derived/B23 (Thr234values (bottom)/237), are NPM normaliz/B23,ed with and theβ-actin. control (-) Densitometry-derived that was set as 1. Data values (bottom) areshown normalized are derived with from the three control independent (-) that experiments. was set β as-actin 1. serves Data as shown the loading are derivedcontrol for from three normalization. (c) Palbociclib-treated ARK2 cells were collected and mRNA expression levels for β independent experiments.ERα, cathepsin D, EBAG9,-actin servesand TFF1/pS2 as the were loading analyzed control using real-time for normalization. qPCR (primers ( cdescribed) Palbociclib-treated in ARK2 cells werethe Methods collected section). and mRNAData are expressionexpressed as levelsmeans ± for standard ERα, cathepsinerrors from three D, EBAG9, independent and TFF1/pS2 were analyzedexperiments. using real-time * p < 0.05qPCR compared (primers with controls. described in the Methods section). Data are expressed as means standard errors from three independent experiments. * p < 0.05 compared with controls. ± Cancers 2019, 11, 1025 4 of 16 Cancers 2019, 11, x 5 of 17

2.2.2.2. Palbociclib and Megestrol AcetateAcetate Synergistically InhibitInhibit Survival,Survival, IncreaseIncrease Apoptosis,Apoptosis, andand IncreaseIncrease thethe ExpressionExpression ofof ERERαα inin EndometrialEndometrial CancerCancer CellsCells BecauseBecause restoration ofof ERαα expression rendersrenders endometrialendometrial cancercancer cellscells susceptible to megestrol acetateacetate treatmenttreatment [[26],26], wewe investigatedinvestigated whetherwhether combinedcombined treatment withwith palbociclibpalbociclib andand megestrolmegestrol acetateacetate couldcould exertexert synergisticsynergistic antiproliferativeantiproliferative eeffects.ffects. Chou-Talalay plots basedbased on the MTTMTT assayassay revealedrevealed thatthat palbociclibpalbociclib andand megestrolmegestrol acetateacetate synergisticallysynergistically inhibited cell viability (Figure 22a,b;a,b; SupplementarySupplementary Figure S3a,b). The combination of palbociclib and megestrol acetate also showedshowed aa synergisticsynergistic eeffectffect onon colonycolony formationformation (Figure(Figure2 2cc and and Supplementary Supplementary FigureFigure S3c)S3c) and and induction induction ofof apoptosis—asapoptosis—as reflectedreflected byby higherhigher levelslevels ofof cleavedcleaved PARPPARP (Figure(Figure2 2d).d). The The results results of of tumor tumor xenograft xenograft experimentsexperiments confirmedconfirmed that palbociclibpalbociclib and megestrolmegestrol acetateacetate usedused inin combinationcombination inhibitedinhibited tumortumor growthgrowth toto aa greatergreater extentextent thanthan eithereither drugdrug alonealone (Figure(Figure2 2ee and and Supplementary Supplementary Figure Figure S3d). S3d). AtAt thethe mechanisticmechanistic level, palbociclibpalbociclib was found to specificallyspecifically inhibitinhibit phospho-NPMphospho-NPM/B23/B23 (Thr199)(Thr199) andand waswas ableable toto upregulateupregulate ERERαα expressionexpression (Figure(Figure2 f).2f). TheThe combinationcombination ofof palbociclibpalbociclib andandletrozole letrozole also also showedshowed synergisticsynergistic antiproliferativeantiproliferative eeffectsffects againstagainst tumortumor cellcell growthgrowth(Supplementary (Supplementary Figure Figure S4). S4).

Figure 2. Cont. Figure 2. Cont. Cancers 2019, 11, 1025 5 of 16 Cancers 2019, 11, x 6 of 17

Figure 2. Palbociclib renders endometrial cancer cells susceptiblesusceptible to megestrol acetate by inducing ERα expression. ( (aa,,b)) ARK2 ARK2 cells cells were were treated with vehicle (-) or different different doses of palbociclib alone (0, 2.5, 5, 5, 10, 10, and and 20 20 µM),µM), megestrol megestrol acetate acetate alone alone (0, (0,2.5, 2.5, 5, 10, 5, 10,and and 20 µM), 20 µ M),or their or their combination combination for 72 for h. 72 h. Cell survival was assessed using the MTT assay. Data are expressed as fold change SD relative Cell survival was assessed using the MTT assay. Data are expressed as fold change ± SD± relative to tovehicle-treated vehicle-treated cells. cells. All experiments All experiments were were performe performedd in triplicate in triplicate (left panel). (left panel). The synergistic The synergistic effect eofff ectpalbociclib of palbociclib and megestrol and megestrol acetate acetate was analyzed was analyzed with withthe CompuSyn the CompuSyn software software (right (right panel). panel). (c) (ARK2c) ARK2 cells cells were were treated treated with with vehicle vehicle (-), (-),different different doses doses of palbociclib of palbociclib alone alone (0, (0,1.25, 1.25, 2.5, 2.5, 5, 10, 5, 10,and and 20 µ µ 20µM),M), megestrol megestrol acetate acetate alone alone (0, (0,1.25, 1.25, 2.5, 2.5, 5, 5,10, 10, and and 20 20 µM),M), or or a a combination (palbociclib plus megestrol acetate) for 5 days. Colony formatio formationn was analyzed with the clonogenic assay. (d) ARK2 µ µ cells werewere treatedtreated withwith vehicle vehicle (-), (-), palbociclib palbociclib (0, (0, 5, 5, and and 10 10M), µM), megestrol megestrol acetate acetate (0, 5,(0, and 5, and 10 10M), µM), or a combinationor a combination (palbociclib (palbociclib plus plus megestrol megestrol acetate) acetat fore) 24 for h. 24 Cleaved h. Cleaved PARP PARP protein protein levels levels (as an (as index an of apoptosis) were analyzed by western blotting. Values measured on densitometry (shown at the index of apoptosis) were analyzed by western blotting. Values measured on densitometry (shown at bottom) were normalized with that observed in vehicle-treated cells (set at 1). All experiments were the bottom) were normalized with that observed in vehicle-treated cells (set at 1). All experiments performed in triplicate, and β-actin served as the loading control for normalization. (e) ARK2 were were performed in triplicate, and β-actin served as the loading control for normalization. (e) ARK2 inoculated into nude mice, inhibitory effect of palbociclib plus megestrol acetate on cancer growth were inoculated into nude mice, inhibitory effect of palbociclib plus megestrol acetate on cancer in a xenograft tumor model. (f) Extracts from tumors exposed to palbociclib, megestrol acetate, or a growth in a xenograft tumor model. (f) Extracts from tumors exposed to palbociclib, megestrol combination were immunoblotted with antibodies raised against ERα, phospho-NPM/B23 (Thr199), acetate, or a combination were immunoblotted with antibodies raised against ERα, and β-actin. Densitometry-derived values (bottom) are normalized with the control that was set as 1. phospho-NPM/B23 (Thr199), and β-actin. Densitometry-derived values (bottom) are normalized Data shown are derived from three independent experiments. β-actin serves as the loading control with the control that was set as 1. Data shown are derived from three independent experiments. for normalization. β-actin serves as the loading control for normalization. 2.3. Phospho-NPM/B23 (Thr199) Is Involved in Endometrial Tumorigenesis 2.3. Phospho-NPM/B23 (Thr199) Is Involved in Endometrial Tumorigenesis To further investigate the role of phospho-NPM/B23 (Thr199) and phospho-NPM/B23 (Thr234/237) in endometrialTo further cancer, investigate we examined the role their of immunohistochemical phospho-NPM/B23 expression(Thr199) inand pathological phospho-NPM/B23 specimens obtained(Thr234/237) from in premenopausal endometrial cancer, women. we Phospho-NPM examined their/B23 (Thr199)immunohistochemical histoscores were expression significantly in higherpathological in endometrial specimens cancer obtained than infrom normal prem endometrialenopausal tissuewomen. (p =Phospho-NPM/B230.020), whereas di ff(Thr199)erences werehistoscores less obvious were significantly for phospho-NPM higher in/B23 endometrial (Thr234/237) cancer (Figure than3 ain and normal Supplementary endometrial Figure tissue S5).(P = Representative0.020), whereas images differences of phospho-NPM were less obvious/B23 (Thr199) for phospho-NPM/B23 expression in malignant (Thr234/237) and normal (Figure endometrial 3a and tissueSupplementary are provided Figure in Figure S5). 3Representativeb. images of phospho-NPM/B23 (Thr199) expression in malignantTo shed and more normal light endometrial on the impact tissue of NPM are provided/B23 dephosphorylation in Figure 3b. and phosphorylation on ERα expression,To shed HEC1B more cellslight were on the transiently impact of transfected NPM/B2 with3 dephosphorylation wild-type NPM/B23 and (WT), phosphorylation a dephospho-mimetic on ERα NPMexpression,/B23 mutant HEC1B (Thr199A), cells orwere a phospho-mimetic transiently transfected NPM/B23 mutantwith wild-type (Thr199D). OverexpressionNPM/B23 (WT), of thea dephospho-mimetic NPMNPM/B23/B23 mutant mutant (Thr199A) (Thr199A) promoted, or ERa αphospho-mimeexpression at bothtic mRNANPM/B23 and mutant protein levels(Thr199D). (Figure Overexpression3c,d). Moreover, of phospho-NPMthe dephospho-mimetic/B23 (Thr199) NPM/B23histoscores mutant were (Thr199A) significantly promoted higher ER inα ERexpressionα negative at endometrialboth mRNA cancerand protein specimens levels than (Figure in ER 3c,d).α positive Moreover, patients phospho-NPM/B23 (p = 0.026; Supplementary (Thr199) Figurehistoscores S6). Conversely,were significantly overexpression higher in of theERα phospho-mimetic negative endometrial NPM/ B23cancer mutant specimens (Thr199D) than promoted in ERα bothpositive cell proliferationpatients (p and= 0.026; colony Supplementary formation (Figure Figure3e,f). S6). All ofConversely, the subsequent overexpression experiments of were the phospho-mimetic NPM/B23 mutant (Thr199D) promoted both cell proliferation and colony formation (Figure 3e,f). All of the subsequent experiments were therefore focused on Cancers 2019, 11, 1025 6 of 16 Cancers 2019, 11, x 7 of 17 thereforephospho-NPM/B23 focused on phospho-NPM (Thr199). These/ B23results (Thr199). confirmed These thatresults phospho-NPM/B23 confirmed that(Thr199) phospho-NPM expression /B23 (Thr199)levels expression were inversely levels associated were inversely with ERα associated in clinical specimen. with ERα in clinical specimen.

FigureFigure 3. Role 3. Role of phospho-NPMof phospho-NPM/B23/B23 (Thr199)(Thr199) expressionexpression in in endometrial endometrial tumorigenesis. tumorigenesis. (a) The (a) The immunohistochemicalimmunohistochemical expression expression ofof phospho-NPMphospho-NPM/B23/B23 (Thr199) (Thr199) was was analyzed analyzed in both in bothnormal normal endometrialendometrial tissues tissues obtained obtained from from patients patients whowho underwent hyst hysterectomyerectomy for forbenign benign gynecologic gynecologic conditionsconditions (n = (n27) = 27) and and endometrial endometrial cancer cancer ((nn == 37; Supplementary Supplementary Table Table S1). S1). (b) (Representativeb) Representative immunohistochemicalimmunohistochemical staining staining of of phospho-NPM phospho-NPM/B2/B233 (Thr199) (Thr199) in inbenign benign endometrial endometrial tissue tissue (upper (upper panel)panel) and and endometrial endometrial cancer cancer (EMCA;(EMCA; lower pane panel).l). Magnification: Magniﬁcation: 100×; 100scale ;bar: scale 50 bar:µm. ( 50c) µm. × (c) ERERα-positiveα-positive HEC1B HEC1B cellscells were transfected transfected with with NPM/B23 NPM/B23 (WT), (WT), the thedephospho-mimetic dephospho-mimetic NPM/B23 NPM /B23 mutantmutant (Thr199A), (Thr199A),or or the the phospho-mimetic phospho-mimetic NPMNPM//B23B23 mutant mutant (Thr199D) (Thr199D) for for24 h. 24 Cell h. Celllysates lysates were were resolved on SDS-PAGE and subjected to immunoblotting with antibodies raised against ERα and β-actin. Densitometry-derived values (bottom) are normalized with the NPM/B23 (WT), which was set as 1. Data shown are derived from three independent experiments. β-actin serves as the loading control for CancersCancers2019, 201911, 1025, 11, x 8 of 717 of 16

resolved on SDS-PAGE and subjected to immunoblotting with antibodies raised against ERα and β-actin. Densitometry-derived values (bottom) are normalized with the NPM/B23 (WT), which was set as 1. Data shown are derived from three independent experiments. β-actin serves as the loading normalization. (d) HEC1B cells were transfected with NPM/B23 (WT), the dephospho-mimetic NPM/B23 control for normalization. (d) HEC1B cells were transfected with NPM/B23 (WT), the mutant (Thr199A), or the phospho-mimetic NPM/B23 mutant (Thr199D) for 24 h. RT-qPCR was used dephospho-mimetic NPM/B23 mutant (Thr199A), or the phospho-mimetic NPM/B23 mutant to analyze(Thr199D) mRNA for expression24 h. RT-qPCR levels was using used the to reported analyze primers.mRNA expression * p < 0.05 versuslevels using control. the Experimentsreported were performed in triplicate and data are expressed as means standard errors of the mean. (e,f) ARK2 primers. * p < 0.05 versus control. Experiments were performed± in triplicate and data are expressed as cellsmeans were transfected ± standard witherrors the of NPM the mean./B23 (WT), (e,f) ARK2 dephospho-mimetic cells were transfected NPM/B23 with mutant the NPM/B23 (Thr199A), (WT), or the phospho-mimeticdephospho-mimetic NPM/ B23NPM/B23 mutant mutant (Thr199D) (Thr199A), for 24 h. or For the cell phospho-mimetic proliferation assay, NPM/B23 cells were mutant seeded at a density(Thr199D) of 10 4forcells 24 h. per For 96 cell well proliferation in complete assa mediumy, cells were for 24 seeded h. The at cells a density were of grown 104 cells for per an additional96 well 24 hin after complete the addition medium of for BrdU. 24 h. ForThe colonycells were formation, grown for the an cellsadditional were 24 seeded h after at the a densityaddition of of 5000BrdU. cells per 6For wells colony in complete formation, medium the cells forwere 6 days.seeded Data at a aredensity expressed of 5000 ascells means per 6 wellsstandard in complete errors medium from three ± independentfor 6 days. experiments. Data are expressed * p < 0.05 as means versus ± standard control. errors from three independent experiments. * p < 0.05 versus control. 2.4. CDK6-Mediated Phosphorylation of NPM/B23 Promotes ERα Expression 2.4.We CDK6-Mediated next sought to Phosphoryla identify thetion kinase of NPM/B23 that specifically Promotes ERα mediated Expression palbociclib-induced NPM/B23 dephosphorylation.We next sought Because to identify palbociclib the kinase is capable that specifically of inhibiting mediated both CDK4 palbociclib-induced and CDK6, ablation NPM/B23 of either enzymedephosphorylation. was achieved through Because specificpalbociclib siRNAs. is capable ARK2 of andinhibiting HEC1B both cells CDK4 in which and CDK6, expression ablation of eitherof CDK4either or CDK6 enzyme was was silenced achieved were through characterized specific siRN by reducedAs. ARK2 phospho-NPM and HEC1B cells/B23 in (Thr199)which expression and increased of ERαeitherexpression CDK4 (Figureor CDK64 ).was To silenced further were investigate characterized whether by CDK4reduced or phospho-NPM/B23 CDK6 can directly (Thr199) interact and with NPMincreased/B23, immunoprecipitation ERα expression (Figure experiments 4). To further were performed. investigate Thewhether results CDK4 revealed or CDK6 that CDK6can directly was able interact with NPM/B23, immunoprecipitation experiments were performed. The results revealed to interact with NPM/B23 (Figure5a), whereas CDK4 was not (Figure5b). We therefore implemented that CDK6 was able to interact with NPM/B23 (Figure 5a), whereas CDK4 was not (Figure 5b). We an in vitro kinase assay to assess whether CDK6 is actually capable of phosphorylating NPM/B23. therefore implemented an in vitro kinase assay to assess whether CDK6 is actually capable of By monitoringphosphorylating phosphorylation NPM/B23. By levels monitoring of endogenous phosphorylation NPM/B23 levels in immunoprecipitates,of endogenous NPM/B23 we foundin that purifiedimmunoprecipitates, recombinant we CDK6 found/cyclin that D1 purified was able recombinant to add phosphate CDK6/cyclin groups D1 at thewas Thr199 able to residue add of NPMphosphate/B23. Notably, groups this at specificthe Thr199 kinase residue activity of NPM/B2 was abrogated3. Notably, upon this palbociclib specific kinase treatment activity (Figure was5 c). Takenabrogated together, upon these palbociclib results indicate treatment that (Figure (1) palbociclib 5c). Taken is together, capable ofthese inducing results dephosphorylationindicate that (1) of NPMpalbociclib/B23 at is the capable Thr199 of inducing site and dephosphorylation (2) CDK6 is involved of NPM/B23 in the at phosphorylation the Thr199 site and of (2) NPM CDK6/B23 is in endometrialinvolved cancer.in the phosphorylation of NPM/B23 in endometrial cancer.

FigureFigure 4. CDK6-mediated 4. CDK6-mediated phosphorylation phosphorylation ofof NPMNPM/B23/B23 promotes ER ERαα expression.expression. (a,b (a) ,ARK2b) ARK2 or or HEC1BHEC1B cells cells were were transiently transiently transfected transfected with with control control siRNA siRNA (si-C), (si-C), CDK4 CDK4 siRNA siRNA (si-CDK4), (si-CDK4), or CDK6or CDK6 siRNA (si-CDK6) for 72 h. Cell lysates were subjected to western blot using antibodies raised siRNA (si-CDK6) for 72 h. Cell lysates were subjected to western blot using antibodies raised against ERα, CDK4, CDK6, phospho-NPM/B23 (Thr199), NPM/B23, and β-actin. Densitometry-derived values (bottom) are normalized with the control (si-C), which was set as 1. Data shown are derived from three independent experiments. β-actin serves as the loading control for normalization. Cancers 2019, 11, x 9 of 17

against ERα, CDK4, CDK6, phospho-NPM/B23 (Thr199), NPM/B23, and β-actin. Densitometry-derived values (bottom) are normalized with the control (si-C), which was set as 1. Data shown are derived from three independent experiments. β-actin serves as the loading control Cancers 2019, 11, 1025 8 of 16 for normalization.

Figure 5. CDK6-mediated NPM/B23 hyperphosphorylation. (a) ARK2 whole-cell lysates were Figure 5. CDK6-mediated NPM/B23 hyperphosphorylation. (a) ARK2 whole-cell lysates were immunoprecipitated (IP) with an anti-CDK6 antibody (α-CDK6) and subsequently analyzed by immunoprecipitated (IP) with an anti-CDK6 antibody (α-CDK6) and subsequently analyzed by immunoblotting with an anti-CDK6 antibody or an anti-NPM/B23 antibody. A control IgG antibody immunoblotting with an anti-CDK6 antibody or an anti-NPM/B23 antibody. A control IgG antibody (α-IgG) was used for mock immunoprecipitation. (b) ARK2 cells were transiently transfected (α-IgG) was used for mock immunoprecipitation. (b) ARK2 cells were transiently transfected with withpcDNA pcDNA (V), (V), HA-CDK4, HA-CDK4, or HA-CDK6-expressing or HA-CDK6-expressing vectors vectors for 24 h. for ARK2 24 h. whole-cell ARK2 whole-cell lysates were lysates wereimmunoprecipitated immunoprecipitated (IP)(IP) with with an ananti-HA anti-HA antibody antibody (α-HA) (α-HA) and andsubsequently subsequently analyzed analyzed by by immunoblottingimmunoblotting using using an an anti-HA anti-HA antibody antibody or an an anti-NPM/B23 anti-NPM/B23 antibody. antibody. TheThe asterisks asterisks indicate indicate non-speciﬁcnon-specific bands. bands. (c) Lysates (c) Lysates from ARK2from cellsARK2 underwent cells underwent immunoprecipitation immunoprecipitation using an using anti-NPM an /B23 antibody.anti-NPM/B23 An in vitroantibody.kinase An assayin vitro was kinase performed assay was on performed immunoprecipitates on immunoprecipitates using a recombinantusing a rCDK6recombinant/cyclinD1 rCDK6/cyclinD1 kinase (as described kinase in(as thedescribed Methods in the section). Methods The section). resulting The productsresulting products were probed withwere the indicatedprobed with antibodies the indicated (HC, antibodi heavyes chain; (HC, LC,heavy light chain; chain). LC, light chain).

2.5. Phosphorylated2.5. Phosphorylated NPM NPM/B23/B23 (Thr199) (Thr199) Forms Forms aa Preferential Complex Complex with with CDK6 CDK6 NPMNPM/B23/B23 was was found found to form to form a protein a protein complex complex with CDK6with CDK6 (Figure (Figure5a). Notably, 5a). Notably, this interaction this wasinteraction signiﬁcantly was weakened significantly upon weakened treatment upon withtreatm palbociclibent with palbociclib (Figure 6(Figurea). The 6a). association The association between between NPM/B23 and CDK6 was further confirmed when Thr at position 199 was replaced with NPM/B23 and CDK6 was further conﬁrmed when Thr at position 199 was replaced with Ala in the Ala in the phospho-mimetic mutant NPM/B23 (Thr199D) (Figure 6b). Altogether, these data indicate phospho-mimeticthat the NPM/B23/CDK6 mutant NPM interaction/B23 (Thr199D)critically depends (Figure on6 b).phosphorylation Altogether, theseat the dataThr199 indicate residue. that the NPM/B23/CDK6 interaction critically depends on phosphorylation at the Thr199 residue. Cancers 2019, 11, 1025 9 of 16 Cancers 2019, 11, x 10 of 17

FigureFigure 6.6. PhosphorylatedPhosphorylated NPMNPM/B23/B23 (Thr199)(Thr199) isis capablecapable ofof associatingassociating withwith CDK6.CDK6. ((aa)) ARK2ARK2 werewere treatedtreated withwith 22 µµMM palbociclibpalbociclib forfor 2424 hh (at(at thethe experimentalexperimental conditionsconditions describeddescribed inin thethe MethodsMethods section).section). Whole-cellWhole-cell lysates lysates were were immunoprecipitated immunoprecipitated (IP) (IP) with with an anti-CDK6an anti-CDK6 antibody antibody (α-CDK6) (α-CDK6) and subsequentlyand subsequently analyzed analyzed by immunoblotting by immunoblotting with an with anti-CDK6 an anti-CDK6 antibody antibody or an anti-NPM or an anti-NPM/B23/B23 antibody. (antibody.b) ARK2 cells(b) ARK2 were cells transfected were transfected with the with dephospho-mimetic the dephospho-mimetic NPM/B23 NPM/B23 mutant mutant (Thr199A) (Thr199A) or the phospho-mimeticor the phospho-mimetic NPM/B23 NPM/B23 mutant (Thr199D)mutant (Thr199D) for 24 h and for immunoprecipitated24 h and immunoprecipitated (IP) with an (IP) anti-CDK6 with an antibodyanti-CDK6 (α -CDK6).antibody Subsequently, (α-CDK6). Subsequently, immunoblotting immunoblotting with an anti-CDK6 with an antibody anti-CDK6 or an antibody anti-NPM or/B23 an antibodyanti-NPM/B23 was performed. antibody was A control perfor IgGmed. antibody A control (α-IgG) IgG was antibody used for (α mock-IgG) immunoprecipitationwas used for mock (HC,immunoprecipitation heavy chain; LC, (HC, light chain).heavy chain; LC, light chain).

3.3. DiscussionDiscussion InIn thisthis study,study, wewe demonstrateddemonstrated thatthat anan increasedincreased expressionexpression ofof phosphorylatedphosphorylated NPMNPM/B23/B23 (at(at Thr199)Thr199) inin cancercancer cellscells promotespromotes itsits interactioninteraction withwith CDK6CDK6 andand playsplays aa keykey rolerole inin endometrialendometrial α tumorigenesis.tumorigenesis. ByBy inhibiting inhibiting the the CDK4 CDK4/6/6 kinase, kinase, palbociclib palbociclib disrupts disrupts this this interaction interaction and promotesand promotes ER expression,ERα expression, ultimately ultimately sensitizing sensitizing hormone-refractory hormone-refr endometrialactory endometrial cancer cellscancer to endocrinecells to endocrine therapy (Figuretherapy7 ).(Figure 7). TheThe potentialpotential clinical clinical usefulness usefulness of palbociclibof palbociclib in patients in patients with endometrial with endometrial malignancies malignancies remains unclear.remains Aunclear. phase IIA clinicalphase II trial clinical (NCT02730429) trial (NCT02730429) is currently is examiningcurrently examining the association the association of letrozole, of anletrozole, aromatase an aromatase inhibitor, with inhibitor, palbociclib with /palbociclibplacebo in/placebo patients within patients metastatic with endometrial metastatic endometrial cancer, but resultscancer, are but still results unavailable. are still unavailable. Our current Our preclinical current data preclinical indicate data that indicate the combination that the combination of palbociclib of andpalbociclib letrozole and can letrozole exert synergistic can exert synergistic effects against effects tumor against growth tumor (Supplementary growth (Supplementary Figure S4). Figure S4). TheThe PALOMAPALOMA 2 phase2 phase III trial III demonstratedtrial demonstrated that in previouslythat in previously untreated patientsuntreated with patients ER-positive, with HER2-negativeER-positive, HER2-negative advanced breast advanced cancer, breast the combination cancer, the ofcombination palbociclib of and palbociclib letrozole and resulted letrozole in a significantlyresulted in longera significantly progression-free longer progression-free survival compared survival with letrozole compared alone with (24.8 letrozole versus 14.5 alone months, (24.8 respectively,versus 14.5 months,p < 0.001) respectively, [18]. Moreover, p < the0.001) PALOMA [18]. Moreover, 3 phase III the trial PALOMA showed that3 phase the progression-freeIII trial showed survivalthat the ofprogression-free patients treated survival with palbociclib of patients plus fulvestranttreated with was palbociclib significantly plus longer fulvestrant than that was of patientssignificantly who receivedlonger than fulvestrant that of alonepatients (9.5 who versus received 4.5 months, fulvestrant respectively, alone p(9.5< 0.001) versus [19 4.5]. months, respectively, p < 0.001) [19]. Cancers 2019, 11, 1025 10 of 16 Cancers 2019, 11, x 11 of 17

Figure 7. Schematic representation of the inhibitory effects of palbociclib on endometrial cancer cell Figure 7. Schematic representation of the inhibitory effects of palbociclib on endometrial cancer cell growth. Palbociclib is capable of inhibiting phosphorylation of NPM/B23 (Thr199) and promoting ERα expression,growth. Palbociclib ultimately is sensitizingcapable of hormone-refractoryinhibiting phosphorylation endometrial of NPM/B23 cancers to (Thr199) endocrine and therapy. promoting ERα expression, ultimately sensitizing hormone-refractory endometrial cancers to endocrine Intherapy. our study, palbociclib and megestrol acetate were found to exert synergistic antiproliferative effects against ERα-negative ARK2. However, the results obtained in our xenograft model revealed that palbociclibIn our study, alone palbociclib was capable and ofmegestrol reducing acetate cell proliferation were found in to ER exertα-positive synergistic HEC1B antiproliferative cells, but not ineffects ERα -negativeagainst ER ARK2α-negative cells (FigureARK2. 2However, and Supplementary the results obtained Figure S3). in our The xenograft key relevance model of revealed ER α in mediatingthat palbociclib palbociclib alone activitywas capable is consistent of reducing with literaturecell proliferation data obtained in ERα on-positive breast cancerHEC1B [28 cells,]. Our but data not pointin ER toα-negative a therapeutic ARK2 utility cells of(Figure palbociclib 2 and plus Supplementary megestrol acetate Figure either S3). The in ER keyα-positive relevance endometrial of ERα in cancermediating or for palbociclib reversing hormoneactivity is therapy consistent resistance with lit inerature Erα-negative data obtained endometrial on breast tumors. cancer We postulated [28]. Our thatdata the point effect to ofa combinedtherapeutic treatment utility of would palbociclib be the activationplus megestrol of Erα acetateby palbociclib, either in which ERα-positive leads to growthendometrial inhibition cancer by megestrolor for reversin acetate.g hormone Because palbociclibtherapy resistance is not a specific in Erα NPM-negative/B23 inhibitor,endometrial we cannottumors. rule We out postulated other mechanisms that the thateffect activate of combined ERα to resulttreatment in decreased would be tumor the growthactivation by megestrol.of Erα by palbociclib,NPM/B23 which is a nuclear leads proteinto growth found inhibition to be overexpressed by megestrol in severalacetate. malignancies Because palbociclib [13]. Interestingly, is not a dephosphorylationspecific NPM/B23 inhibitor, of NPM/ B23we hascannot been ru shownle out other to result mechanisms in restoration that ofactivate E2F1-dependent ERα to result DNA in repairdecreased [29]. tumor Phosphorylation growth by megestrol. of the Thr234 /237 sites in hepatocellular carcinoma cells was related to invasivenessNPM/B23 and is migration a nuclear [30 protein]. Phosphorylation found to ofbe NPMoverexpressed/B23 at Thr199 in isseveral mediated malignancies by CDK2/cyclin [13]. E,Interestingly, with phospho-NPM dephosphorylation/B23 (Thr199) of being NPM/B23 involved ha ins bothbeen centrosome shown to duplication result in andrestoration pre-mRNA of processingE2F1-dependent [31]. We DNA showed repair that [29]. phospho-NPM Phosphorylat/B23ion (Thr199) of the is Thr234/237 a substrate ofsites CDK6-cyclin in hepatocellular D1 and phosphorylationcarcinoma cells was of NPM related/B23 isto ainvasiveness prerequisite forand CDK6 migration interaction. [30]. Phosphorylation of NPM/B23 at Thr199Numerous is mediated studies by haveCDK2/cyclin identified E, factorswith ph thatospho-NPM/B23 may influence (Thr199) palbociclib being activity, involved including in both Rbcentrosome [22], hormone duplication receptor and status pre-mRNA [18], p16 expression processing [32 [31].], and We PTEN showed [24]. Ourthat current phospho-NPM/B23 data indicate that(Thr199) palbociclib is a substrate treatment of reducedCDK6-cyclin phospho-NPM D1 and phosphorylation/B23 (Thr199) and of NPM/B23 induced ER isα a expressionprerequisite both for inCDK6 human interaction. endometrial cancer cells and in a xenograft tumor model. Moreover, IHC revealed that Numerous studies have identified factors that may influence palbociclib activity, including Rb [22], hormone receptor status [18], p16 expression [32], and PTEN [24]. Our current data indicate that palbociclib treatment reduced phospho-NPM/B23 (Thr199) and induced ERα expression both in human endometrial cancer cells and in a xenograft tumor model. Moreover, IHC revealed that Cancers 2019, 11, 1025 11 of 16 phospho-NPM/B23 (Thr199) is overexpressed in endometrial cancer, suggesting that it could serve as a biomarker to guide palbociclib treatment in future clinical trials. NPM/B23 has the potential to act as an oncoprotein and may play a role in targeted cancer therapy. Several molecules that are capable of interacting with NPM/B23 have shown promising antiproliferative potential [33], albeit they are unsuitable for drug development. Although palbociclib is not a direct inhibitor of NPM/B23, it can to some extent disrupt its interaction with CDK6.

4. Materials and Methods

4.1. Cell Cultures HEC1B endometrial cancer cells were purchased from the American Type Culture Collection (Manassas, VA, USA). ARK2 endometrial cancer cells were obtained from Dr. Santin (Yale University, School of Medicine, New Haven, CT, USA) [34]. Ishikawa endometrial cancer cells were kindly provided by Dr. Nishida (National Hospital Organization, Kasumigaura Medical Center, Japan). HEC1B cells and Ishikawa cells were grown in α-MEM medium containing 15% (v/v) fetal bovine serum (FBS). ARK2 cells were grown in RPMI-1640 medium containing 10% (v/v) FBS.

4.2. Antibodies, Reagents, and Plasmids Mouse monoclonal antibodies against NPM/B23, β-actin, CDK4, and CDK6 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal antibodies against PARP, ERα, and phospho-NPM/B23 (Thr234/237) and rabbit monoclonal antibodies raised against phospho-NPM/B23 (Ser125) and HA were purchased from Abcam (Cambridge, MA, USA). The rabbit polyclonal antibody against phospho-NPM (Thr199) was from Cell Signaling Technology (Danvers, MA, USA). Further, anti-PR rabbit monoclonal antibody was obtained from GeneTex (San Antonio, TX, USA); Palbociclib and megestrol acetate were obtained from Pﬁzer Manufacturing Deutschland GmbH (Freiburg, Germany) and TTY Biopharm Company (Taipei City, Taiwan), respectively. All chemicals were from Sigma (St. Louis, MO, USA) unless otherwise indicated. The expression vector plasmids for HA-CDK4 and HA-CDK6 were from Addgene (Cambridge, MA, USA). The expression vector plasmids for NPM/B23 (WT), NPM/B23 (Thr199A), and NPM/B23 (Thr199D) were previously described in detail [29].

4.3. DNA Transfection Experiments In overexpression experiments, NPM/B23 (WT), NPM/B23 (Thr199A), NPM/B23 (Thr199D), HA-CDK4, HA-CDK6, and pcDNA (V) were transfected into ARK2 cells using the Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. Transient transfection was performed 24 h before cell harvesting.

4.4. Western Blot Analysis The western blot protocol was previously described in detail [25,26]. In brief, cells were harvested, washed twice in phosphate-buffered saline (PBS), and then lysed in ice-cold RIPA lysis buffer (1% Triton X-100, 1% NP40, 0.1% SDS, 0.5% DOC, 20 mM Tris-HCl pH 7.4, 150 mM NaCl, cocktail protease inhibitor, and phosphatase inhibitor cocktail) (Won-Won Biotechnology, Taipei, Taiwan) for 30 min. Lysates were boiled in 4 sample buffer dye and subjected to 10% SDS-PAGE. After separation, proteins × were electrotransferred onto a nitrocellulose membrane (Amersham Pharmacia Biotech/GE, Healthcare, Piscataway, NJ, USA). The reported primary and secondary antibodies were subsequently used to probe the blots. The immunoreactive bands were detected with the enhanced chemiluminescence reaction (Millipore, Billerica, MA, USA).

4.5. RNA Extraction and Real-time QPCR Total RNA was extracted from cell lines using the TOOLSmart RNA Extractor (Biotools Co., Ltd. Taipei City, Taiwan) according to the manufacturer’s instructions. RNA samples Cancers 2019, 11, 1025 12 of 16 were subjected to quantitative real-time QPCR (RT-QPCR) analysis. Transcription levels were normalized to GAPDH values of each sample. Primer sequences were as follows: ERα, 50-GGAGGGCAGGGGTGAA-30 (sense), 50-GGCCAGGCTGTTCTTCTTAG-30 (antisense); NPM/B23, 50-GGGGCTTTGAAATAACACCA-30 (sense), 50-GAACCTTGCTACCACCTCCA-30 (antisense); cathepsin D, 50-GTACATGATCCCCTGTGAGAAGGT-30 (sense), 50-GGGACAGCTTGT AGCCTTTGC-30 (antisense); EBAG9, 50-GATGCACCCACCAGTGTAAAGA-30 (sense), 50-AGTCAGGTTCCAGTTGTTCCAAAG-30 (antisense); TFF1/pS2, 50-ACATGGAAGGATTT GCTGATA -30 (sense), 50-TTCCGGCCATCTCTCACTAT -30 (antisense); GAPDH: 50-GGTATCG TGGAAGGACTCATGAC-30 (sense), 50-ATGCCAGTGAGCTTCCCGT-30 (antisense). Ampliﬁcations were carried out with an initial denaturation at 95 ◦C for 10 min, followed by 45 cycles of 95 ◦C for 15 s and 60 ◦C for 1 min. An ABI PRISM 7900 HT instrument (Applied Biosystems, Foster City, CA, USA) was used for all reactions. Each measurement was performed in duplicate and the threshold cycle (Ct) was determined.

4.6. Cell Viability Assay Approximately 1 104 cells were seeded into each well of a 96-well culture plate for 24 h. Cells in × serum-free media were treated with palbociclib, megestrol acetate, or letrozole. The MTT viability assay was then performed by adding MTT (5 mg/mL, 25 µL) into each well. After a 4 h incubation at 37 ◦C, the supernatant was discarded and DMSO (100 µL) was added to each well. The mixture was shaken to dissolve the formazan, and the absorbance was read at 570 nm in a multiwell spectrophotometer (VICTOR 2; Perkin Elmer GMI, Ramsey, MN, USA).

4.7. Clonogenic Assays Approximately 5000 cells treated with diﬀerent dose of palbociclib, megestrol acetate, or letrozole were seeded into 6-well dishes and maintained for 5 14 days. Cells were subsequently ﬁxed in 30% − methanol/12.5% acetic acid and stained with Brilliant Blue R to visualize the colonies.

4.8. Animals and Treatment Six-week-old female BALB/c nude mice were obtained from the National Laboratory Animal Center, Taipei City, Taiwan. The study protocol was reviewed and approved by the Animal Care Committee of the Chang Gung Memorial Hospital Institutional Review Board (approval number 2016101703 and 2017081501). ARK2 or HEC1B cells were harvested, washed, and resuspended in Hanks’ balanced salt solution (HBSS) at 107 cells/mL. Tumors were established through the subcutaneous inoculation of the cell suspension (100 µL) into the lateral hind leg of nude mice aged 6 8 weeks. − After 28 days, animals were treated with palbociclib (100 mg/kg given orally 5 days a week) and/or megestrol acetate (10 mg/kg once per week). Specifically, mice were randomized to four different treatment arms (n = 3 each), as follows: (1) vehicle; (2) palbociclib alone; (3) megestrol acetate alone; and (4) their combination. During the course of treatment, tumor growth was monitored on a weekly basis. Tumor volume (cm3) was calculated as in vivo proxy of tumor mass. At the end of the experiments, tumors were excised and extracted using the RIPA buffer. Extracts were subjected to western blotting as described above.

4.9. Synergistic Effect of the Combined Drug Treatment To assess the synergistic effect of palbociclib, megestrol acetate, or letrozole, we calculated the combination index (CI) using the CompuSyn software. Drug interactions were examined according to the median-effect principle of the Chou-Talalay method [35], which is based on the use of the MTT assay and the colony formation assay [36]. Cancers 2019, 11, 1025 13 of 16

4.10. RNA Interference Procedures Cells were transiently transfected either with speciﬁc siRNAs targeting NPM/B23, CDK4, CDK6, or control siRNA (Ambion, Austin, TX, USA) using the Lipofectamine RNAiMAX reagent (Invitrogen). In brief, Lipofectamine RNAiMAX was incubated with the Opti-MEM medium without phenol red (Invitrogen) for 5 min at room temperature. Speciﬁc siRNAs were added to the Lipofectamine RNAiMAX mixture and incubated at room temperature for 30 min to promote the formation of the transfection complex. Transfection mixtures were subsequently added to cells in the Opti-MEM medium. After 72 h, cells were harvested for subsequent PAGE and western blot analysis. The sequences of NPM/B23 siRNA were as follows: 50-CUGGUGCAAAGGAUGAGUUTT-30 (sense) and 50-AACUCAUCCUUUGCACCAGTT-30 (antisense), CAGUGGUCUUAAGGUUGAATT-30 (sense) and 50-UUCAACCUUAAGACCACUGTT-30 (antisense), GGAAGAUGCAGAGUCAGAATT-30 (sense) and 50-UUCUGACUCUGCAUCUUCCTT-30 (antisense), whereas the sequences of CDK4 siRNA were 50-GAUGCGCCAGUUUCUAAGATT-30 (sense) and 50-UCUUAGAAACUGGCGCAUCTT-30 (antisense), 50-CAGAGAUGUUUCGUCGAAATT-30 (sense) and 50-UUUCGACGAAACAUCUCUGTT-30 (antisense), 50-CUCUGCAGCACUCUUAUCUTT-30 (sense) and 50-AGAUAAGAGUGCUGCAGAGTT-30 (antisense), 50-CAGCACUCUUAUCUACAUATT-30 (sense) and 50-UAUGUAGAUAAGAGUGCUGTT-30 (antisense). The sequences of CDK6 siRNA were as follows: 50-GAGUAGUUCUCUCUAACUATT-30 (sense) and 50-UAGUUAGAGAGAACUACUCTT-30 (antisense), 50-GCAGAAAUGUUUCGUAGAATT-30 (sense) and 50-UUCUACGAAACAUUUCUGCTT-30 (antisense), 50-GACUCAAGGUGGUCAGUAATT-30 (sense) and 50-GGAGAGUAGUUCUCUCUAATT-30 (antisense), 50-UUAGAGAGAACUACUCUCCTT-30 (antisense). The sequences of negative-control siRNA were 50-UAACGACGCGACGACGUAA-30 (sense) and 50-UUACGUCGUCGCGUCGUUA-30 (antisense).

4.11. In Vitro Kinase Assay Cells’ extracts were subjected to immunoprecipitation using the anti-NPM/B23 antibody. Immunocomplexes were washed with WCE buﬀer and kinase reaction buﬀer (25 mM Tris (pH 7.5), 5 mM β-glycerophosphate, 2 mM DTT, 0.1 mM Na3VO4, and 10 mM MgCl2). One µg of CDK6/Cyclin D1 (Invitrogen) was added to the immunocomplexes for 1 h at 30 ◦C. For inhibition studies, immunocomplexes and CDK6/Cyclin D1 were incubated in the presence of 2 µM palbociclib. The degree of phosphorylation was analyzed with both SDS-PAGE and immunoblotting.

4.12. Immunoprecipitation Cells were harvested and washed twice in ice-cold PBS. Cell pellets were subsequently resuspended in ice-cold WCE lysis buffer (20 mM HEPES, 10% glycerol, 0.5% Triton X-100, 0.2 M sodium chloride, 1 mM EDTA, 1 mM EGTA, and protease inhibitor cocktail) for 30 min and centrifuged at 12,000 rpm at 4 ◦C for 30 min. Equal amounts of each cell extract protein were incubated with the indicated antibodies (2 µg) at 4 ◦C for 2 h. Immune complexes were captured with protein G-sepharose (30 µL; Upstate Biotechnology, Lake Placid, NY, USA) for 2 h at 4 ◦C under rotation. The protein G-antigen-antibody complexes were washed four times with WCE lysis buffer and boiled in 2 urea sample buffer dye × (100 mM Tris-HCl, pH 6.8, 4% SDS, 0.2% bromophenol blue, 20% glycerol, 200 mM β-mercaptoethanol, 8 M urea) for subsequent PAGE and western blot analysis.

4.13. Immunohistochemistry and Clinical Tissue Specimens The immunohistochemical study was approved by the local Institutional Review Board (approval number 201601370B0 and 201701722B0). Immunohistochemistry (IHC) was performed as previously described in detail [26]. In brief, 4-µm-thick formalin-fixed, paraffin-embedded tissue slices were deparaffinized in xylene and rehydrated through graded washes of ethanol in water. Sections were stained with the anti-phospho-NPM/B23 (T199) or ERα antibody using a Basic DAB detection kit (Ventana Medical Systems, Tucson, AZ, USA) in an automated IHC stainer. Hematoxylin was used Cancers 2019, 11, 1025 14 of 16 for counterstaining. A semiquantitative immunostaining score (histoscore) was calculated as the percentage of positive cells multiplied by their staining intensity (0 = negative, 1 = weak, 2 = moderate, 3 = strong). Consequently, the histoscore ranged from a minimum of 0 to a maximum of 300 (i.e., 100% multiplied by 3). IHC was also performed on a commercially available endometrial cancer tissue array (EMC1021; US Biomax Inc., Rockville, MD, USA).

4.14. Proliferation Assay To study the eﬀects of NPM/B23, dephospho-mimetic NPM/B23 mutant (Thr199A) or the phospho-mimetic NPM/B23 mutant (Thr199D) mediated cell proliferation. Cells were seeded at a density of 104 cells per well in complete medium for 24 h. The cells were grown for an additional 24 h after the addition of BrdU, and DNA synthesis was assayed using an enzyme-linked immunosorbent assay for BrdU incorporation (Roche Applied Science, Indianapolis, IN, USA).

5. Conclusions In summary, our results demonstrated that palbociclib promoted NPM/B23 dephosphorylation at Thr199—an eﬀect mediated by disruption of CDK6 kinase activity. This may have therapeutic relevance and can prompt further preclinical and clinical studies on the potential usefulness of palbociclib as an NPM/B23 modulator in endometrial cancer.

Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6694/11/7/1025/s1, Figure S1: PR expression was not affected by palbociclib treatment, Figure S2: Palbociclib induces ERα expression in late-passage (n > 40) Ishikawa endometrial cancer cells, Figure S3: The combination of palbociclib and megestrol acetate exerts antitumor effects in endometrial cancer cells and in a xenograft tumor model, Figure S4: The combination of palbociclib and letrozole exerts synergistic antitumor effects in endometrial cancer cells, Figure S5: Immunohistochemical expression of phospho-NPM/B23 (Thr234/237) in normal endometrium and endometrial cancer, Figure S6: Immunohistochemical expression of phospho-NPM/B23 (Thr199) in endometrial cancer. Table S1: Clinical information of patients with endometrial cancer. Author Contributions: Conceptualization, C.-Y.L. and A.C.; Formal analysis, C.-Y.L., L.-Y.L., and A.C.; Funding acquisition, C.-H.L.; Investigation, T.-H.W., C.-L.H., and C.-L.T.; Methodology, C.-Y.L. and L.-Y.L.; Project administration, C.-H.L.; Resources, C.-L.T. and C.-H.L.; Supervision, A.C.; Writing—original draft, C.-Y.L. and A.C.; Writing—review & editing, A.C. and C.-H.L. Funding: This study was financially supported by grants from the Ministry of Science and Technology (106-2320-B-182A-007 to C.-Y.L.) and the Chang Gung Foundation (CMRPG3H0331 to C.-Y.L. and CORPG3H0471/0481 to A.C.). The authors are grateful to Jung-Erh Yang for her excellent technical assistance. Formalin-fixed paraffin-embedded specimens were kindly provided by the Tumor Bank. Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Lortet-Tieulent, J.; Ferlay, J.; Bray, F.; Jemal, A. International Patterns and Trends in Endometrial Cancer Incidence, 1978–2013. J. Natl. Cancer Inst. 2018, 110, 354–361. [CrossRef][PubMed] 2. Health Promotion Administration Ministry of Health and Welfare Taiwan. Cancer Registry Annual Report, 2015 Taiwan; Health Promotion Administration Ministry of Health and Welfare Taiwan: Taipei, Taiwan, 2017. 3. Creasman, W.T.; Miller, D.S. Adenocarcinoma of the uterine corpus. In Clinical Gynecologic Oncology; Di Saia, P.J., Creasman, W.T., Mannel, R.S., McMeekin, D.S., Mutch, D.G., Eds.; Elsevier: Philadelphia, PA, USA, 2012; pp. 121–154. 4. Creasman, W.T.; Soper, J.T.; McCarty, K.S.; McCarty, K.S., Jr.; Hinshaw, W., Sr.; Clarke-Pearson, D.L. Inﬂuence of cytoplasmic steroid receptor content on prognosis of early stage endometrial carcinoma. Am. J. Obstet. Gynecol. 1985, 151, 922–932. [CrossRef] 5. Thigpen, J.T.; Brady, M.F.; Alvarez, R.D.; Adelson, M.D.; Homesley, H.D.; Manetta, A.; Soper, J.T.; Given, F.T. Oral medroxyprogesterone acetate in the treatment of advanced or recurrent endometrial carcinoma: A dose-response study by the Gynecologic Oncology Group. J. Clin. Oncol. 1999, 17, 1736–1744. [CrossRef] [PubMed] Cancers 2019, 11, 1025 15 of 16

6. Carlson, M.J.; Thiel, K.W.; Leslie, K.K. Past, present, and future of hormonal therapy in recurrent endometrial cancer. Int. J. Women’s Health 2014, 6, 429–435. 7. Jongen, V.; Briet, J.; de Jong, R.; ten Hoor, K.; Boezen, M.; van der Zee, A.; Nijman, H.; Hollema, H. Expression of estrogen receptor-alpha and -beta and progesterone receptor-A and -B in a large cohort of patients with endometrioid endometrial cancer. Gynecol. Oncol. 2009, 112, 537–542. [CrossRef][PubMed] 8. Kuukasjarvi, T.; Kononen, J.; Helin, H.; Holli, K.; Isola, J. Loss of estrogen receptor in recurrent breast cancer is associated with poor response to endocrine therapy. J. Clin. Oncol. 1996, 14, 2584–2589. [CrossRef][PubMed] 9. Peter, M.; Nakagawa, J.; Doree, M.; Labbe, J.C.; Nigg, E.A. Identification of major nucleolar proteins as candidate mitotic substrates of cdc2 kinase. Cell 1990, 60, 791–801. [CrossRef] 10. Negi, S.S.; Olson, M.O. Effects of interphase and mitotic phosphorylation on the mobility and location of nucleolar protein B23. J. Cell Sci. 2006, 119, 3676–3685. [CrossRef] 11. Chan, P.K.; Aldrich, M.; Cook, R.G.; Busch, H. Amino acid sequence of protein B23 phosphorylation site. J. Biol. Chem. 1986, 261, 1868–1872. 12. Okuwaki, M.; Tsujimoto, M.; Nagata, K. The RNA binding activity of a ribosome biogenesis factor, nucleophosmin/B23, is modulated by phosphorylation with a cell cycle-dependent kinase and by association with its subtype. Mol. Biol. Cell 2002, 13, 2016–2030. [CrossRef] 13. Okuda, M. The role of nucleophosmin in centrosome duplication. Oncogene 2002, 21, 6170–6174. [CrossRef] 14. Adon, A.M.; Zeng, X.; Harrison, M.K.; Sannem, S.; Kiyokawa, H.; Kaldis, P.; Saavedra, H.I. Cdk2 and Cdk4 regulate the centrosome cycle and are critical mediators of centrosome amplification in p53-null cells. Mol. Cell. Biol. 2010, 30, 694–710. [CrossRef] 15. Sarek, G.; Jarviluoma, A.; Moore, H.M.; Tojkander, S.; Vartia, S.; Biberfeld, P.; Laiho, M.; Ojala, P.M. Nucleophosmin phosphorylation by v-cyclin-CDK6 controls KSHV latency. PLoS Pathog. 2010, 6, e1000818. [CrossRef] 16. Hanahan, D.; Weinberg, R.A. Hallmarks of cancer: The next generation. Cell 2011, 144, 646–674. [CrossRef] 17. Hamilton, E.; Infante, J.R. Targeting CDK4/6 in patients with cancer. Cancer Treat. Rev. 2016, 45, 129–138. [CrossRef] 18. Finn, R.S.; Martin, M.; Rugo, H.S.; Jones, S.; Im, S.A.; Gelmon, K.; Harbeck, N.; Lipatov, O.N.; Walshe, J.M.; Moulder, S.; et al. Palbociclib and Letrozole in Advanced Breast Cancer. N. Engl. J. Med. 2016, 375, 1925–1936. [CrossRef] 19. Cristofanilli, M.; Turner, N.C.; Bondarenko, I.; Ro, J.; Im, S.A.; Masuda, N.; Colleoni, M.; DeMichele, A.; Loi, S.; Verma, S.; et al. Fulvestrant plus palbociclib versus fulvestrant plus placebo for treatment of hormone-receptor-positive, HER2-negative metastatic breast cancer that progressed on previous endocrine therapy (PALOMA-3): Final analysis of the multicentre, double-blind, phase 3 randomised controlled trial. Lancet Oncol. 2016, 17, 425–439. 20. Hortobagyi, G.N.; Stemmer, S.M.; Burris, H.A.; Yap, Y.S.; Sonke, G.S.; Paluch-Shimon, S.; Campone, M.; Blackwell, K.L.; Andre, F.; Winer, E.P.; et al. Ribociclib as First-Line Therapy for HR-Positive, Advanced Breast Cancer. N. Engl. J. Med. 2016, 375, 1738–1748. [CrossRef] 21. Beaver, J.A.; Amiri-Kordestani, L.; Charlab, R.; Chen, W.; Palmby, T.; Tilley, A.; Zirkelbach, J.F.; Yu, J.; Liu, Q.; Zhao, L.; et al. FDA Approval: Palbociclib for the Treatment of Postmenopausal Patients with Estrogen Receptor-Positive, HER2-Negative Metastatic Breast Cancer. Clin. Cancer Res. 2015, 21, 4760–4766. [CrossRef] 22. DeMichele, A.; Clark, A.S.; Tan, K.S.; Heitjan, D.F.; Gramlich, K.; Gallagher, M.; Lal, P.; Feldman, M.; Zhang, P.; Colameco, C.; et al. CDK 4/6 inhibitor palbociclib (PD0332991) in Rb+ advanced breast cancer: Phase II activity, safety, and predictive biomarker assessment. Clin. Cancer Res. 2015, 21, 995–1001. [CrossRef] 23. Tanaka, T.; Terai, Y.; Ashihara, K.; Fujiwara, S.; Tanaka, Y.; Sasaki, H.; Tsunetoh, S.; Ohmichi, M. The efficacy of the cyclin-dependent kinase 4/6 inhibitor in endometrial cancer. PLoS ONE 2017, 12, e0177019. [CrossRef] 24. Dosil, M.A.; Mirantes, C.; Eritja, N.; Felip, I.; Navaridas, R.; Gatius, S.; Santacana, M.; Colas, E.; Moiola, C.; Schoenenberger, J.A.; et al. Palbociclib has antitumour effects on Pten-deficient endometrial neoplasias. J. Pathol. 2017, 242, 152–164. [CrossRef] 25. Chao, A.; Lin, C.Y.; Tsai, C.L.; Hsueh, S.; Lin, Y.Y.; Lin, C.T.; Chou, H.H.; Wang, T.H.; Lai, C.H.; Wang, H.S. Estrogen stimulates the proliferation of human endometrial cancer cells by stabilizing nucleophosmin/B23 (NPM/B23). J. Mol. Med. 2013, 91, 249–259. [CrossRef] Cancers 2019, 11, 1025 16 of 16

26. Lin, C.Y.; Chao, A.; Wang, T.H.; Lee, L.Y.; Yang, L.Y.; Tsai, C.L.; Wang, H.S.; Lai, C.H. Nucleophosmin/B23 is a negative regulator of estrogen receptor alpha expression via AP2gamma in endometrial cancer cells. Oncotarget 2016, 7, 60038–60052. 27. Grisendi, S.; Mecucci, C.; Falini, B.; Pandolfi, P.P. Nucleophosmin and cancer. Nat. Rev. Cancer 2006, 6, 493–505. [CrossRef] 28. Finn, R.S.; Dering, J.; Conklin, D.; Kalous, O.; Cohen, D.J.; Desai, A.J.; Ginther, C.; Atefi, M.; Chen, I.; Fowst, C.; et al. PD 0332991, a selective cyclin D kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal estrogen receptor-positive human breast cancer cell lines in vitro. Breast Cancer Res. 2009, 11, R77. [CrossRef] 29. Lin, C.Y.; Tan, B.C.; Liu, H.; Shih, C.J.; Chien, K.Y.; Lin, C.L.; Yung, B.Y. Dephosphorylation of nucleophosmin by PP1beta facilitates pRB binding and consequent E2F1-dependent DNA repair. Mol. Biol. Cell 2010, 21, 4409–4417. [CrossRef] 30. Ching, R.H.; Lau, E.Y.; Ling, P.M.; Lee, J.M.; Ma, M.K.; Cheng, B.Y.; Lo, R.C.; Ng, I.O.; Lee, T.K. Phosphorylation of nucleophosmin at threonine 234/237 is associated with HCC metastasis. Oncotarget 2015, 6, 43483–43495. [CrossRef] 31. Tarapore, P.; Shinmura, K.; Suzuki, H.; Tokuyama, Y.; Kim, S.H.; Mayeda, A.; Fukasawa, K. Thr199 phosphorylation targets nucleophosmin to nuclear speckles and represses pre-mRNA processing. FEBS Lett. 2006, 580, 399–409. [CrossRef] 32. Konecny, G.E.; Winterhoff, B.; Kolarova, T.; Qi, J.; Manivong, K.; Dering, J.; Yang, G.; Chalukya, M.; Wang, H.J.; Anderson, L.; et al. Expression of p16 and retinoblastoma determines response to CDK4/6 inhibition in ovarian cancer. Clin. Cancer Res. 2011, 17, 1591–1602. [CrossRef] 33. Di Matteo, A.; Franceschini, M.; Chiarella, S.; Rocchio, S.; Travaglini-Allocatelli, C.; Federici, L. Molecules that target nucleophosmin for cancer treatment: An update. Oncotarget 2016, 7, 44821–44840. [CrossRef] 34. Zhao, S.; Choi, M.; Overton, J.D.; Bellone, S.; Roque, D.M.; Cocco, E.; Guzzo, F.; English, D.P.; Varughese, J.; Gasparrini, S.; et al. Landscape of somatic single-nucleotide and copy-number mutations in uterine serous carcinoma. Proc. Natl. Acad. Sci. USA 2013, 110, 2916–2921. [CrossRef] 35. Chou, T.C. Drug combination studies and their synergy quantification using the Chou-Talalay method. Cancer Res. 2010, 70, 440–446. [CrossRef] 36. Chao, A.; Lin, C.Y.; Wu, R.C.; Lee, Y.S.; Lee, L.Y.; Tsai, C.L.; Yang, L.Y.; Liu, H.; Chen, S.J.; Wang, T.H.; et al. The combination of everolimus and terameprocol exerts synergistic antiproliferative effects in endometrial cancer: Molecular role of insulin-like growth factor binding protein 2. J. Mol. Med. 2018, 96, 1251–1266. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).