RSK Promotes Prostate Cancer Progression in Bone through ... · RSK expression increases...

Oncogenes and Tumor Suppressors

RSK Promotes Prostate Cancer Progression inBone through ING3, CKAP2, and PTK6-MediatedCell SurvivalGuoyu Yu1, Yu-Chen Lee1, Chien-Jui Cheng2,3, Chuan-Fen Wu4, Jian H. Song5,Gary E. Gallick5, Li-Yuan Yu-Lee6, Jian Kuang4, and Sue-Hwa Lin1,5

Abstract

Prostate cancer has a proclivity to metastasize to bone. Themechanism by which prostate cancer cells are able to survive andprogress in the bonemicroenvironment is not clear. Identificationof molecules that play critical roles in the progression of prostatecancer in bone will provide essential targets for therapy. Ribo-somal S6 protein kinases (RSK) have been shown to mediatemany cellular functions critical for cancer progression. WhetherRSK plays a role in the progression of prostate cancer in bone isunknown. IHC analysis of human prostate cancer specimensshowed increased phosphorylation of RSK in the nucleus ofprostate cancer cells in a significant fraction of human prostatecancer bone metastasis specimens, compared with the primarysite or lymph node metastasis. Expression of constitutively activemyristylated RSK in C4-2B4 cells (C4-2B4/RSK) increased theirsurvival and anchorage-independent growth compared with C4-2B4/vector cells. Using anorthotopic bone injectionmodel, it was

determined that injecting C4-2B4/RSK cells into mouse femursenhanced their progression in bone compared with control cells.In PC3-mm2 cells, knockdown of RSK1 (RPS6KA1), the predom-inant RSK isoform, but not RSK2 (RPS6KA2) alone, decreasedanchorage-independent growth in vitro and reduced tumor pro-gression in bone and tumor-induced bone remodeling in vivo.Mechanistic studies showed that RSK regulates anchorage-inde-pendent growth through transcriptional regulation of factors thatmodulate cell survival, including ING3, CKAP2, and PTK6.Together, these data provide strong evidence that RSK is animportant driver in prostate cancer progression in bone.

Implications: RSK, an important driver in prostate cancer pro-gression in bone, has promising potential as a therapeutic targetfor prostate cancer bone metastasis. Mol Cancer Res; 13(2); 348–57.�2014 AACR.

IntroductionProstate cancer is the second leading cause of cancer-related

death in men in the United States. Mortality from prostate canceris due mainly to development of metastasis in bone. Prostatecancer has a proclivity to metastasize to bone. One critical featureformetastatic prostate cancer cells to colonize in bone is to survivein the bone microenvironment. The mechanism by which pros-tate cancer cells are able to survive and progress within the bonemicroenvironment is not clear. Identification of molecules that

play critical roles in the progression of prostate cancer in bonewillprovide targets for therapy.

Ribosomal S6 protein kinase (RSK) is a family of signaltransducing Ser/Thr kinases. Four isoforms, RSK1–4, have beenreported in mammalian cells (for review, see refs. 1–4). Thebest functionally characterized isoforms are RSK1 and RSK2.Each RSK isoform contains two nonidentical kinase domains,one at the N-terminus and one at the C-terminus. Phosphor-ylation of RSKs at Ser/Thr, which occurs at multiple sites, isrequired for RSK activation (4) and the N-terminal kinasedomain is primarily responsible for substrate phosphorylation(5). RSKs phosphorylate many proteins, both cytosolic andnuclear (2). The many effects of RSKs on various proteins maycontribute to the observations that RSKs mediate wide-rangingcellular processes, including proliferation (6–8), migration (9),and invasion (1).

Expression of RSK1 and 2 proteins, analyzed by Western blotanalysis, has been previously shown to increase in prostate cancerwhen the cancer is localized in the primary site (8). However,whether expression of RSKs is increased in bone metastases isunknown, likely due to the lack of suitable RSK antibody for IHCanalysis. Clark and colleagues (8) also showed that RSK inhibitiondecreases the proliferation of cancer cells, including LNCaP andPC3 prostate cancer cells and MCF-7 breast cancer cells, but notnormal breast epithelial cells MCF-10A (8). These observationssuggest that RSKs are involved in prostate cancer progression.Whether RSKs play a role in prostate cancer bone metastasis isunknown.

1Department of Translational Molecular Pathology, The University ofTexas M. D. Anderson Cancer Center, Houston, Texas. 2Department ofPathology, College of Medicine, Taipei Medical University, Taipei,Taiwan. 3Department of Pathology,Taipei Medical University Hospital,TaipeiMedical University,Taipei,Taiwan. 4Departmentof ExperimentalTherapeutics, The University of Texas M. D. Anderson Cancer Center,Houston,Texas. 5Department of Genitourinary Medical Oncology,TheUniversity of Texas M. D. Anderson Cancer Center, Houston, Texas.6Department of Medicine, Baylor College ofMedicine, Houston,Texas.

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

G. Yu, Y.-C. Lee, and C.-J. Cheng contributed equally to this article.

Corresponding Author: Sue-Hwa Lin, The University of Texas M.D. AndersonCancer Center, 1515 Holcombe Boulevard, Houston, TX 77030. Phone: 713-794-1559; Fax: 713-794-4672; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-14-0384-T

�2014 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 13(2) February 2015348

on September 9, 2021. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst September 4, 2014; DOI: 10.1158/1541-7786.MCR-14-0384-T

http://mcr.aacrjournals.org/

In this study, we examined the role of RSKs in prostate cancerbone metastasis. Our studies showed that expression of RSKs inprostate cancer cells increases cell survival and anchorage-inde-pendent growth in vitro and enhances prostate cancer progressionin bone in vivo.

Materials and MethodsMaterials

C4-2B4-LT and PC3-mm2-LT, expressing luciferase and redfluorescence protein Tomato, were generated as described previ-ously (10, 11). The authenticity of PC3-mm2 and C4-2B4 celllines was confirmed by fingerprinting. pGIPZ lentiviral humanPTK6 shRNA was from Thermo Scientific. RSK1, pRSK(T359/S363), CKAP2, b-actin antibodies were from Santa Cruz Biotech-nology. Anti-RSK2 antibody (clone Y83) was from Epitomics.Antibodies against total RSK (RSK1/RSK2/RSK3), p38-MAPK(D13E1), phospho p38-MAPK (Thr180/Tyr182) (D3F9), SAPK/JNK (56G8), p-SAPK/JNK (Thr183/Tyr185) (81E11) were fromCell Signaling Technology. Antibodies against PTK6 and ING3were from Proteintech. ThemyrRSK plasmid was kindly providedby Dr. John Blenis (Harvard Medical School).

ImmunohistochemistryFormalin-fixed, paraffin-embedded human prostate cancer

specimens fromprimary tumor (20 cases), lymphnodemetastasis(19 cases), and bone metastasis (20 cases) were obtained fromM.D. Anderson Cancer Center (MDACC) Prostate Cancer TissueBank through an institutional-approved IRB protocol. Immuno-histochemistry using pRSK(T359/S363) antibodies (Santa CruzBiotechnology) was performed using procedures described pre-viously (11). The staining was defined as positive when >10% ofthe tumor cells in the specimen were immunoreactive.

Generation of C4-2B4 cells overexpressing myrRSKcDNA encoding myrRSK (12) was inserted into bicistronic

retroviral vector pBMN-I-GFP. C4-2B4/RSK cells were generatedfrom C4-2B4-LT cells transduced with retrovirus generated frompBMN-RSK-GFP vector and selected by FACS. C4-2B4-LT cellstransduced with empty vector (C4-2B4/vector) were generatedsimilarly.

Western blotting analysisProtein concentration was determined by Coomassie Plus

assay. Proteins were separated in SDS-PAGE and immunoblottedas indicated.

Cell proliferation and soft agar colony assayCell proliferation was determined by viable cell counting. The

soft agar colony assay was performed as described by Giancottiand Ruoslahti (13).

Intrabone injection, bioluminescence imaging, and mCTThe luciferase-expressing prostate cancer cells were injected

into the right femurs of male SCID mice. Tumor growth wasmonitored weekly using bioluminescence imaging (BLI) with anIVIS Imaging System (Xenogen). Femurs and tibias were fixed informaldehyde, and specimens were imaged on an Explore LocusRS pre-clinical in vivo scanner (GEMedical Systems). Imageswerereconstructed and analyzed in MicroView (Parallax Innovations,Inc.).

Knockdown of RSK1 or RSK2 in PC3-mm2 cellsLentivirus-expressing shRNAs were generated by cotransfecting

RSK1 shRNA or RSK2 shRNA plasmids with pCMV-dR8.2 dvprand pCMV-VSVG packaging plasmids into 293FT cells. PC3-mm2-LT cells transduced with pLKO or pLKO-shRSK1 vectorswere selected by 5 mg/mL puromycin to generate PC3-pLKO andPC3-shRSK1, respectively. RSK2 knockdown PC3-mm2 cells weregenerated similarly using pGIPZ-shRSK2. To generate PC3-shRSK1/2 double knockdown cells, PC3-shRSK1 cells wereinfected with lentivirus from pGIPZ-shRSK2 and the cells selectedby FACS through GFP expressed from pGIPZ.

Reverse transcription and quantitative PCR analysisTotal RNA was extracted from cells using the RNeasy Mini Kit

(Qiagen). Quantitative real-time RT-PCR (qRT-PCR) was per-formed, using GAPDH as a control. The PCR primer sequencesare listed in Supplementary Table S1.

Generation of C4-2B4/RSK and C4-2B4/vector cellsoverexpressing ING3 and CKAP2

cDNAs encoding human ING3 (accession number: NM_019071.2) and CKAP2 (accession number: BC136332.1) werecloned by PCR and used to generate retroviral vectors. Theprimer sequences used are listed in Supplementary Table S1.C4-2B4/RSK and C4-2B4/vector cells infected with retrovirusescontaining ING3 or CKAP2 cDNA were selected by 400 mg/mLG418.

Figure 1.Expression of pRSK(T359/S363) in human prostate cancer (PCa). A,immunohistochemistry of pRSK(T359/S363) in human prostate cancerspecimens. Paraffin-embedded human prostate cancer specimens wereimmunostained with anti-pRSK(T359/S363) antibody. Primary tumorsare mainly negative (19/20) for pRSK(T359/S363). B, lymph node metastasisare mainly negative (18/19) for pRSK(T359/S363). C, a significant fraction ofbone metastasis specimens are positive (8/20, using 10% as cutoff)for pRSK(T359/S363). The pRSK(T359/S363) protein tends to localize in thenucleus (see enlargement in inset). B, bone; T, tumor. D, percentage ofprostate cancer specimens that are positive with pRSK(T359/S363).Magnification, �200.

RSK in Prostate Cancer Bone Metastasis

www.aacrjournals.org Mol Cancer Res; 13(2) February 2015 349




Knockdown PTK6 in C4-2B4/RSK cellsLentivirus-expressing PTK6 shRNAs were generated as

described above and used to infect C4-2B4/RSK cells.

Resultsp-RSK expression is increased in human prostate cancerspecimens

To examine whether RSKs are activated during prostate cancerprogression, we performed IHC staining of paraffin-embeddedhuman prostate cancer specimens. RSK isoforms are composed oftwo distinct kinase domains, both of which are activated byphosphorylation. Dalby and colleagues (14) showed that one ofthe mechanisms for RSK activation involves ERK-mediated phos-phorylation of two adjacent conserved residues (T359 and Ser363in RSK1) in the linker region of RSK, and S363 phosphorylationdirectly activates the N-terminal kinase domain. We used anti-pRSK(T359/S363) antibody to immunostain human prostatecancer specimens that represent various stages of prostate cancerprogression from primary prostate cancer to lymph node andbone metastasis. In primary prostate cancer, 19 of 20 primary

tumors were negative with anti-pRSK(T359/S363) staining(according to 10% cutoff value; Fig. 1A). In the lymph nodemetastases specimens, 18 of 19 were negative (Fig. 1B). In con-trast, in specimens frombonemetastases, eight of 20 cases stainedpositive (Fig. 1C and Supplementary Fig. S1). The differences inpRSK(T359/S363) expression between bone metastasis andlymph node metastasis or primary tumors are significant (P ¼0.003; Fig. 1D). In addition, the majority of pRSK(T359/S363) inbone metastasis specimens are localized in the nucleus (Fig. 1C,inset; Supplementary Fig. S1), consistent with the reports by Zhaoand colleagues (15) and Chen and colleagues (16) that RSKsundergo translocation to the nucleus following growth factorstimulation. These observations suggest that RSKs are activatedin a significant fraction of metastatic prostate cancer cells in boneandmay play a role in the progression of prostate cancer in bone.

RSK expression increases anchorage-independent growth ofC4-2B4 cells

To examine the role of RSK in prostate cancer bone metastasis,we expressed myrRSK, a constitutively activated RSK (12), in

Figure 2.Effect of myrRSK overexpression onC4-2B4 growth in bone. A, luciferase-labeled C4-2B4 cells expressingmyrRSK or empty vector wereselected for positivity with TomatoRFP and GFP (left). Western blottingfor the expression of RSK, pRSK-T359/S363 in C4-2B4/vector andC4-2B4/RSK cells. B, no differencein C4-2B4/vector and C4-2B4/RSKcell proliferation when culturedin 10% FBS. C, effects of serumconcentrations on the survival of C4-2B4/vector and C4-2B4/RSK cells. D,anchorage-independent growth ofC4-2B4/vector andC4-2B4/RSK cells.E, growth of C4-2B4/vector andC4-2B4/RSK cells in bone asmeasured by bioluminescence. F,quantification of the tumor growth inbone based on the BLI. Average BLI�SEMwas shown. The bioluminescencesignals were higher in C4-2B4/RSKtumors compared with those inC4-2B4/vector. � , P < 0.05.

Yu et al.

Mol Cancer Res; 13(2) February 2015 Molecular Cancer Research350




C4-2B4 cells, which is an androgen-independent subline derivedfrom LNCaP (17). To allow for in vivo BLI, myrRSK in a bicistronicretroviral vector (pBMN-myrRSK-GFP) was transduced into C4-2B4-LT cells that stably expressed luciferase and Tomato redfluorescence protein, also introduced by bicistronic vector(pBMN-Luc-Tomato). C4-2B4-LT cells expressing both luciferaseandmyrRSK (C4-2B4/RSK)were selected by FACS for cells expres-sing both GFP and Tomato (Fig. 2A, left). Control C4-2B4/vectorcells were generated by transducing C4-2B4-LT with pBMN-I-GFPvector. Western blot analyses showed that overexpression of RSKin C4-2B4 cells not only increased the levels of total RSK but alsothe phosphorylation of RSK at T359/S363, consistent with theexpression of a constitutively activated RSK (Fig. 2A, right).

Next,we examined the effects of RSKonC4-2B4proliferation asRSKs are known to regulate cell-cycle progression through mod-ulation of protein factors that play a role in G1, G1–S, or G2–Mtransition (for review, see ref. 2). In prostate cancer, studies byClark and colleagues (8) showed that treatment of LNCaP andPC3 with the RSK inhibitor SL0101 led to inhibition of prostatecancer cell proliferation. We found that there is no significantdifference in cell proliferation between C4-2B4/RSK and thecontrol C4-2B4/vector cells when the cells were cultured inmediacontaining 10% (Fig. 2B) or 5% (Fig. 2C) FBS, indicating thatexpression of myrRSK in C4-2B4 cells does not affect theirproliferation under standard culture condition. However, whenthese cells were cultured in media containing lower concentra-

tions of FBS, i.e., 1.0%or 0.1%, the number of C4-2B4/RSK cells issignificantly higher compared with C4-2B4/vector cells (Fig. 2C).These observations suggest that expression of myrRSK confers asurvival advantage for C4-2B4 cells in low serum growthconditions.

Next, we examinedwhether the RSK-mediated survival can leadto increased anchorage-independent growth in soft agar. Whenplated in soft agar, noneor very few colonies, if any,were observedin C4-2B4/vector cells (Fig. 2D), whereas a significant number ofcolonies were observed in C4-2B4/RSK plates (Fig. 2D). Thus,myrRSK expression confers anchorage-independent growth ofC4-2B4 cells.

Overexpression of constitutively activated RSK in C4-2B4increases C4-2B4 tumor progression in bone

One critical feature for metastatic prostate cancer cells toprogress in bone is to survive in the bone microenvironment. Itis possible that myrRSK-mediated anchorage-independentgrowth of C4-2B4 cells may enhance C4-2B4 tumor growth inbone. To examine the effect of myrRSK on C4-2B4 tumor growthin bone, the C4-2B4/vector and C4-2B4/RSK cells were injectedinto femurs of SCID mice, and the tumor growth in bone wasmonitored by bioluminescence. The bioluminescence signalswere significantly higher in C4-2B4/RSK tumors compared withthose in C4-2B4/vector tumors (Fig. 2E). Quantification of thebioluminescence showed that expression of myrRSK led to more

Figure 3.Characterization of PC3-mm2 cells with knockdown of RSK1 and/or RSK2. A, PC3-shRSK1/2 double knockdown cells stably expressed shRNAs for both RSK1 andRSK2. To generate PC3-shRSK1/2 double knockdown cells, PC3-shRSK1 cells were infected with lentivirus from pGIPZ-shRSK2. The cells were selected by usingpuromycin and followed with FACS for expressing both Tomato (red) and GFP (green) fluorescence proteins for the expression of luciferase and shRSK2,respectively. B, Western blots of the PC3-mm2 cell lines that stably express the RSK1, RSK2, or RSK1/2 shRNAs with antibodies against RSK1, RSK2, total RSK, orpRSK-T359/S363. Vector-transfected (pLKO, pGIPZ) PC3-mm2 cells were used as controls. Cell lysates were used directly for Western blotting except forpRSK-T359/S363, in which cell lysates were first immunoprecipitated with anti-pRSK-T359/S363 antibody followed with Western blot with antibody against totalRSK. C, effects of serum concentrations on the proliferation of PC3-mm2 cells with knockdown of RSKs. D, anchorage-independent growth of PC3-mm2 cellswith knockdown of RSKs.






than a 10-fold increase in tumor growth compared with vector-transfected cells (Fig. 2F). Together, these results indicate thatincreased expression of activated RSK in C4-2B4 cells enhancestheir growth in bone, likely through an increase in cell survival.

Knockdown of RSK1 or RSK2 expression decreases anchorage-independent growth of PC3-mm2 cells

We further determined the effect of RSK knockdown on pros-tate cancer progression in bone. Four RSK isoforms, RSK1–4, havebeen reported inmammalian cells. Among them, RSK1 and RSK2have been studied more extensively. Thus, we determined theirroles in prostate cancer growth in bone. PC3-mm2 cells that werederived from prostate cancer bone metastasis and exhibit robusttumor growth in bone were selected for the study. LentiviralshRNAs were used to knockdown RSK1 or RSK2, individually ortogether. The lentiviral vector pLKO, with a puromycin selectionmarker, was used to knock down RSK1 and the lentivral vectorpGIPZ, with both puromycin and GFP selection markers, wasused to knock down RSK2. PC3-mm2 cells with knockdown ofboth RSK1 and RSK2 were generated by selecting for the expres-sion of shRSK1 through puromycin resistance followed withtransfection of pGIPZ-shRSK2 and selection for the expressionof shRSK2 through fluorescent activated cell sorting for GFP. As aresult, the PC3-mm2 RSK1/2 double knockdown cells wereresistant to puromycin (for shRSK1), and positive for GFP (forshRSK2) and Tomato (for luciferase; Fig. 3A). Western blottingusing antibodies specific to RSK1 or RSK2 was used to confirmknockdown of RSK1 or RSK2 in these cell lines. As expected, thelevels of RSK1 or RSK2 proteins were significantly reduced com-

pared with pLKO or pGIPZ vector-transfected cells (Fig. 3B). Todetermine the proportion of RSK1 or RSK2 in total RSK in PC3-mm2 cells, Western blot of total RSKwas performed on these RSKknockdown cells. As shown in Fig. 3B, RSK1 is the major RSK inPC3-mm2 cells as knockdown of RSK1 led to a significant reduc-tion (78%)of total RSK,whereas knockdownofRSK2only led to amodest (8%) decrease in total RSK. Western blotting with anti-pRSK antibodies showed that knockdown of RSK also resulted indecreases in the phosphorylation of RSK at T359/S363 (Fig. 3B).

We further examined the effect of RSKknockdownonPC3-mm2cell survival. Similar to C4-2B4 cells, we found that knockdown ofRSKs inPC3-mm2 cells hadno effect on cell proliferationwhen thecells were grown in medium with 10% serum (Fig. 3C, left).However, knockdown of RSK1 or both RSK1/2 led to a decreasein cell survival when cultured in 1% serum condition (Fig. 3C,right). Decreases in colony formation in soft agar were alsoobserved in PC3/shRSK1 and PC3/RSK1/2 cells compared withcontrols (Fig. 3D). The differences in the anchorage-independentgrowth of RSKs knockdownPC3-mm2 cells are not nearly asmuchas that observed in myrRSK-overexpressing C4-2B4 cells, suggest-ing that RSK1 is one of many factors that contribute to theanchorage-independent growth of PC3-mm2 cells.

Knockdown of RSK on PC3-mm2 tumor progression in boneNext, we examined the effects of RSKknockdownon the growth

of PC3-mm2 cells in bone. PC3-mm2 cells transducedwith vector(pLKO, pGIPZ) or with single or double RSK1 and RSK2 knock-down were injected into mouse femurs. Tumor growth wasmonitored weekly by BLI, and the tumor growth at 3 weeks was

Figure 4.Effect of RSK knockdown on PC3-mm2 growth in bone. A, PC3-mm2cells with knockdowns of RSK1, RSK2,or both RSK1/2 were injected in tomouse femurs. Control vector-transfected cells (pLKO and pGIPZ)were injected in parallel. BLI at 3weeks after inoculation is shown. B,semiquantification of tumor growthbased on signals from BLI. AverageBLI� SEM was shown. Knockdown ofRSK1 or both RSK1/2 resulted insignificant decreases in PC3-mm2tumor growth in bone. �, P < 0.05. C,mCT of tumor-bearing femurs. D, bonevolume of tumor-bearing bonedetermined by mCT analysis. Bonevolume within 11 mm from femur headwas calculated.

Yu et al.





shown in Fig. 4A. Quantification of bioluminescence signalsdemonstrated that knockdown of RSK1 in PC3-mm2 cells led toa significant inhibitionof their progression inbone,whereas RSK2knockdowndidnot (Fig. 4B). Knockdownof bothRSK1andRSK2generated a reduction similar to knockdown of RSK1 alone (Fig.4B). These results show that RSK1 is largely responsible for RSK-mediated PC3-mm2 progression in bone, likely due to the higherlevels of RSK1 in PC3-mm2 cells (Fig. 3B).

PC3-mm2 is known to induce strong osteolytic bone lesions.Thus, we also evaluated the effects of RSK1 and RSK2 on PC3-mm2–inducedbone changes. Analysis of tumor-bearing femurs byX-ray showed that PC3-pLKO and PC3-pGIPZ cells induced strongosteolytic lesions (Supplementary Fig. S2). Non–tumor-bearingfemurs do not show radiographic changes. Knockdown of RSK1 orboth RSK1 and RSK2 showed a reduction of the osteolytic lesioncompared with PC3 vector-injected femurs (Supplementary Fig.S2). Knockdown of RSK2 showed moderate reduction of theosteolytic response (Supplementary Fig. S2). To better define theeffect of RSK knockdown on tumor-induced bone changes, thefemurs were subjected to mCT analysis (Fig. 4C) and the bonevolumes determined (Fig. 4D). When compared with PC3-pLKOor PC3-pGIPZ–bearing femurs, tumor-bearing femurs with PC3-shRSK1or PC3-shRSK1/2 cells hadhigher bone volumes (Fig. 4D),likely due to the slower growth of PC3-shRSK1 or PC3-shRSK1/2cells in bone. These observations suggest that knockingdownRSK1or RSK1/2 in PC3-mm2 cells led to a decrease in tumor growth inbone accompanied by a decrease in bone destruction. Together,both overexpression and knockdown approaches support a role ofRSK in prostate cancer progression in bone.

Signal pathways regulating RSK-mediated survival in C4-2B4cells

We then examined signal pathways that regulate RSK-mediatedevents. RSKs have been shown to affect many cellular functionsthrough multiple signal pathways. As our data above indicatedRSK regulated the survival of prostate cancer cells, we focus onsignal pathways thatmay be involved in regulating RSK-mediatedsurvival. Jin and colleagues (18) reported that RSK2 mediatessurvival of head and neck cancer cells by both transcription-independent, e.g., phosphorylation of p38MAPK and SAPK/JNK,and transcription-dependent mechanisms, e.g., expression ofinhibitor of growth protein 3 (ING3), cytoskeleton-associatedprotein-2 (CKAP2), and protein-tyrosine kinase (PTK6/Brk).Thus, we examined whether these mechanisms are also involvedin RSK-mediated survival of prostate cancer cells. The effect ofmyrRSK expression on p38 MAPK and SAPK/JNK phosphoryla-tion was examined byWestern blot analysis. As shown in Fig. 5A,the phosphorylation of p38 MAPK and SAPK/JNK was decreasedsignificantly in C4-2B4/RSK cells compared with C4-2B4/vectorcells (Fig. 5A), suggesting that inhibition of these proapoptoticfactors may be involved in RSK-mediated survival in prostatecancer cells. The effects of RSK expression on ING3, CKAP2, andPTK6 expression were also examined. As shown in Fig. 5B–D (leftplots), expression of myrRSK in C4-2B4 cells led to decreases inthemessage levels of the tumor suppressors ING3 andCKAP2 andan increase in the message for PTK6, an intracellular nonreceptortyrosine kinase, as determined by RT-PCR. qRT-PCR furthershowed that expression of myrRSK in C4-2B4 cells led to 65%and 80% decreases in ING3 and CKAP2 messages, respectively,and a 40.0-fold increase in PTK6 message (Fig. 5B–D, middleplots). Western blot showed that the changes in the protein levels

correspond to the changes in the levels of messages (Fig. 5B–D,right plots).

ING3 is involved in RSK-mediated C4-2B4 cell survivalWe examinedwhether the decrease inmessage levels of ING3 is

involved in RSK-mediated anchorage-independent growth. ING3is a tumor suppressor that has been shown to be involved inapoptosis and cell cycle (19). To examine whether decrease ofING3 expression is involved in the RSK-mediated prostate cancercell survival, we overexpressed ING3 in C4-2B4/RSK cells using abicistronic retroviral vectorwith a neomycin selectionmarker. C4-2B4/vector cells transfected with empty neomycin vector were

Figure 5.Effects of RSK on the phosphorylation of p38-MAPK and SAPK/JNK and theexpression of PTK6, ING3, and CKAP2. A, RSK decreases the phosphorylationof p38-MAPK and SAPK/JNK. V, C4-2B4/vector; RSK, C4-2B4/RSK. B to D,RSK downregulates ING3, CKAP2, and upregulates PTK6 expression. Leftplots, RT-PCR. Middle plots, qRT-PCR. Right plots, Western blot. "�", <0.05.






used as a control. The C4-2B4/RSK/ING3 and control cells wereselected for neomycin resistance (for ING3 vector) as well as GFPpositivity (for myrRSK vector). As shown in Fig. 6A, expression ofING3 in C4-2B4/RSK cells restored ING3 to a slightly higher levelthan that in control C4-2B4/vector cells. C4-2B4/vector cellsoverexpressing ING3 (C4-2B4/vector/ING3) were also generatedto test the effect of ING3 on C4-2B4 cells. When these cells werecultured in medium containing low serum, i.e., 1% FBS, over-expression of ING3 decreased the cell number when cultured formore than four days (Fig. 6B). Expression of ING3 also abrogatedRSK-mediated anchorage-independent growth in C4-2B4/RSKcells (Fig. 6C). Overexpression of ING3 in C4-2B4/vector cellsdid not have significant effects on either assay (Fig. 6B and C).These results suggest that ING3 plays a role in RSK-mediated C4-2B4 cell survival.

Involvement of CKAP2 in RSK-mediated C4-2B4 cell survivalCKAP2 is a cytoplasmic protein associated with cytoskeletal

fibers (20). CKAP2 enhances wild-type (WT) p53 activity andtriggers G1 arrest and apoptosis in a p53-dependent manner (21).To examine whether CKAP2 is involved in the RSK-mediatedprostate cancer cell survival, we overexpressed CKAP2 in C4-2B4/RSK and C4-2B4/vector cells (Fig. 6D). When these cells werecultured in medium containing 0.1% FBS for three days, CKAP2overexpression decreased C4-2B4/RSK cell numbers (Fig. 6E).Overexpression of CKAP2 in control C4-2B4/vector cells alsodecreased cell numbers (Fig. 6E), consistent with the function ofCKAP2 in cell-cycle inhibition (21–23). Importantly, overexpres-sion of CKAP2 inhibits RSK-mediated anchorage-independentgrowth in soft agar assay (Fig. 6F). These observations suggest thatCKAP2 is also involved in the RSK-mediated prostate cancersurvival.

Involvement of PTK6 in RSK-mediated C4-2B4 cell survivalPTK6, which is upregulated by RSK activation in C4-2B4 cells

(Fig. 5D), has been shown tomediate a range of cellular processesrelated to the development or maintenance of malignancy (24,25). Irie and colleagues (26) reported that PTK6 regulates IGF-1–induced anchorage-independent survival of mammary epithelialcells. In addition, Harvey and colleagues (27) showed that PTK6/Brk enhances breast carcinoma cell survival in suspension. Theseobservations suggest that PTK6 may also play a role in RSK-mediated anchorage-independent growth. To test this, PTK6 inC4-2B4/RSK cells was knocked down by lentiviral vectors expres-sing PTK6 shRNAs or non-silencing (NS) shRNA. The C4-2B4/RSK/shPTK6 cells (clones #4 and #5) and C4-2B4/RSK/shNScontrol cells were selected for puromycin resistance (for shRNAvectors) as well as GFP positivity (for RSK vectors). Westernblotting using antibodies specific to PTK6 confirmed that C4-2B4/RSK/shPTK6-#4 and #5 clones have significantly reducedPTK6 levels compared with shNS vector-transfected cells (Fig.7A). In the survival assay with low serummedium, knockdown ofPTK6 slightly decreased the RSK-mediated survival in C4-2B4/RSK cells after culturing for more than five days (Fig. 7B). Thechanges in survival in low serummedium from PTK6 knockdownare not as striking as those observed with ING3 and CKAP2overexpression. However, knockdown of PTK6 significantlydecreased anchorage-independent growth compared with C4-2B4/RSK-shNS control cells (Fig. 7C). On the basis of theseobservations, we propose that RSK-mediated transcriptional reg-ulation of ING3, CKAP2, and PTK6 plays a role in RSK-mediatedC4-2B4 cell survival in vitro, and possibly C4-2B4 progression inbone in vivo (Fig. 7D).

DiscussionSurvival in the bone marrow microenvironment is one of

the key steps for the metastatic growth of prostate cancercells in bone. Our studies showed that RSK increases pros-tate cancer cell survival in vitro and progression in bone invivo. In addition, we showed that RSK phosphorylation isincreased in human prostate cancer bone metastasis speci-mens compared with those in primary site or lymph nodemetastasis. Together, our studies provide evidence that RSKis an important driver for the progression of prostate cancerin bone and suggest that RSK is a promising therapy targetfor prostate cancer bone metastasis.

Figure 6.Effects of ING3 and CKAP2 on RSK-mediated cell survival and anchorage-independent cell growth. A to C, overexpression of ING3 in C4-2B4/RSK cells.D to F, overexpression of CKAP2. A andD,Western blots for the expression ofING3 or CKAP2, respectively. B and E, cell numbers when cultured under lowserum condition. C and F, anchorage-independent growth. "�", <0.05.

Yu et al.





RSKsmediatedifferent cellular functions in a context-dependentmanner (1). RSKs are known to regulate cell-cycle progression (forreview, see ref. 2). Studies by Clark and colleagues (8) showed thatRSK inhibition decreases theproliferation of cancer cells, includingLNCaP andPC3prostate cancer cells andMCF-7breast cancer cells,but not normal breast epithelial cellsMCF-10A (8), suggesting thatcancer cells may exploit RSK-mediated pathways for proliferation.Although Clark and colleagues (8) showed that treatment ofLNCaP and PC3 cells with RSK inhibitor SL0101 led to inhibitionof prostate cancer cell proliferation, we found that overexpressionof myrRSK in C4-2B4 or knockdown of RSK1 in PC3-mm2 cellsmainly affected anchorage-independent growth, but not cell pro-liferation, when the cells were cultured under standard culturecondition. The differences between our studies and those of Clarkand colleagues (8) may be due to the difference in cell lines, i.e.,LNCaP versus C4-2B4, and the approach, i.e., inhibitor versus geneexpression, used in these studies.

RSK has been shown to exhibit transcriptional regulation ofgene expression. Indeed, we showed that in human prostatecancer bonemetastasis specimens, phosphorylated RSK is mainlylocalized in the nucleus, suggesting a possible role in generegulation. Studies by Chen and colleagues (16) showed thatnuclear RSKplays a role inmodulating c-Fosphosphorylation.Wefound that RSKs transcriptionally regulated the expression ofseveral apoptosis-related genes, including PTK6, ING3, andCKAP2 (18), in C4-2B4 prostate cancer cell line. As PTK6, ING3,and CKAP2 are CREB target genes (18), RSK may regulate thesegenes throughCREBphosphorylation. Interestingly, the functionsof ING3 and CKAP2 are p53 dependent, and C4-2B4 has WT p53(28). We found that knockdown of RSK1/2 in PC3-mm2 cells didnot affect the expression of ING3 or CKAP2 (data not shown),likely due to the lack of functional p53 in PC3-mm2 cells (28).How RSK regulates survival of PC3-mm2 cells is not known andrequires further analysis.

Figure 7.Effects of PTK6 on RSK-mediated cellsurvival and anchorage-independentcell growth. A, Western blot for theexpression of PTK6. B, cell numberswhen cultured under low serumcondition. C, knockdown of PTK6attenuates RSK-mediated anchorage-independent growth. D, proposedmodel of RSK-mediated C4-2B4 cellsurvival and progression in bone.Increase in RSK activity leads totranscriptional downregulation of ING3and CKAP2 and upregulation of PTK6inC4-2B4 cells. These lead to increasedsurvival of C4-2B4 cells in vitro andprogression in bonein vivo. "�", <0.05.






Our observation that RSKs play a role in prostate cancer bonemetastasis suggests that RSKs have potential to be a therapy targetfor prostate cancer bone metastasis. As Clark and colleagues (8)showed that RSK inhibition preferentially inhibits the prolifera-tion of both androgen-dependent LNCaP and androgen-indepen-dent PC3 prostate cancer cells, but not normal breast epithelialcells, these interesting observations also suggest that it is possibleto preferentially inhibit RSK activity in cancer cells. RSK has beenshown to be a druggable target. Three different classes of RSKinhibitors, SL0101 (29), FMK (30) and BI-D1870 (31), have beenidentified (32). However, a relatively high concentration, i.e., inmmol/L range, is required for these inhibitors to elicit an effect oncellular activity in vitro. Clark and colleagues (8) showed thatSL0101 at a concentration of 20 mmol/L inhibited LNCaP or PC3proliferation in vitro. Jin and colleagues (18) showed anoikissensitization of several cancer cell lines by using SL0101 andFMK at the concentrations of 100 mmol/L and 10 mmol/L, respec-tively. Similarly, a high concentration of BI-D1870 (�10 mmol/L)was also required for detecting its effect in cell lines (31). Inaddition, BI-D1870 also has effects on polo-kinase 1 activity (31),leading to failure of cells to undergo cytokinesis in BI-D1870–treated C4-2B4 cells (data not shown). Development of RSKinhibitors with improved affinity, biologic stability, and mem-brane permeability of SL0101 has been reported (33, 34). Thus, itis likely that a clinically applicable inhibitor for RSK will beavailable in the near future. Alternatively, it is possible to targetdownstreammediators of RSK. Further development of inhibitorsof RSK activity or RSK-mediated signal pathways will have poten-tial in the treatment of prostate cancer bone metastasis.

Metastatic prostate cancer cells in bonemay affect bonehomeo-stasis, resulting in osteoblastic or osteolytic responses. PC3 cellshave been shown to generate osteolytic bone lesions. Tumor cellscan modulate osteolytic program through various mechanisms.Hall and colleagues (35) showed that knockdown of DKK-1 inPC3 cells inhibited cell proliferation, through induction ofp21CIP-1/WAF-1, resulting in a reduction of osteolytic lesion inbone. In contrast, Dutta and colleagues (36) showed that integrinavb6 expression promotes an osteolytic program in PC3 cells by

upregulating MMP2, without affecting tumor growth in bone. Inthis study, we found that knockdown of RSK1 in PC3-mm2 cellsleads to a decrease in cell survival, resulting in a decrease ofosteolytic response. Thus, therapies targeting RSK may also helpin reducing tumor-associated bone lesions.

In conclusion, our studies identify a novel role of RSK inprostate cancer bone metastasis through both transcriptional andposttranslational modulation of pathways that regulate anchor-age-independent growth. Further development of strategies thatinhibit RSK-mediated survival in prostate cancer is warranted.

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

Authors' ContributionsConception and design: G. Yu, Y.-C. Lee, J.H. Song, G.E. Gallick, L.-Y. Yu-Lee,J. Kuang, S.-H. LinDevelopment of methodology: G. Yu, Y.-C. Lee, C.-F. Wu, J.H. Song, J. KuangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y.-C. Lee, C.-J. Cheng, C.-F.Wu, J.H. Song, G.E. GallickAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): G. Yu, Y.-C. Lee, C.-J. Cheng, C.-F. Wu, J.H. Song,G.E. Gallick, L.-Y. Yu-Lee, J. Kuang, S.-H. LinWriting, review, and/or revision of the manuscript: Y.-C. Lee, C.-J. Cheng,J.H. Song, G.E. Gallick, L.-Y. Yu-Lee, J. Kuang, S.-H. LinAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): G. Yu, Y.-C. Lee, S.-H. LinStudy supervision: S.-H. Lin

Grant SupportThisworkwas supported by grants from theNIHCA174798, P50CA140388,

CA16672, the Prostate Cancer Foundation, DOD (PC093132), DOD(PC080847), and the Cancer Prevention and Research Institute of Texas (CPRITRP110327).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received July 10, 2014; revised August 20, 2014; accepted August 24, 2014;published OnlineFirst September 4, 2014.

References1. Sulzmaier FJ, Ramos JW. RSK isoforms in cancer cell invasion and metas-

tasis. Cancer Res 2013;73:6099–105.2. Romeo Y, Zhang X, Roux PP. Regulation and function of the RSK family of

protein kinases. Biochem J 2012;441:553–69.3. Lara R, Seckl MJ, Pardo OE. The p90 RSK family members: common

functions and isoform specificity. Cancer Res 2013;73:5301–8.4. Anjum R, Blenis J. The RSK family of kinases: emerging roles in cellular

signalling. Nat Rev Mol Cell Biol 2008;9:747–58.5. BjorbaekC, ZhaoY,MollerDE.Divergent functional roles for p90rsk kinase

domains. J Biol Chem 1995;270:18848–52.6. Wang R, Jung SY, Wu CF, Qin J, Kobayashi R, Gallick GE, et al. Direct roles

of the signaling kinase RSK2 in Cdc25C activation during Xenopus oocytematuration. Proc Natl Acad Sci U S A 2010;107:19885–90.

7. Wu CF, Liu S, Lee YC, Wang R, Sun S, Yin F, et al. RSK promotes G2/Mtransition through activating phosphorylation of Cdc25A and Cdc25B.Oncogene 2014;33:2385–94.

8. Clark DE, Errington TM, Smith JA, FriersonHF Jr., WeberMJ, LanniganDA.The serine/threonine protein kinase, p90 ribosomal S6 kinase, is animportant regulator of prostate cancer cell proliferation. Cancer Res2005;65:3108–16.

9. Zhang X, Lavoie G, Fort L, Huttlin EL, Tcherkezian J, Galan JA, et al. Gab2phosphorylation by RSK inhibits Shp2 recruitment and cell motility. MolCell Biol 2013;33:1657–70.

10. Huang CF, Lira C, Chu K, Bilen MA, Lee YC, Ye X, et al. Cadherin-11increasesmigration and invasionof prostate cancer cells and enhances theirinteraction with osteoblasts. Cancer Res 2010;70:4580–9.

11. Chu K, Cheng CJ, Ye X, Lee YC, Zurita AJ, Chen DT, et al. Cadherin-11promotes the metastasis of prostate cancer cells to bone. Mol Cancer Res2008;6:1259–67.

12. Richards SA, Dreisbach VC, Murphy LO, Blenis J. Characterization ofregulatory events associated with membrane targeting of p90 ribosomalS6 kinase 1. Mol Cell Biol 2001;21:7470–80.

13. Giancotti FG, Ruoslahti E. Elevated levels of the alpha 5 beta 1 fibronectinreceptor suppress the transformed phenotype of Chinese hamster ovarycells. Cell 1990;60:849–59.

14. Dalby KN, Morrice N, Caudwell FB, Avruch J, Cohen P. Identification ofregulatory phosphorylation sites in mitogen-activated protein kinase(MAPK)-activated protein kinase-1a/p90rsk that are inducible by MAPK.J Biol Chem 1998;273:1496–505.

15. Zhao Y, Bjorbaek C, Weremowicz S, Morton CC, Moller DE. RSK3 encodesa novel pp90rsk isoform with a unique N-terminal sequence: growthfactor-stimulated kinase function and nuclear translocation. Mol Cell Biol1995;15:4353–63.

16. Chen RH, Abate C, Blenis J. Phosphorylation of the c-Fos transrepressiondomain by mitogen-activated protein kinase and 90-kDa ribosomal S6kinase. Proc Natl Acad Sci U S A 1993;90:10952–6.


Yu et al.




17. Thalmann GN, Sikes RA, Wu TT, Degeorges A, Chang SM, Ozen M, et al.LNCaP progression model of human prostate cancer: androgen-indepen-dence and osseous metastasis. Prostate 2000;44:91–103.

18. Jin L, LiD, Lee JS, Elf S, Alesi GN, Fan J, et al. p90 RSK2mediates antianoikissignals by both transcription-dependent and -independent mechanisms.Mol Cell Biol 2013;33:2574–85.

19. Nagashima M, Shiseki M, Pedeux RM, Okamura S, Kitahama-Shiseki M,Miura K, et al. A novel PHD-finger motif protein, p47ING3, modulatesp53-mediated transcription, cell cycle control, and apoptosis. Oncogene2003;22:343–50.

20. Maouche-Chretien L, Deleu N, Badoual C, Fraissignes P, Berger R, GaulardP, et al. Identification of a novel cDNA, encoding a cytoskeletal associatedprotein, differentially expressed in diffuse large B cell lymphomas. Onco-gene 1998;17:1245–51.

21. Tsuchihara K, Lapin V, Bakal C, Okada H, Brown L, Hirota-Tsuchihara M,et al. Ckap2 regulates aneuploidy, cell cycling, and cell death in a p53-dependent manner. Cancer Res 2005;65:6685–91.

22. Seki A, Fang G. CKAP2 is a spindle-associated protein degraded by APC/C-Cdh1 during mitotic exit. J Biol Chem 2007;282:15103–13.

23. Hong KU, Park YS, Seong YS, Kang D, Bae CD, Park J. Functional impor-tance of the anaphase-promoting complex-Cdh1-mediated degradation ofTMAP/CKAP2 in regulation of spindle function and cytokinesis. Mol CellBiol 2007;27:3667–81.

24. Ostrander JH, Daniel AR, Lange CA. Brk/PTK6 signaling in normal andcancer cell models. Curr Opin Pharmacol 2010;10:662–9.

25. Brauer PM, Tyner AL. Building a better understanding of the intracellulartyrosine kinase PTK6 - BRK by BRK. Biochim Biophys Acta 2010;1806:66–73.

26. Irie HY, Shrestha Y, Selfors LM, Frye F, Iida N, Wang Z, et al. PTK6 regulatesIGF-1-induced anchorage-independent survival. PLoS One 2010;5:e11729.

27. Harvey AJ, Pennington CJ, Porter S, Burmi RS, Edwards DR, Court W, et al.Brk protects breast cancer cells from autophagic cell death induced by lossof anchorage. Am J Pathol 2009;175:1226–34.

28. Carroll AG, Voeller HJ, Sugars L, Gelmann EP. p53 oncogene mutations inthree human prostate cancer cell lines. Prostate 1993;23:123–34.

29. Smith JA, Poteet-Smith CE, Xu Y, Errington TM, Hecht SM, Lannigan DA.Identification of the first specific inhibitor of p90 ribosomal S6 kinase(RSK) reveals an unexpected role for RSK in cancer cell proliferation.Cancer Res 2005;65:1027–34.

30. Cohen MS, Zhang C, Shokat KM, Taunton J. Structural bioinformatics-based design of selective, irreversible kinase inhibitors. Science 2005;308:1318–21.

31. Sapkota GP, Cummings L, Newell FS, Armstrong C, Bain J, Frodin M, et al.BI-D1870 is a specific inhibitor of the p90 RSK (ribosomal S6 kinase)isoforms in vitro and in vivo. Biochem J 2007;401:29–38.

32. Nguyen TL. Targeting RSK: an overview of small molecule inhibitors.Anticancer Agents Med Chem 2008;8:710–6.

33. Mrozowski RM, Vemula R, Wu B, Zhang Q, Schroeder BR, Hilinski MK,et al. Improving the affinity of SL0101 for RSKusing structure-baseddesign.ACS Med Chem Lett 2012;4:175–9.

34. Hilinski MK, Mrozowski RM, Clark DE, Lannigan DA. Analogs of the RSKinhibitor SL0101: optimization of in vitro biological stability. Bioorg MedChem Lett 2012;22:3244–7.

35. Hall CL, Zhang H, Baile S, Ljungman M, Kuhstoss S, Keller ET. p21CIP-1/WAF-1 induction is required to inhibit prostate cancer growth elicited bydeficient expression of the Wnt inhibitor Dickkopf-1. Cancer Res2010;70:9916–26.

36. Dutta A, Li J, Lu H, Akech J, Pratap J, Wang T, et al. Integrin alphavbeta6promotes an osteolytic program in cancer cells by upregulating MMP2.Cancer Res 2014;74:1598–608.






2015;13:348-357. Published OnlineFirst September 4, 2014.Mol Cancer Res Guoyu Yu, Yu-Chen Lee, Chien-Jui Cheng, et al. CKAP2, and PTK6-Mediated Cell SurvivalRSK Promotes Prostate Cancer Progression in Bone through ING3,

Updated version

10.1158/1541-7786.MCR-14-0384-Tdoi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2014/09/05/1541-7786.MCR-14-0384-T.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/13/2/348.full#ref-list-1

This article cites 36 articles, 22 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/2/348.full#related-urls

This article has been cited by 4 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://mcr.aacrjournals.org/content/13/2/348To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-14-0384-T

http://mcr.aacrjournals.org/content/suppl/2014/09/05/1541-7786.MCR-14-0384-T.DC1

http://mcr.aacrjournals.org/content/13/2/348.full#ref-list-1

http://mcr.aacrjournals.org/content/13/2/348.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/2/348


RSK Promotes Prostate Cancer Progression in Bone through ... · RSK expression increases...

Documents

Transcript of RSK Promotes Prostate Cancer Progression in Bone through ... · RSK expression increases...