Salt-Inducible Kinase 2 Regulates Mitotic Progression and Transcription in Prostate … · Cell Cycle and Senescence Salt-Inducible Kinase 2 Regulates Mitotic Progression and Transcription

Cell Cycle and Senescence

Salt-Inducible Kinase 2 Regulates MitoticProgression and Transcription in Prostate CancerH�el�ene Bon1, Karan Wadhwa1, Alexander Schreiner2, Michelle Osborne3,Thomas Carroll4, Antonio Ramos-Montoya1, Helen Ross-Adams1, Matthieu Visser5,Ralf Hoffmann6, Ahmed Ashour Ahmed7,8, David E. Neal1,9,10, and Ian G. Mills1,11,12,13

Abstract

Salt-inducible kinase 2 (SIK2) is amultifunctional kinase of theAMPK family that plays a role in CREB1-mediated gene transcrip-tion and was recently reported to have therapeutic potential inovarian cancer. The expression of this kinase was investigated inprostate cancer clinical specimens. Interestingly, auto-antibo-dies against SIK2 were increased in the plasma of patients withaggressive disease. Examination of SIK2 in prostate cancer cellsfound that it functions both as a positive regulator of cell-cycleprogression and a negative regulator of CREB1 activity. Knock-down of SIK2 inhibited cell growth, delayed cell-cycle progres-sion, induced cell death, and enhanced CREB1 activity. Expres-sion of a kinase-dead mutant of SIK2 also inhibited cell growth,induced cell death, and enhanced CREB1 activity. Treatmentwith a small-molecule SIK2 inhibitor (ARN-3236), currently inpreclinical development, also led to enhanced CREB1 activity

in a dose- and time-dependent manner. Because CREB1 is atranscription factor and proto-oncogene, it was posited that theeffects of SIK2 on cell proliferation and viability might bemediated by changes in gene expression. To test this, geneexpression array profiling was performed and while SIK2knockdown or overexpression of the kinase-dead mutant affect-ed established CREB1 target genes; the overlap with transcriptsregulated by forskolin (FSK), the adenylate cyclase/CREB1pathway activator, was incomplete.

Implications: This study demonstrates that targeting SIK2 genet-ically or therapeutically will have pleiotropic effects on cell-cycleprogression and transcription factor activation, which should beaccounted for when characterizing SIK2 inhibitors. Mol Cancer Res;13(4); 620–35. �2014 AACR.

IntroductionProstate cancer is the most common cancer diagnosed in men

and the second most common cause of cancer-related death afterlung cancer, with around 40,000 men diagnosed and 10,000deaths annually in the United Kingdom (1). If diagnosed early,it is generally successfully treated using radical prostatectomy.However, 10% to 15% of the patients are diagnosed after theircancer has spread and present with advanced or inoperable cancer(2). Because androgens have been shown to play an importantrole in the progression of prostate cancer, removing or blockingthe action of androgens using hormonal therapy, also referred toandrogen deprivation, is the treatment of choice for patients atthis stage. It usually results in a favorable clinical responsewith theinitial regression of at least 80% of hormone-sensitive prostatecancers. However, 10% to 20%ofmen eventually fail this therapybecause tumor cells acquire the capability to grow in the absenceof androgens, and the disease eventually recurs with fatal hor-mone refractory prostate cancer (HRPC; ref. 3). Consequently,new approaches are needed for the treatment of prostate cancer sobiomarkers for diagnosis and novel targets for therapeutic inter-vention are urgently required.

Salt-inducible kinase 2 (SIK2) is a serine/threonine kinase thatbelongs to the calcium calmodulin kinases (CaM) superfamilyand the AMP-activated protein kinases (AMPK) subfamily. Themembers of this family act as sensors of cellular energy changes,and are known to regulate physiologic processes that consume orgenerate ATP to maintain energy homeostasis in the cells (4).Several studies report that SIK2 is activated in refeeding fromstarvation andmodulates homeostasis to help the cells to adapt to

1Uro-Oncology Research Group, Cambridge Research Institute, Cam-bridge, United Kingdom. 2Microscopy and Imaging Core, CambridgeResearch Institute, Cambridge, United Kingdom. 3Genomics Core,Cambridge Research Institute, Cambridge, United Kingdom. 4Bioin-formatics Core, Cambridge Research Institute, Cambridge, UnitedKingdom. 5Health Care Innovation, Philips Research, Eidhoven, theNetherlands. 6MolecularDiagnostics, Philips Research, Eindhoven, theNetherlands. 7Weatherall Institute ofMolecularMedicine, UniversityofOxford, Oxford, United Kingdom. 8Nuffield Department of Obstetricsand Gynaecology, University of Oxford, Oxford, United Kingdom.9Department of Urology, Addenbrooke's Hospital, Cambridge, UnitedKingdom. 10Department of Oncology, University of Cambridge, Cam-bridge, United Kingdom. 11Department of Urology, Oslo UniversityHospital, Oslo, Norway. 12Department of Cancer Prevention, OsloUniversity Hospital, Oslo, Norway. 13Prostate Cancer Research Group,Centre for Molecular Medicine Norway, University of Oslo and OsloUniversity Hospital, Oslo, Norway.

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

D.E. Nea and I.G. Mills contributed equally to this article.

Note: The array data associated with the manuscript is currently deposited atNCBI for embargoed access by the reviewers: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE45711. The embargo will be lifted upon publication.

Current address for Thomas Carroll: MRC Clinical Sciences Centre, Faculty ofMedicine Imperial College London, Hammersmith Hospital Campus, Du CaneRoad, London, W12 0NN.

Current address for Antonio Ramos-Montoya: Oncology iMed, AstraZeneca,Alderley Park, Macclesfield, SK10 4TG, UK.

Corresponding Author: Ian G. Mills, University of Oslo and Oslo UniversityHospital, PO Box 1137, Oslo, N-0318, Norway. Phone: 47-22840767; Fax: 47-22840598; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-13-0182-T

�2014 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 13(4) April 2015620

on June 25, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst December 29, 2014; DOI: 10.1158/1541-7786.MCR-13-0182-T

http://mcr.aacrjournals.org/

metabolic stresses via the regulation of cAMP response element-binding protein (CREB1)–mediated gene transcription (5–11).CREB1 is a transcription factor that binds toDNAsequences calledcAMP response elements (CRE), thereby stimulating gene expres-sion once phosphorylated at Ser133 (12). The activity of CREB1 isenhanced by the transducers of regulated CREB1 activity (TORC)via a phosphorylation-independent interaction with the bZIPDNA-binding/dimerization domain of CREB1 (13). The TORCfamily is composed of TORC1, TORC2, and TORC3 and all threeTORC isoforms contain the putative SIK2 phosphorylation siteand are downstream targets of SIK2 in the regulation of CREB1-mediated gene expression (14).

Few studies have provided evidence of a role for SIK2 in cancer.SIK2 gene is located in the commonly deleted region of breastcancer at 11q23.1 (15–17). In contrast, this region is amplified indiffuse large B-cell lymphoma (DLBCL), which results inincreased SIK2 expression at the RNA and protein levels intumors, where SIK2 was found to regulate survival and glucosemetabolism, suggesting its role in DLBCL progression (18). Inaddition, SIK2 has recently been identified as a centrosomekinase controlling the G2–M phase transition in ovarian cancer.It localizes with the centrosomes where it phosphorylates thecentrosome linker c-Nap1 at Ser2392, resulting in its transloca-tion from the centrosomes to the cytoplasm at the onset ofmitosis to facilitate loss of centriole cohesion as an initial stepin centrosome separation (19).

In this study, we found high levels of auto-antibodies againstSIK2 in plasma of patients with prostate cancer. We show for thefirst time that SIK2 is present in prostate cancer cells, and that itsdepletion inhibits growth and induces cell-cycle arrest and apo-ptosis. Using immunoprecipitation, reporter assays, and cellularfractionation methods, we found that SIK2 interacts with theCREB1 regulator TORC1 and negatively regulates CREB1 activityby interfering with TORC1 nuclear translocation. Using a whole-genome profiling approach, we then identified the genes andnetworks regulated by SIK2.

Materials and MethodsProtein array profiling of plasma samples

The study was performed on a patient cohort comprising menwho have received prostatectomy and for which there was at least5 years follow-up and assembled as part of the ProtecT Study(Prostate testing for cancer and Treatment). The disease of menwith likely indolent cancer is characterized as follows: PSA <10 ng/mL; pStage < ¼ T2; prostate cancer volume < ¼ 0.5 mL,and no PSA recurrence >5 years follow-up, while the disease isclassified as likely aggressive disease if there has been clinical/PSArecurrence, or if in the postsurgery pathology pGleason >6, pStage>T2, or prostate cancer volume >0.5 mL (20). Plasma sampleswere analyzed using ProtoPlexTM immune response assays aspreviously described (21). All experimental work was performedat the service laboratory of the array supplier (Invitrogen).

Cell culture and reagentsLNCaP cells were obtained from the ATCC. C4-2 cells and

C4-2b were obtained fromMD Anderson Cancer Center (Hous-ton, TX0. LNCaP-Bic cells were a kind gift from Dr. Zoran Culig(General Hospital Feldkirch, Feldkirch Austria). DuCaP cellswere a kind gift from Philips Research (Eindhoven, the Nether-lands). SKOv3 cells were a kind gift from Dr. James Brenton

(University of Cambridge, Cambridge, United Kingdom).LNCaP, C4-2, C4-2b, and DuCaP cells were maintained at37�C in RPMI-1640 (Invitrogen) containing L-glutamine andsupplemented with 10% fetal bovine serum (FBS; Gibco) in ahumidified atmosphere supplied with 5%CO2. LNCaP-Bic cellswere cultured under the same conditions but medium wassupplemented with 1 mmol/L bicalutamide (Enzo Life Science).Cells were routinely subcultured at 1:4 using 0.25% Trypsin–EDTA (Invitrogen) when 80% to 90% confluency was reached.

Forskolin was obtained from Sigma-Aldrich. A SIK2 inhibitor,ARN-3236, was obtained from Arrien Pharmaceuticals withreported low nanomolar IC50 for SIK2 both in activity assays andcell-line experiments (22). SIK2 siRNA sequences were obtainedfrom Dharmacon (19). The nontargeting siRNA was obtainedfrom Dharmacon. The pCMV6-Entry-myc/flag construct wasobtained from Origene. The pCMV6-Entry-WT-SIK2-myc/flagand pCMV6-Entry-SIK2-EOS-KI-myc/flag were generated as pre-viously described (19).

Cell transfectionAMAXA nucleofection. For AMAXA nucleofection, 2 � 106 cellswere mixed with 100 mL of Nucleofector solution R (Lonza) and10 mg of DNA construct or 1 mmol/L of siRNA duplexes. Anelectrical current was then applied to the cells (Nucleofectorprogram T-09) to deliver the DNA or siRNAs into the nucleus.Cells were then transferred into culture dishes containing RPMI-1640 supplemented with 10% FBS.

Lipofectamine transfection. For each reaction, siRNA duplexes at40 nmol/L were mixed with 6 mL of Lipofectamine RNAi Max(Invitrogen) and 500 mL of Optimem (Gibco). Mixtures wereincubated for 20 minutes at room temperature and then dis-pensed in a6-well plate before additionof 200,000 cells in2mLofRPMI-1640 supplemented with 10% FBS.

DNA preparation. All the DNA constructs used in this study wereexpanded using the XL10-Gold ultracompetent Escherichia colicells (Agilent Technologies) and were purified using the HiSpeedPlasmid Midi Kit (Qiagen) according to the manufacturer'srecommendations.

Cell counting and cell viabilityCells were seeded in triplicate at a density of 300,000 cells per

well in a 6-well plate. At each time point, the supernatant washarvested to include dead or detached cells and live cells wereharvested using 0.25% Trypsin–EDTA (Invitrogen). Dead cellsand live cells were then pooled together, pelleted, resuspended in500 mL 1� PBS, and transferred to a vial for cell counting andestimation of cell viability using a Beckman Coulter Vi-Cell.

IncuCyte growth assaysCells were seeded in four replicates at a density of 20,000 cells

per well in a 48-well plate. Plates were placed in the IncuCyte andnine time-lapse images of each well were taken at 3-hour intervalsfor 7 days. IncuCyte 2010A software was used to assess changes incell confluence as a surrogate for change in cell number.

MTS Cell proliferation assayCells were seeded in four replicates at a density of 10,000 cells

perwell in a 96-well plate. At each timepoint, 20mLofCellTiter 96AQueous Assay reagent (Promega) was added directly to each well

SIK2 Functions in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 13(4) April 2015 621




with minimal exposure to light. Plates were incubated for 1 hourat 37�C, 5%CO2. Formazan absorption was measured at 490 nmusing an Infinite M200 spectrophotometer (Tecan). The meanabsorbance of wells was displayed as optical density to estimateproliferation status.

Soft agar colony formation assayCell were resuspended in DMEM (Cell Biolabs) supplemented

with 6% FBS and containing 0.4% agar. They were then seeded insix replicates at a density of 1,000 cells per well in a 96-well platecontaining a bottom layer ofDMEM supplementedwith 10%FBSand containing 0.6% agar. Cell-agar suspensionwas overlaidwithmedia containing 10% FBS and cultured for 7 days. After 7 days,the soft agar layer was solubilized, cells were lysed, and thenumber of colonies was determined using the CyQuant GR dyeand measure of fluorescence at 520 nm. To measure colonyformation of cells after transient knockdown, cells were trans-fected with siRNA, trypsinized 24 hours later, and 10,000 cellswere reseeded in soft agar as described above.

Cell-cycle analysisFor DNA content analysis, cells were seeded in triplicate at a

density of 300,000 cells per well in a 6-well plate and were grownfor 48hours or 72hours. At each time point, cells were trypsinizedusing 0.25%Trypsin–EDTA (Invitrogen), were washed in 1� PBS,and were fixed with 1% paraformaldehyde (Electron MicroscopyScience) for 1 hour at 4�C. Cells were then washed in cold 1� PBS(Gibco), resuspended in 80% ice-cold methanol, and stored at�20�C until staining. Methanol-fixed cells were treated with 3mmol/L DAPI (Sigma-Aldrich) overnight at 4�C. Fluorescence-activated cell sorting (FACS) analysis was carried out using a BDLSRII instrument (Becton & Dickinson) and data acquisition wasperformed using BD FACSDiva software (v.5.0.3.). The fluores-cence emitted byDAPIwas collected using aUV-450/50 bandpassfilter. Data were analyzed after doublet discrimination (23) usingthe FlowJo software (TreeStar, v.8.8.4.) and applying the curve-fitting algorithm contained in the software.

Annexin V apoptosis assayCells were seeded in triplicate at a density of 300,000 cells per

well in a 6-well plate. At each time point, the supernatant washarvested to include dead or detached cells and live cells wereharvested using 0.25% Trypsin–EDTA (Invitrogen). Dead cellsand live cells were then pooled together, washed in 1� PBS,resuspended in Annexin V binding buffer (BioLegend), andstained with 5 mg/mL propidium iodide (PI; Sigma-Aldrich) andAnnexin V–Alexa Fluor 647 (BioLegend) for 15 minutes. FACSanalysis was carried out using a BD LSRII instrument (Becton &Dickinson) and data acquisition was performed using BD FACS-Diva software (v.5.0.3.). The fluorescence emitted by PI and AlexaFluor 647 was collected using a Blue-575/26 and a Red-660/20bandpass filters. Data were analyzed after doublet discrimination(23) using the FlowJo software (TreeStar, v.8.8.4.).

ImmunofluoresenceCells were seeded at a density of 50,000 cells per well in a 24-

well plate with coverslips. After at least 48 hours to allow adher-ence, cells were washed in 1� PBS and were fixed with 4%paraformaldehyde (Sigma-Aldrich) for 3minutes. Cells were thenwashed in cold 1� PBS (Gibco), treated with 80% ice-cold

ethanol, and stored at �20�C until staining. For staining, cellswere washed in 1� TBS-0.2% Triton–0.04% SDS, blocked for 30minutes with 1� TBS supplemented with 1.5% bovine serumalbumin (BSA; Sigma-Aldrich) and incubated for 1 hour with aSIK2 (BioLegend) or g-tubulin (Sigma-Aldrich) primary antibodyfollowed by an appropriate Alexa Fluor 488- or Alexa Fluor 594–conjugated secondary antibody (Invitrogen). Cover slips werethen mounted onto slides with Vectashield solution containingDAPI (Vectashield). Images were taken using a Leica Tandemconfocal microscope.

Time-lapse imagingCells were seeded at a density of 40,000 cells per well in a

15 m-Slide 8-well plate (Ibidi). After 36 hours, the plate wasplaced in an incubator in a humidified atmosphere supplied with5% CO2 connected to a Nikon Eclipse TE 2000 microscope(Nikon Instruments Europe). Images were obtained using 40�lens and were collected every 5 minutes for 48 hours from fourdifferent positions per well using NIS-Elements AR 3.2 software(NIS Elements). The time from the start of prometaphase, asevidence by the cells detaching and rounding-up, to the endof cytokinesis, as evidence by the generation of two daughtercells, was scored for individual cells. The exact start time ofprometaphase for each cell was recorded to avoid duplication ofmeasurement.

Preparation of total cell extractsCells were washed, scraped in ice-cold 1� PBS, pelleted, and

resuspended in lysis buffer containing 20 mmol/L HEPES(pH 7.5), 150 mmol/L sodium chloride, 1% NP-40, 0.25%sodiumdeoxycholate, 10% glycerol, 40mmol/L sodium fluoride,12 mmol/L b-glycerophosphate, 1 mmol/L sodium orthovana-date, 0.5 mmol/L EDTA, 2.5 mmol/L EGTA, a protease inhibitorcocktail (Calbiochem), and a phosphatase inhibitor mix (Sigma-Aldrich). The extractswere cleared by centrifugation (8,000� g for10 minutes at 4�C) and supernatants were collected. Proteinquantification was performed using the Quant-iT Protein Assayaccording to the manufacturer's recommendations (Invitrogen).Total cell lysates were boiled at 70�C in 5� SDS sample bufferfor 10 minutes and immunoblot analyses were performed asdescribed below.

For treatment of the extracts with a phosphatase, the sameprocedure was followed but the extracts were prepared inNEBuffer 3 (New England Biolabs) and were sonicated for90 seconds before being treated with alkaline phosphatasecalf intestinal (New England Biolabs) at 1 U/mg of protein for2 hours at 37�C.

Preparation of cellular and nuclear extractsCells were washed, scraped in ice-cold 1� PBS, pelleted, and

resuspended for exactly 10 minutes in lysis buffer containing10 mmol/L HEPES (pH 7.9), 10 mmol/L potassium chloride,1.5 mmol/L magnesium chloride, 0.34 mol/L sucrose, 10%glycerol, 1 mmol/L dithiothreitol, 0.1% Triton, a protease inhib-itor cocktail (Calbiochem), and a phosphatase inhibitor mix(Sigma-Aldrich). The extracts were cleared by centrifugation(1,300 � g for 4 minutes at 4�C) and the cytoplasmic fractionswere collected and cleared by centrifugation (20,000 � g for15 minutes at 4�C). The nuclei were washed twice in lysis buffercontaining 10 mmol/L HEPES (pH 7.9), 10 mmol/L potassiumchloride, 1.5 mmol/L magnesium chloride, 0.34 mol/L sucrose,

Bon et al.

Mol Cancer Res; 13(4) April 2015 Molecular Cancer Research622




10% glycerol, 0.1% Triton, a protease inhibitor cocktail (Calbio-chem), and a phosphatase inhibitor mix (Sigma-Aldrich). Theywere then resuspended in lysis buffer containing 3mmol/L EDTA,0.2 mmol/L EGTA, 1 mmol/L dithiothreitol, 0.1% Triton, aprotease inhibitor cocktail (Calbiochem), and a phosphataseinhibitor mix (Sigma-Aldrich), sonicated for 90 seconds, andcleared by centrifugation (20,000 � g for 15 minutes at 4�C) toobtain the nuclear fractions. Protein quantification was per-formed using the Quant-iT Protein Assay according to the man-ufacturer's recommendations (Invitrogen). Cellular and nuclearextracts were boiled at 70�C in 5� SDS sample buffer for 10minutes and immunoblot analyses were performed as describedbelow.

ImmunoprecipitationTen micrograms of SIK2 (Cell Signaling Technology) or IgG

Rabbit (Vector) antibody was incubated with 50 mL of Protein ADynabeads (Novex by Life Technologies) for 4 hours at 4�C andsubsequently cross-linked with 20 mmol/L dimethyl pimelimi-date in 0.2 mol/L triethanolamine pH 8.2 at room temperature.For each condition, 75 million cells were lysed and cellular andnuclear extracts were prepared as indicated in the sections above.Each fraction was then incubated with the antibody-bound beadsovernight at 4�Cwith rotational mixing. Beads were washed threetimes in 1%RIPA and then placed in 1� Laemli buffer and heatedat 70�C for 10 minutes. Beads were removed on a magnet andimmunoblot analyses on the IP products were performed asdescribed below.

Western blottingImmunoblot analyses were performed using a range between

20 and 100 mg of lysates depending on the abundance of theprotein of interest. The material was electrophoretically resolvedon denaturing sodium dodecyl sulphate–polyacrylamide gelelectrophoresis (SDS-PAGE) gels ranging from 6% to 16% acryl-amide and was transferred to nitrocellulose Immobilon-P mem-branes (Invitrogen) using the iBlot Dry Blotting System (Invitro-gen).Membranes were then blocked in 5%w/vMarvelmilk or 5%BSA (Sigma-Aldrich) w/v in 1� TBS–0.1% Tween for 30 minutesat room temperature and were immunoblotted with b-actin(Abcam), b-tubulin (Abcam), Bak (Cell Signaling Technology),Bax (Cell Signaling Technology), Bid (Cell Signaling Technol-ogy), CREB1 (Cell Signaling Technology), pCREB1 (Cell Sig-naling Technology), cyclin D1 (Cell Signaling Technology),histone H3 (Abcam), p21 (Cell Signaling Technology), p27(Cell Signaling Technology), PARP (Cell Signaling Technolo-gy), SIK2 (BioLegend, Sigma-Aldrich or Cell Signaling Tech-nology), TORC1 (Cell Signaling Technology), TORC2 (CellSignaling Technology), and TORC3 (Cell Signaling Technolo-gy) primary antibodies overnight at 4�C. Membranes were thenwashed five times in 1� TBS–0.1% Tween and were incubatedwith an appropriated horseradish peroxidase (HRP)–conjugat-ed secondary antibody (Dako) for 1 hour at room temperature.Membranes were then washed five times in 1� TBS–0.1%Tween and proteins were visualized using ECL Plus WesternBlotting Detection System (GE Healthcare).

RNA extraction and cDNA synthesisCells were washed, scraped in ice-cold 1� PBS, pelleted, and

resuspended in 1-mL TRizol reagent (Sigma-Aldrich). Sampleswere incubated for 5 minutes at room temperature, mixed with

200 mL of chloroform, and cleared by centrifugation (12,000 � gfor 15 minutes at 4�C). The aqueous upper phase was saved andincubated with 500 mL isopropanol for 10 minutes at roomtemperature. Samples were centrifuged (12,000 � g for 15 min-utes at 4�C), supernatants were removed, and pellets were washedwith 1 mL 75% ethanol. Ethanol was discarded after centrifuga-tion and pellets were resuspended in RNase-free H2O. RNAquantification was assessed using NanoDrop 1000 (ThermoScientific). One microgram of total RNA was reverse transcribedusing the High Capacity RNA-to-cDNA MasterMix (Applied Bio-systems) according to the manufacturer's recommendations.

Quantitative real-time PCRReactions were performed in triplicate using 5 mL of FAST SYBR

GreenPCRMaster Mix (Applied Biosystems), 2 pmol of primers(Supplementary Table S1), and 1 mL of cDNA in a total volume of10mL. The cycling conditionswere 20 seconds at 95�C, 40 cycles of1 second at 95�C, and 20 seconds at 60�C. All reactions wereperformed on the ABI PRISM 7900 HT Sequence DetectionSystem (Applied Biosystems) and data were analyzed withApplied Biosystems Sequence Detection Software (v.2.3.).

Relative expression levels were calculated using the followingformula:

(efficiency of test PCR(control sample Ct � test sample Ct))/(efficiency of control PCR(control sample Ct � test sample Ct))

with PCR efficiency ¼ 10(1/-standard curve slope) ¼ 2

Ct represents the number of cycles required before an arbitrarythreshold is reached. Test PCR represents the gene of interest.Control PCR represents the control genes. Test sample representsthe sample of interest (e.g., cell line or condition). Control samplerepresents the reference sample (e.g., parental cell line or cells att ¼ 0).

Data were normalized to housekeeping genes b-actin andsuccinate dehydrogenase, and the ratio between the sample ofinterest and the reference sample was calculated.

Illumina HumanHT-12 v4 Expression BeadChipCells were washed, scraped in ice-cold 1� PBS, pelleted, and

resuspended in 1 mL of TRI reagent solution (Ambion). RNAswere extracted using RiboPure Kit according to themanufacturer'sinstructions (Ambion). RNA quantification was assessed usingNanoDrop 1000 (Thermo Scientific) and RNA integrity bymicro-analysis (Agilent Bioanalyzer). RNAs were diluted to 22.7 ng/mLand 250 ng were used as the input for cRNA conversion thatincludes amplification and biotin labeling, using the IlluminaTotalPrep-96 Kit (Ambion). Purified, quality-controlled (Bioana-lyzer and Nanodrop spectrophotometer), and normalized cRNAwas hybridized to arrays according to the Illumina protocol(Illumina, WGGX DirectHyb Assay Guide 11286331 RevA). Rawimage files were processed and analyzed using the BeadArraypackage from Bioconductor (24).

Reporter assaysFor reporter assays, cells were transfected with 2 mg of CRE(1)

reporter construct (Panomics) using Amaxa nucleofection asdescribed above. Cells were then seeded in 12 replicates at adensity of 150,000 cells per well in a 24-well plate. At each timepoint, cells were scraped directly in culture medium, transferredinto a 96-well plate, pelleted, washed in ice-cold 1� PBS, and






resuspended in20mL 1� lysis buffer (Promega). Lysateswere thentransferred into a dark 96-well plate and luciferase assays wereperformed using a Dual-Luciferase Reporter Assay System (Pro-mega) and an automatic injection luminometer instrument(PHERAstar FS, BMG Labtech). Hundred microliters of Dual-GloLuciferase reagent was injected to each well, mixed, and Fireflyluciferase activity wasmeasured 10minutes later. Then, 100 mL ofDual-Glo Stop&Glo reagent was added to each well, mixed, andRenilla luciferase activity was measured 10 minutes later. Activitywas determined by calculating the ratio of Firefly luciferase toRenilla luciferase for each well and replicates were averaged.

ResultsSIK2 is found in plasma and is a marker for prostate cancer

To identify novel biomarkers for prostate cancer in biologicfluids, we conducted a study using protein arrays to detect theimmune reactivity of plasma samples obtained from 21patients with indolent disease and 23 with aggressive disease(21). Auto-antibody responses were able to discriminatebetween indolent and aggressive cases and identified SIK2 asoverexpressed in patients with aggressive disease comparedwith patients with indolent disease (P ¼ 2.32E-04; Fig. 1A).The same samples were used to generate a receiver operatingcharacteristic (ROC) curve (25) for SIK2 in plasma and con-firmed that SIK2 auto-antibodies could be a good blood markerfor prostate cancer (Fig. 1B).

SIK2 knockdown and kinase-dead SIK2 reduce proliferationand viability of LNCaP cells

To assess the role of SIK2 in prostate cancer, we transientlyoverexpressed a construct containing the wild-type sequence ofthe kinase (pCMV6-Entry-SIK2-myc/flag) and the correspondingconstruct containing the kinase-dead sequence (pCMV6-Entry-SIK2-EOS-KI-myc/flag) in LNCaP cells (Fig. 2A). The N-terminalof SIK2 contains the serine/threonine kinase domain with lysine49 (K49) located at the putative ATP-binding site that is essentialfor SIK2 kinase activity. The SIK2-mutant construct used had thislysine replaced with a methionine (K49M), so the activity of thekinase was abolished (26).

Cell proliferation and viability assays were then conductedusing a Beckman Coulter Vi-Cell and Trypan Blue staining.

Reduced cell growth could be observed after 4 and 7 days ofgrowth when the kinase-dead construct was overexpressed inthe LNCaP cell line compared with the wild-type and emptyvectors (Fig. 2B). In addition, LNCaP cells overexpressing thekinase-dead mutant had reduced cell viability comparedwith the empty vector and wild-type SIK2 (75.6%, 91.3%, and92.0% viable cells, respectively, after 4 days of growth; Fig. 2C).To compensate for limitations on cell numbers and variabletransfection efficiencies following transient transfection, wewent on to generate LNCaP cells stably overexpressing the sameconstructs and confirmed these phenotypic effects (Supplemen-tary Fig. S1). However, the very low proliferation rate of thekinase-dead overexpressing stable cell line resulted in thedeselection of the cells overexpressing the construct over time(data now shown).

We then used multiple siRNA sequences (19) to transientlyknockdown SIK2 in LNCaP cells. Cell proliferation and viabilityafter 4 and 7 days of knockdownweremeasured using a BeckmanCoulter Vi-Cell and Trypan Blue staining. SIK2 knockdowninduced a significant reduction in cell proliferation and cellviability with up to 70% cell death after 7 days of growth. Themagnitude of inhibition of cell proliferation and viability corre-latedwith that of SIK2depletion. The cytotoxic effectwas themostpronounced for siRNA C, which was also the siRNA which gavethe most marked phenotype in ovarian cells and vice versa forsiRNA A (Fig. 2B and C; ref. 19). The phenotypic effects observedafter SIK2 knockdown in LNCaP cells were also observed usingothermethods to assess cell proliferation (Supplementary Fig. S1)and in a selection of other model prostate cancer cell lines(Supplementary Fig. S2).

SIK2 knockdown induces apoptosis in LNCaP cellsThe significant reduction in cell viability observed after knock-

down of SIK2 in LNCaP cells prompted us to consider whether ithas a role in apoptosis.We carried out Annexin V assays on LNCaPcells transfected with all three SIK2-siRNA sets (A, B, and C)and the nontargeting siRNA (NT). PI and Annexin V stainingswere performed on the cells at day 2, 3, 4, and 7 after knockdown,and the percentages of viable, necrotic, and apoptotic cells wereevaluated using flow cytomery and an appropriate gatingstrategy. Figure 3A shows the percentage of apoptotic cells alongthe time course following transfection with all three SIK2 siRNAs

A

AggressiveIndolent

SIK2

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

B SIK2

6050403020100

AUC = 0.8P < 0.0001

100908070

100−SpecitivityS

ensi

tivity

0

10

20

30

40

50

60

70

80

90

100Wilcoxon P = 2.32 × 10-4

Figure 1.SIK2 autoantibodies are found inplasma and are a blood marker forprostate cancer. A, SIK2autoantibodies were detected usingprotein arrays (ProtoPlexTM;Invitrogen) in plasma samples of twopatient cohorts comprising 21 patientswith indolent prostate cancerand 23 patients with aggressiveprostate cancer (n ¼ 44; n ¼ 23 inpositive group and n ¼ 21 in thenegative group; refs. 20, 21). B, thesame samples were used to generatea ROC curve (25) for plasma SIK2.Area under the curve (AUC) for SIK2is 0.8 with a P value of <0.0001.

Bon et al.





and the nontargeting siRNA. Significant differences were observedwhen the cells were transfected with all three SIK2-siRNAs, show-ing a clear induction of apoptosis after SIK2 knockdown (up to60%with siRNAs B and C, and 40%with siRNA A). Interestingly,when the cells were transfected with siRNA C, apoptosis occurredfaster, with 40% of cells apoptotic only 2 days after knockdown.Representative FACS profiles for the nontargeting siRNA andSIK2-siRNA C are shown in Fig. 3B.

To gain further insight into the mechanisms involved, wethen looked at the levels of the key apoptosis regulators Bid,Bax, Bak, caspase-3, and PARP in the LNCaP cell line knockeddown for SIK2 (Fig. 3C). SIK2 immunoblotting confirmed theefficiency of the knockdown with all three siRNAs and the equalamount of b-actin in all cell extracts confirmed that cellularintegrity was unaffected. Bid levels were found to be decreasedafter SIK2 knockdown, which suggested it had been cleavedinto its truncated form tBid, although the lower molecularweight band could not be detected (15 kDa). Bax and Badlevels were increased after knockdown with all three siRNAs,and a clear reduction in the levels of caspase-3 and PARPaccompanied by the appearance of lower molecular weightbands (17 and 89 kDa, respectively) were observed afterknockdown with SIK2-siRNA C, which suggests that caspase-

3 and PARP were cleaved. These results indicate that SIK2knockdown induces apoptosis in LNCaP cells via a mechanisminvolving the proapoptotic proteins Bid, Bax, and Bak, and thesubsequent cleavage of caspase-3 and PARP. Again, the effectsseen were the most apparent with siRNA C and least apparentwith the less efficient siRNA A.

SIK2 knockdown induces a cell-cycle arrest in G1-phase anddelays mitotic progression of LNCaP cells

Following on from Ahmed and colleagues (19) showing thatSIK2 is a centrosome kinase required for cell-cycle progressionfrom G1 to S and for the initiation of mitosis, we questionedwhether the effects observed on cell proliferation after SIK2knockdown in LNCaP cells were a consequence of its implicationin cell-cycle regulation.

The population distribution in G1, S, and G2–M phases of thecell cycle 48 and 72 hours after SIK2 knockdown in LNCaP cellswas assessed using flow cytometry andDAPI staining. There was asignificant increase in the proportion of cells in G1-phase whenSIK2 was knocked down with all three siRNAs versus the non-targeting siRNA control (68.9%, 69.9%, and 75.3% cells in G1 forsiRNAs A, B, and C, respectively, versus 65.5% for NT-siRNA at48 hours; Fig. 4A and B). This indicated that SIK2 inhibition

b-Actin

SIK2

Day 4 Day 7

NT A B C NT A B C

38

MW (kDa)

102

EV WT

SIK2

b-Actin

KI EV WT KI

Day 4 Day 7

38

MW (kDa)

102

A

B

C

Cell proliferation

KIEVWTKIEVWTKIEVWT0.0

0.5

1.0

1.5

Day 0

**

**N

um

ber

of

viab

le c

ells

(x1

06)

Day 4 Day 7

Cell proliferation

CBANTCBANT0.0

0.2

0.4

0.6

0.8

1.0

Day

0

Day 4

***

***

******

***

Nu

mb

er o

f vi

able

cel

ls (

x10

6)

Day 7

Cell viability

KIEVWTKIEVWTKIEVWT0

20

40

60

80

100 **

Via

ble

cel

ls (

%)

Day 0 Day 4 Day 7

Cell viability

CBANTCBANT0

20

40

60

80

100

* ***

***

**

***

***V

iab

le c

ells

(%

)

Day

0

Day 4 Day 7

Figure 2.Kinase-dead SIK2 and SIK2knockdown reduce proliferation andviability of LNCaP cells. A, SIK2overexpression and knockdown wereconfirmed by Western blot analysis atdays 4 and 7 after transfection with awild-type SIK2-Myc construct, akinase-dead SIK2-Myc construct (KI)or the corresponding empty-vectorconstruct (EV), and after transfectionwith the control nontargeting siRNA(NT) or SIK2-siRNAs (A–C). Celllysates were resolved by SDS-PAGE,transferred to nitrocellulosemembrane and immunoblotted forSIK2 (Sigma-Aldrich). b-Actin(Abcam) was used as a loadingcontrol. B, cell proliferation of LNCaPcells overexpressing wild-type orkinase-dead SIK2 or knocked down forSIK2 was assessed by counting thecells using a Beckman Coulter Vi-Celland Trypan Blue staining. The numberof viable cells at days 0, 4, and 7 isshown (n ¼ 3; error bars representmean � SEM; � , P < 0.05; �� , P < 0.01;�� , P < 0.001). C, cell viability wasassessed with a Beckman CoulterVi-Cell and Trypan Blue staining. Thepercentage of viable cells at days 0, 4,and 7 is shown (n ¼ 3; error barsrepresent mean � SEM; �, P < 0.05;�� , P < 0.01; �� , P < 0.001).






efficiently inhibited the progression of the cells through the cellcycle by arresting them in G1-phase. Consistent with this pheno-type was an increase in p21 expression with all three siRNAs (Fig.4C). There was also a decrease in cyclinD1 level with siRNAC andan increase in p27 was observed after knockdown with siRNA A,reflecting the cells inability to exit G1-phase after SIK2 knockdown(Supplementary Fig. S3). These data support the phenotypeobserved using flow cytometry, indicating that SIK2 knockdowninduces a cell-cycle arrest in G1 via the upregulation of p21 andp27 and the downregulation of cyclin D1.

A G1 cell-cycle arrest can be a consequence of the inabilityof the cells to enter mitosis and then escaping into G1. Given

that SIK2 has been shown to play a key role in the initiation ofthe G2–M phase transition by localizing at the centrosomes inovarian cancer cells (19), we decided to test whether this wasalso the case in prostate cancer cells. g-Tubulin is a proteinmember of the microtubule organizing center (MTOC), whichplays a role in the nucleation and polar orientation of micro-tubules. It localizes at the centrosome and is commonly used as acentrosomemarker (27). Therefore, we used SIK2 and g-Tubulincostainings and immunofluorescence to look at a potentialcolocalization between the kinase and the centrosomes in ourcells. LNCaP cells were stained using antibodies for SIK2 andg-Tubulin. Staining was also performed on the ovarian cancer

0 102 103 104 105

red 660/20-A

0

102

103

104

105

blue

575

/26-

A

0.836 5.7

70.722.8

0 102 103 104 105

red 660/20-A

0

102

103

104

105

blue

57

5/2

6-A

3.41 1.43

59.235.9

Day 7Day 4Day 3Day 2

0 102 103 104 105

red 660/20-A

0

102

103

104

105

blu

e 5

75

/26

-A

0

102

103

104

105

blu

e 5

75

/26

-A

0.158 6.54

22.870.6

0.668 19 0.516 23

679.4943.237.2

0 102 103 104 105

red 660/20-A

0 102 103 104 105

red 660/20-A0 102 103 104 105

red 660/20-A

0

102

103

104

105

blu

e 5

75

/26

-A

0

102

103

104

105

blu

e 5

75

/26

-A

0.0732 3.92

17.178.9

0 102 103 104 105

red 660/20-A

0

102

103

104

105

blue

575

/26

-A

1.07 7.74

24.966.3

0 102 103 104 105

red 660/20-A

0

102

103

104

105

blu

e 5

75

/26

-A

0.455 2.15

6.4990.9

NT-

siR

NA

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

SIK

2-si

RN

A C

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

Necrotic

ApoptoticViable

b-Actin

SIK2

Bax

PARP

Caspase-3

38

MW (kDa)

102102

17

17

31

Bak

Bid17

NT A B C NT A B C

24

A

B

CApoptosis

NTA B C NTA B C NTA B C NTA B C0

20

40

60

80

100

Day 2

******

***

***

******

***

******

Ap

op

toti

c ce

lls (

%)

Day 3 Day 4 Day 7

Figure 3.SIK2 knockdown induces apoptosis in LNCaP cells. A, apoptosis of LNCaP cells knocked down for SIK2 was assessed using flow cytometry by simultaneouslystaining the cells with an Annexin V antibody (BioLegend) and the DNA dye PI (Sigma-Aldrich). The percentage of apoptotic cells at days 2, 3, 4, and 7 aftertransfection with the nontargeting siRNA (NT) and SIK2-siRNAs (A–C) is shown (n ¼ 3; error bars represent mean � SD; ��, P < 0.01; ��, P < 0.001).B, dot-plots are representative of the gating strategy used on LNCaP cells transfected with the control nontargeting siRNA (NT) and SIK2-siRNA C. Stainingallows the viable cells (double negative), the apoptotic cells (Annexin V positive), and the necrotic cells (Annexin V and PI double positive) to be distinguished.C, Bid, Bak, Bax, PARP, and caspase-3 protein expression was assessed by Western blot analysis 4 days after transfection of LNCaP cells with the controlnontargeting siRNA (NT) and SIK2-siRNAs (A–C). Cell lysateswere resolved by SDS-PAGE, transferred to nitrocellulosemembrane, and immunoblotted for Bid, Bak,Bax, PARP, and caspase-3 (Cell Signaling Technology). SIK2 levels after knockdown are also shown (Sigma-Aldrich). b-Actin (Abcam) was used as a loadingcontrol. Cleaved PARP and caspase-3 are indicated with arrows.

Bon et al.





cell line SKOv3 for a positive control. Images are shown inSupplementary Fig. S4A. We could reproduce the data publishedby Ahmed and colleagues (19) showing a clear colocalizationof SIK2 at the centrosomes in ovarian cancer cells in interphase

and in mitosis. We also observed colocalization between SIK2and g-tubulin in LNCaP cells, providing some evidence ofSIK2 colocalization at the centrosomes and a potential role inthe regulation of mitosis in prostate cancer cells. This

050K

100K150K

200K250KUV B-A

050K

100K150K

200K250KUV B-A

NT-siRNASIK2-siRNA ASIK2-siRNA BSIK2-siRNA C

48 h 72 h

A

B

C D B

SIK2

p21

b-Actin

MW (kDa)NT A C

38

102

17

NT A B C

0

20

40

60

80

100

**

**

*

***

***

*

***

***

*

Cell cycle48 h

Pro

po

rtio

n o

f to

tal c

ell n

um

ber

(%

)

NT A B C

0

20

40

60

80

100

**

**

* *

*

**

*

*** *

**

Cell cycle72 h

Pro

po

rtio

n o

f to

tal c

ell n

um

ber

(%

)

%G1

%S%G2–M

Time lapse

Mea

n m

ito

tic

pro

gre

ssio

n t

ime

(min

)

NT A B C0

500

1,000

1,500 P = 8.82E–05

P = 3.75E–02

P = 4.22E–02

Figure 4.SIK2 knockdown induces a cell-cycle arrest in G1-phase anddelaysmitotic progression of LNCaP cells. A, the population distribution of LNCaP cells transiently knockeddown for SIK2 was assessed by cell-cycle analysis using DAPI staining and flow cytometry. The percentage of cells in G1, S, and G2–M phases of the cell cycle 48and 72 hours after transfection with the control nontargeting siRNA (NT) and SIK2-siRNAs (A–C) is shown (n ¼ 3; error bars represent mean � SD; �, P < 0.05;�� , P < 0.01; �� , P < 0.001). B, representative flow cytometry profiles for each condition. C, p21 protein expression was assessed by Western blot analysis 48 hoursafter transfection of LNCaP cells with the control nontargeting siRNA (NT) and SIK2-siRNAs (A–C). Cell lysates were resolved by SDS-PAGE, transferred tonitrocellulose membrane, and immunoblotted for p21 (Cell Signaling Technology). SIK2 levels after knockdown are also shown (Sigma-Aldrich). b-Actin (Abcam)was used as a loading control. D, LNCaP cells were transiently knocked down for SIK2 for 36 hours and monitored for 48 hours (one image every 5 minutes)usingbright-fieldphase-contractmicroscopy. The time interval from the start of prometaphase (cells detaching and rounding-up) to the endof cytokinesis (generationof twodaughter cells)was estimated for 20 individual cellsper condition. Themeanmitoticprogression timeafter transfectionwith thecontrol nontargeting siRNA(NT)and SIK2-siRNAs (A–C) is shown (n ¼ 20; error bars represent mean � SD; � , P < 0.05; �� , P < 0.01; �� , P < 0.001).






colocalization was particularly apparent in metaphase cells, andto a lesser extent in anaphase and interphase cells (Supplemen-tary Fig. S4B). The predominant detection of a single band usingWestern blotting that was enhanced after overexpression of thewild-type or kinase-dead (KI) constructs and dissipated aftersiRNA depletion confirmed the specificity of the antibody aspreviously reported (Supplementary Fig. S5; ref. 19).

The colocalization of SIK2 with the centrosomes in prostatecancer cells strongly suggests that SIK2 might be involved in theinitiation of centrosome splitting and the subsequent inductionof cell entry intomitosis as it was described in ovarian cancer cells(19). We consequently used bright field microscopy and time-lapse imaging to test whether cells knocked down for SIK2 haddelayedmitotic progression. LNCaP cells were transfectedwith allthree SIK2-siRNA sets (A, B, and C) and the nontargeting siRNA(NT), andwere grown for 36 hours before beingmonitored for 48hours using time-lapse imaging. The time from the start ofprometaphase, as evidence by the cells detaching and round-ing-up, to the end of cytokinesis, as evidence by the generation oftwo daughter cells, was scored for 20 individual cells per condi-tion. Results showed that loss of SIK2 resulted in a dramaticincrease in the mean mitotic progression time (52 minutes incontrol cells transfectedwith nontargeting siRNA versus 112, 183,and 437 minutes following SIK2 knockdown; Fig. 4D). Impor-tantly, many of the SIK2-depleted cells either did not exit mitosisduring the time of recording or failed to undergo cytokinesis soexited into G1 or entered apoptosis (data now shown). This wasparticularly apparent with siRNA C where 15 cells out of the 20counted underwent apoptosis, which represents 75% of the cellsthat corroborates the data shown in Fig. 3A. Interestingly, out ofthose 15 cells that underwent apoptosis, two managed to com-plete mitosis but entered apoptosis as soon as cytokinesis wascompleted (data not shown). These data strongly suggest thatSIK2 is involved in the initiation of cell entry into mitosis (G2 toM) or in the progression through mitosis as it was described inovarian cancer cells (19). Therefore, the cell-cycle arrest observedin G1 after SIK2 depletion (Fig. 4A) is the result of cells failing tocomplete mitosis and escaping into G1.

SIK2 regulates CREB1 activity via the phosphorylation ofTORC1 and its sequestration in the cytoplasm

The importance of SIK2 in prostate cancer cells growth andsurvival and the significant impact of the kinase-dead mutant onthe cells phenotype implicate SIK2 and its downstream targets inprostate cancer progression. Several studies have pointed to a rolefor SIK2 in the regulation of metabolic and survival pathways viathe repression of CREB1-dependent transcription following thephosphorylation of the transducers of regulated CREB1 activity(TORC; refs. 5–11). Prostate cancer therapies typically involve theimposition of stress on cancer cells through hormone-deprivationand -related approaches. To provide a more complete under-standing of the impact of SIK2 on prostate cancer cells, we testedwhether changing SIK2 expression levels might affect CREB1-dependent gene transcription via TORC phosphorylation anddegradation or cellular relocalization in our cell lines. First, weused reporter assays and nuclear fractionation methods to testwhether CREB1 activity and TORC1 phosphorylation and degra-dation and/or localization were under the control of SIK2 inLNCaP cells.

Using a luciferase reporter construct containing the specificDNA-binding sequences of CREB1 (CRE) and a constitutively

expressing Renilla luciferase construct, we monitored the activityof CREB1 24 hours following SIK2 knockdown and 12 hoursfollowing overexpression of wild-type and kinase-dead SIK2.CREB1 activity also was assessed 12 hours after treatment withForskolin, an inducer of CREB1 that raises the intracellular levelsof cAMP and induces CREB1 activity through the activation ofcAMP-dependent protein kinase (PKA), and the subsequent phos-phorylation of CREB1 at Ser133 (28). As expected, the controlForskolin induced a dramatic increase of CREB1 activity. SIK2depletion using three independent siRNAs or ectopic expressionof kinase-dead SIK2 (KI) resulted in a significant increase in CREB1activity. Conversely, CREB1 activity was significantly reducedwhen thewild-type kinase was overexpressed (Fig. 5A).We also ob-served a dose-dependent increase in CREB1 activity along timewhen using the SIK2 inhibitor ARN-3236 (Supplementary Fig. S6)that confirmed the role of the kinase in CREB1 regulation.

On the basis of the observation of a potential regulation ofCREB1 activity by SIK2, we tested whether we could validate adirect interaction between the kinase and its well-known sub-strate, the transducers of regulated CREB1 activity (TORC),TORC1 (14). We could demonstrate that SIK2 and TORC1 direct-ly interact by coimmunoprecipitating both proteins in C4-2b, aprostate cancer cell line expressing high endogenous levels of thekinase (Fig. 5B).

Protein extracts of LNCaP cells transfected with EV, WT, and KIwere then prepared to examine the effects of the kinase on theexpression levels of TORC1. Western blot analyses showed thatoverexpression of the wild-type kinase increased the steady-statelevels of TORC1 but also revealed the presence of a double bandwhen the cells were transfected with the wild-type kinase (Fig. 5C;�CIP). This prompted us to explore whether this was the result ofTORC1 phosphorylation by SIK2. The same extracts were there-fore also treated with a phosphatase alkaline (Fig. 5C;þCIP), andthe disappearance of the upper band after treatment with thephosphatase confirmed the phosphorylation status of TORC1following overexpression of the wild-type kinase. This confirmedthe findings of others depicting role of SIK2 in the prevention ofTORC degradation via its phosphorylation (6–9, 11). Further-more, we observed that the levels of phospho-CREB1 wereunchanged following overexpression of the wild-type andkinase-dead SIK2, which suggests that CREB1 activity is regulatedvia a phosphorylation-independent mechanism.

Isolated cytoplasmic and nuclear fractions for EV, WT, and KIwere then prepared to examine the effects ofwild-type and kinase-dead SIK2 on the cellular distribution of TORC1. Western blotanalyses performedon lysates obtained after cellular fractionationconfirmed the presence of the transcription factor CREB1 and itsphosphorylated form pCREB1 in the nucleus. They also showed aclear accumulation of TORC1 in the cytoplasmic fraction whenthe cells were transfected with wild-type SIK2. Conversely, lowerlevels of TORC1were found in the nuclear fractionswhen the cellswere transfected with the wild-type kinase compared with theempty vector or the kinase-dead mutant. The change in TORC1localization after overexpression of wild-type and kinase-deadSIK2 suggests it relates to the activity of the kinase and not merelyits presence, because the kinase-dead mutant was not able tomaintain TORC1 in the cytoplasm. Surprisingly, thismutant doesnot appear to induce more TORC1 translocation to the nucleusthan the empty vector (similar levels of TORC1 were found in thenucleuswithKI comparedwith EV)but therewas a clear inductionof CREB1 activity with kinase-dead SIK2 in the reporter assay.

Bon et al.





Interestingly, the cellular localization of the wild-type and kinase-dead SIK2 was also investigated and high levels of the mutantprotein were found in the nuclear fractions compared with the

wild-type. This would indicate that kinase-dead SIK2 does notonly permit TORC1 translocation to the nucleus but also parti-cipates in its transactivation (Fig. 5D).

A Reporter assayCRE(1)

EV WT KI0.0

0.5

1.0

1.5

2.0

2.5

P < 0.001

P < 0.001

b-Tubulin

Histone H3

CREB1

pCREB1

TORC1

SIK2

EV WT KI EV WT KI

Cytopl. Nucl.

MW (kDa)

52

76

17

38

38

102

B

b-Actin

CREB1

pCREB1

SIK2

TORC1

EV WT KI EV WT KI

–CIP +CIP

D

MW (kDa)

38

38

38

102

76

V eh . F or sk .0.0

2.0

4.0

6.0

8.0C

RE

(1)

acti

vity

P < 0 .0 01

IP

MW (kDa)

102

76TORC1

SIK2

SP

Cyt

op

. Ig

G

Cyt

op

. in

pu

tN

ucl

. in

pu

t

Cyt

op

. SIK

2

Nu

cl. S

IK2

Nu

cl. I

gG

Nu

cl. S

IK2

Cyt

op

. SIK

2

Cyt

op

. Ig

G

Nu

cl. I

gG

C

CR

E(1

) ac

tivi

ty

Reporter assayCRE(1)

Reporter assayCRE(1)

NT A B C0.0

1.0

2.0

3.0

4.0

P < 0.05

P < 0.05 P < 0.05

CR

E(1

) ac

tivi

ty

Figure 5.SIK2 regulates CREB1 activity via the phosphorylation of TORC1 and its sequestration in the cytoplasm. A, LNCaP cells were transiently transfected with SIK2-siRNAs(A–C), the control nontargeting siRNA (NT), a wild-type SIK2-Myc construct, a kinase-dead SIK2-Myc construct (KI), and the corresponding empty-vectorconstruct (EV). A control Forskolin treatment (Forsk., 10 mmol/L) (þ) or vehicle (DMSO) (�) was also included. CREB1 activity was assessed using a CRE(1) reporterconstruct (Affymetrix) and a dual-luciferase reporter assay system (Promega) 12 hours after Forskolin treatment or SIK2 overexpression and 24 hours after SIK2knockdown (n¼ 12; error bars representmean� SD). B, SIK2was immunoprecipitated (Cell Signaling Technology) in C4-2b cells total protein lysates and immunoblotanalyses of TORC1 were performed. Total cell lysates (input) and immunoprecipitation products (SP and IP) were resolved by SDS-PAGE, transferred tonitrocellulose membrane and immunoblotted for TORC1 (Cell Signaling Technology). SIK2 levels after immunoprecipitation are also shown (Sigma-Aldrich). C,phosphorylation status ofCREB1, pCREB1, andTORC1was checkedafter treatment of total protein lysates obtained fromLNCaP cells 12 hours after transfectionwithEV,WT, and KI using an alkaline phosphatase (New England Biolabs; þ CIP). Cell lysates were resolved by SDS-PAGE, transferred to nitrocellulose membrane, andimmunoblotted for CREB1, pCREB1, and TORC1 (Cell Signaling Technology). SIK2 levels after overexpression are also shown (Sigma-Aldrich). b-Actin (Abcam) wasused as a loading control. D, cellular fractionation was performed on LNCaP cells 12 hours after transfection with EV, WT, and KI. CREB1, pCREB1, and TORC1 proteinexpression in cytoplasmic (cytop.) and nuclear (nucl.) fractions was assessed by Western blot analysis. Cell lysates were resolved by SDS-PAGE, transferred tonitrocellulose membrane, and immunoblotted for CREB1, pCREB1, and TORC1 (Cell Signaling Technology). SIK2 localization is also shown (Sigma-Aldrich). b-Tubulinand histone H3 (Abcam) were used to confirm the purity of the fractionation.






Although we could nicely confirm increased steady-state levelsof the other two TORC isoforms—TORC2 and TORC3—(Sup-plementary Fig. S6B;�CIP), and their phosphorylation followingoverexpression of the wild-type kinase (Supplementary Fig. S6BþCIP), unfortunately, the examination of their cellular distribu-tion has beenmore difficult because TORC2 and TORC3were notdiscernible in the nuclear fractions despite our efforts to increasethe concentrations of the lysates. When we could see a strongaccumulation of TORC2 and TORC3 in the cytoplasmic fractionwhen the cells were transfected with wild-type SIK2, the twoisoforms could not be detected in the nuclear fractions (Supple-mentary Fig. S6C). This issue can be due to a combination ofTORC2 and TORC3 being far less abundant than TORC1 in ourcell line and/or the quality of the antibodies used (SupplementaryFig. S5). Therefore, at that stage, we could not confirm their role inCREB1 regulation because their presence in the nucleus remainedunproven.

SIK2 regulates a cell-cycle network and an ER stress network viaCREB1-dependent transcription

After discovering a possible role for the transcription factorCREB1 and its coactivators, TORCs downstream of SIK2, weperformed gene expression profiling using microarray to definegenes and networks regulated by the SIK2. On the basis of thereporter assay, this approach was used in follow-up to provide anunbiased assessment of the impact on gene expression includingbut also extending beyond established CREB1 target genes.

For comparability, the experiment was conducted under thesame conditions as the reporter assay and cellular fractionationexperiments. LNCaP cells were transiently transfected with thewild-type SIK2 construct, the kinase-dead SIK2 construct (KI) andsiRNA C (SIK2si) because this was the siRNA that gave thestrongest phenotype in all the functional work performed. Theseconditions allowed us to identify a list of SIK2-regulated genesand pathways. We also included the condition where the cellswere stimulated with Forskolin (Forsk.) to generate a signature ofCREB1-regulated genes. The entire list of differentially expressedgenes (DEG) identified in each condition is shown in Supple-mentary Table S2.

We first looked at the DEG in the conditions where SIK2 wasknocked down (SIK2si), kinase-dead SIK2 was overexpressed(KI), and when the cells were treated with Forskolin (Forsk.)because all three conditions induced CREB1 activity (Fig. 5A). Bycomparing the gene signatures in those conditions, we aimed toidentify SIK2-regulated genes dependent on CREB1 activity. Anumber of publications have defined CREB1 target genes basedon chromatin immunoprecipitation and expression array profil-ing under conditions of Forskolin treatment and information isavailable in addition from a searchable online database (http://natural.salk.edu/CREB/). We therefore compared the genes thatwere differentially expressed in our cell lines with those reportedto be CREB1 targets in other studies (29, 30) and selectedinhibitor of DNA-binding 1 (ID1), nuclear receptor subfamily4 group Amember 1 (NR4A1), homeobox A5 (HOXA5), and salt-inducible kinase 1 (SIK1) for validation. Of these genes, ID1 wasincreased in expression by Forskolin treatment but also by knock-down of SIK2 and overexpression of the kinase-dead mutant. Incontrast, the other genes were either only induced by Forskolintreatment (e.g., SIK1) or indeed were inhibited in all conditions(e.g., HOXA5). This suggests that while CREB1 activity isenhanced by all three treatments, the impact of SIK2 knockdown

or overexpression of kinase-dead SIK2 is pleiotropic and extendsto other transcriptional regulators. Beyond that, the difference inForskolin response in our study versus others, for example inrespect to HOXA5 expression, suggests that cell type dependencyalso plays a role and that systematic unraveling of the CREB1regulome and SIK2-dependency will require subsequent in-depthstudy in prostate cells. Cell-type dependency in the CREB1 reg-ulome has been previously reported with Zhang and colleagues(30), demonstrating on overlap of less than 5%.

We then went on to select a number of additional genes forvalidation based on their consistent pattern of expression changein the arraydata under all three conditions. These additional geneswere CCR4 carbon catabolite repression 4-like (CCRN4L), DnaJ(Hsp40) homolog, subfamily C, member 12 (DNAJC12), ERBBreceptor feedback inhibitor 1 (ERRFl1), and nuclear factor of klight polypeptide gene enhancer in B-cells inhibitor, zeta(NFKBIZ), which were upregulated in all three conditions, andhomo sapiens chromosome 1 open reading frame 112(C1orf112), forkead box O3 (FOXO3), and protein tyrosinephosphatase, receptor type, F (PTPRF), which were downregu-lated in the same conditions (Fig. 6A). Interestingly none of thesegenes have been reported to be CREB1 target genes in the data-base. We used quantitative real-time PCR (qRT-PCR) to validatethose findings and confirmed that FOXO3, HOXA5, and PTPRFwere significantly downregulated in all three conditions whileERRFl1 was significantly upregulated (Supplementary Fig. S7).

Around half of the total number of genes downregulated afterSIK2 knockdown overlapped with the condition where the cellswere treated with Forskolin (46%) that confirmed that SIK2regulates gene expression via CREB1, but surprisingly, only12% of the genes upregulated by LNCaP-SIK2-KI and 10% ofthe genes downregulated were similar to the genes differentiallyexpressed after Forskolin treatment. This suggests that the impactof the kinase-dead mutant on gene transcription probablyinvolves transcription factors other than CREB1 (Fig. 6A). Thedetailed gene lists are shown in Supplementary Table S3.

The small number of SIK2-regulated genes dependent onCREB1 activity identified in our study is not amenable to apathway analysis but transcriptional regulation is an overrepre-sented functionwhen considering all eight using theDAVIDGeneOntology web tool. Four fall into this functional classification[FOXO3 (31), HOXA5 (32), NFKBIZ (33), and CCRN4L (34)]. Inaddition, two of the eight genes have been reported to be negativeregulators of receptor tyrosine kinases (ERRFI1 and PTPRF;refs. 35, 36). Clearly, therefore, although the number of genesis small, their impact on gene regulation and signaling is poten-tially significant.

To get a broader insight into the genes and networks down-stream of SIK2, we submitted to the Gene Ontology databaseIngenuity (IPA) the complete list of DEG after wild-type SIK2 andkinase-dead SIK2 overexpression and after SIK2 knockdown. Thetopfive enrichednetworks associatedwith the categories "diseasesand disorders" and "molecular and cellular functions" for eachcondition are shown in Supplementary Fig. S8A. It is clear thatnetworks related to cancerwere considerably affected by triggeringSIK2 in our prostate cancer cell line (cancer was the top disease infour out of six gene lists). This is particularly important becauseonly one study to date has demonstrated a role for SIK2 in cancerprogression (19). Themolecular and cellular functions repeatedlyassociated with the DEG identified in this study related to cellulargrowth and proliferation, cell death and survival, cell cycle, and

Bon et al.





transcription. These findings agree with those of our functionalstudy showing the involvement of SIK2 in prostate cancer cellgrowth, cell cycle, and survival, aswell as gene transcription via theregulation of the TORC/CREB1 transcriptional pathway.

We then interrogated the Gene Ontology database GeneCodis(37, 38) to identify the enriched biologic processes and thecommon transcriptional mediators underpinning the molecularfunction "cell cycle" after SIK2 knockdown and "cell death andsurvival" after overexpression of kinase-dead SIK2. The list of 211genes downregulated after SIK2 knockdownwas submitted to thedatabase and the top 10 biologic processes associated with thisgene list were identified (Supplementary Fig. S8B). It is clear thatprocesses associated with cell-cycle progression, and especiallymitotic progression, were overrepresented. The gene lists of thetop four biologic processes were then submitted for DNA motif

analysis in promoters within 2 kb of the transcription start site.There was a particular enrichment in E2F-binding motifs, indi-cating that a majority of these genes are regulated by the E2Ffamily of transcription factors. Interestingly, motif enrichmentusing the CREB1-binding sites mapped by Zhang and colleagues(29, 30) by chromatin immunoprecipitation also reported thatE2F motifs were the most significantly over-represented in thosesites (P value, 10�42).

The same procedure was followed using the 152 genes upre-gulated after overexpression of kinase-dead SIK2. We identified aparticular enrichment for processes related to response to stresswith pathways relating to endoplasmic reticulum (ER) stress,unfolded protein response (UPR), and protein folding. Theinduction of a set of genes involved in stress response might bean unattended effect following the ectopic expression of a

Figure 6.SIK2 regulates cell cycle and a ER stress response via CREB1-dependent transcription in prostate cancer cells. A, Venn diagrams summarize how many geneswere found to be downregulated or upregulated in LNCaP cells after SIK2 knockdown (SIK2si), overexpression of kinase-dead SIK2 (KI) or treatment withForskolin (Forsk.). Overlaps between the three conditions are also shown. Seventy-eight percent of the genes downregulated after SIK2 knockdown are dependenton CREB1 activity. Ten percent and 12% of the genes down- and upregulated after overexpression of kinase-dead SIK2 are dependent on CREB1 activity.B, proposedmechanism. (i)Wild-type SIK2phosphorylates TORC,which results in TORC sequestration andaccumulation in the cytoplasm. TORC translocation to thenucleus is inhibited, so CREB1 activity is repressed, which results in cell-cycle progression and cell survival. (ii) Kinase-dead SIK2 (KI) cannot phosphorylate TORC.SIK2-KI and TORC form a complex that translocates to the nucleus where it activates CREB1 and other transcription factors, such as HSF, IRF, and NF-kB,which results in ER stress response and cell death.






recombinant protein into the cells. Despite this, pathways relatingto protein folding and response to stress were not enriched afteroverexpression of the wild-type kinase (Supplementary Fig. S8A)and from the gene lists generated, a subset of those stress geneswere distinctly being expressed after overexpression of the kinase-dead protein versus the wild-type (DNAJA1, DNAJB9, DNAJC12,DNAJC3, HSP90AA1, HSP90AB1, HSP90AB3P, HSP90AB4P,HPSA4, HSPA4L, HSPA5, HSPB1, and HSPD1; SupplementaryTable S2), which reflects the distinct effect of themutant kinase onER stress genes transcription. This was further supported by anenrichment for transcription factors such as heat shock factor(HSF), a transcriptional activator of heat shock genes, interferonregulatory factor 3 (IRF), a transcription factor involved in theregulation of interferon-regulated genes, and nuclear factork-light-chain-enhancer of activated B (NF-kB), a transcriptionalactivator involved in cellular responses to stimuli, such as stress,when the kinase-dead protein was overexpressed (SupplementaryFig. S8B).

The detailed lists of genes involved in the processes of"mitotic cell cycle," "protein folding," and "response to stress"are shown in Supplementary Table S4. Owing to the profoundimpact of SIK2 knockdown in reducing cell proliferation andviability it was not possible to perform rescue experiments withSIK2 overexpression constructs against a knockdown back-ground. Further dissection of these processes will therefore bepursued in future studies.

Taken together, these data suggest a hypotheticalmechanism inwhich when SIK2 is present and catalytically active, it phosphor-ylates TORC1 that results in its sequestration in the cytoplasm.When in the cytoplasm, TORC1 is stabilized and not degraded asdescribed in other studies, but its translocation into the nucleus isinhibited so CREB1-dependent gene transcription is repressed,which results in cell-cycle progression and ultimately, cell survival(Fig. 6B, i). When SIK2 is present but unable to phosphorylateTORC1, it is translocated to the nucleus with TORC1, probably asa complex. Once in the nucleus, the TORC1–SIK2–KI complexcould act as a CREB1 transactivator and also triggers the activationof other transcription factors such as HSF, IRF, and NF-kB, whichresults in ER stress response and ultimately, cell death (Fig. 6B, ii).

DiscussionThis is the first study to report that SIK2 auto-antibodies are

highly expressed in patients with aggressive disease and poorprognosis and that they could be a good blood marker for thediagnosis of prostate cancer. Currently, no single cohort existswith sufficient clinical follow-up as well as matched blood andtissue samples in which to test for SIK2 auto-antibody levels,lSIK2 tissue expression and prognosis. Biobanks are beingdeveloped that will allow this to be addressed in the longerterm future and this analysis will, from a clinical perspective, bevital in determining whether SIK2 inhibitors really should beintroduced in prostate cancer patients. The case for developingSIK2 inhibitors was made earlier in ovarian cancer by Ahmedand colleagues (19). They showed that that SIK2 is a centro-some kinase required for the initiation of mitosis so we firstconsidered a potential role for SIK2 in cell growth and cell-cycleregulation in our cells.

In our study, we characterized the phenotypic effects of SIK2overexpression and knockdown on prostate cancer cell lines forthe first time. Our results showed that SIK2 knockdown induces a

pronounced reduction in cell growth in a selection of prostatecancer cell lines (LNCaP, LNCaP-Bic, C4-2, and DuCaP) and thisis accompanied by a G1 cell-cycle arrest involving the cell-cycleregulators p21, p27, and cyclin D1 in LNCaP cells. We couldreproduce the findings from research into ovarian cancer cellsusing our prostate cancer cell lines and showed that SIK2 colo-calizes at the centrosomes in cells inmetaphase and anaphase andregulates cells entry or progression intomitosis. Themost intrigu-ing aspect of our findings was the pronounced inhibition of cellproliferation observed when the activity of the kinase was abol-ished (overexpression of a kinase-deadmutant of SIK2) in LNCaPcells. This was sufficiently potent that it led to a deselection of cellstransfect with the kinase-dead mutant during our attempts toderive stable cell lines expressing the mutant. The significance ofthe kinase-dead mutant on cell growth has not been reportedbefore and indicates the importance of SIK2 downstream targetsin prostate cancer progression. SIK2 mutations are detectable inTCGA (The Cancer Genome Atlas) data (39) and in the CancerCell-Line Encyclopedia (CCLE; ref. 40); however, none have beenfind in the catalytic site with the exception of a Lysine-49 tothreonine mutation reported in an intestinal cell line in theCCLE dataset. This residue is the same one that was mutated tocreate the kinase-dead mutant, SIK2-KI. Several studies haveidentified a role for SIK2 and its downstream target TORC in theregulation of cellular energy homeostasis under certain stressconditions (5–11). It is important to take into account that SIK2is a member of the AMPK family, the members of which aremetabolic master kinases involved in the regulation of a numberof metabolic processes including b-oxidation of fatty acids, lipo-genesis, protein and cholesterol synthesis, as well as apoptosis(41, 42). It was therefore crucial to consider that SIK2 is not only acentrosome kinase but it may also be implicated in the regulationof biologic processes essential for the survival of prostate cancercells. Indeed, a new function for SIK2 described in our work andpreviously unreported in other cancer types is its role in cellsurvival. We showed that SIK2 depletion induces cell death andthat inhibition of its kinase activity also dramatically affects cellviability. This supports findings by Ahmed and colleagues (19),who suggested that SIK2 could present novel avenues for thedevelopment of novel cancer therapies. The authors suggest thatSIK2 inhibitors could be used in combination with paclitaxel, adrug of category of the taxanes, which disrupts microtubulesfunction so inhibits the process of cell division. Drugs thatinterfere with the mitotic process are commonly used for thetreatment of many types of cancers (43). For instance, docetaxel,another taxane family member, is used for the treatment ofadvanced prostate cancer. The cytotoxic activity of this type ofdrug in combination with SIK2 inhibitors might be exerted inprostate cancer compared with other cancer types because ourdata indicate that SIK2 not only interferes with cell-cycle progres-sion but also induces caspase-dependent apoptosis.

We then demonstrated that SIK2 regulates the activity of thetranscription factor CREB1 by interacting with its transactivatorTORC1 and altering its cellular distribution via its phosphoryla-tion (Fig. 5B–D). Overexpression of a kinase-dead mutantreduced steady-state levels but also led to a loss of a band doubletthat could also be removed by phosphatase treatment, indicatingthat it represented the phosphorylated form of the protein (Fig.5C). These results confirmed findings published by others show-ing that SIK2 phosphorylates TORCs that results in their seques-tration in the cytoplasm, but they did not corroborate the findings

Bon et al.





of Koo and colleagues (6), Dentin and colleagues (7), Wang andcolleagues (8), Ryu and colleagues (9), or Sasaki and colleagues(11) who reported that accumulation of TORC within the cyto-plasm was followed by its degradation. Interestingly, we alsoshowed that the cellular localization of the kinase-dead mutantwas nuclear compared with the wild-type kinase. This suggeststhat SIK2 regulation of CREB1-mediated transcription might bedirect. Only SIK1 has been shown to act directly on CREB1 in thenucleus and all studies that report a role for SIK2 in CREB1-mediated transcription have implicated the transactivatorsTORCs. However, Horike and colleagues (26) have shown thatSIK2 was mostly present in the cytoplasm in mouse adipose 3T3-L1 cells, but could be translocated to the nucleus when itsphosphorylatable Ser587 was replaced with Alanine (S587A).The authors suggested that SIK2 could act as a direct repressorfor CREB1-mediated gene transcription, as was the case withSIK1 (26). Here, we suggest that kinase-dead SIK2 might bepart of a complex with TORC that directly induces CREB1 trans-activation in the nucleus. We used a gene expression profilingapproach to define the genes and networks regulated by the SIK2/TORC/CREB1 transcriptional pathway. We compared the genesdifferentially expressed in our studywith two published studies inwhich CREB1 target genes were identified (29, 30). Out of thegenes we identified, HOXA5, was a known CREB1-regulated gene(29), which makes it an interesting candidate involved in SIK2signaling viaCREB1-mediated gene transcription.However, itwasdownregulated in response to Forskolin treatment, SIK2 deple-tion, and kinase-dead SIK2 overexpression that did not corrob-orate the enhanced CREB1 activity observed in our reporter assay.Zhang and colleagues (30) further reported that certain CREB1targets are silenced by CpG island hypermethylation, suggestingthat other genomic factors can affect patterns of these genes. Oneof the genes we selected for validation based on Ravnskjaer andcolleagues's (29) study, ID1, was reported to be never hyper-methylated in Zhang and colleagues's (30) and was robustlyinduced in our experiment.

We also found that about only 10% of the genes regulated byLNCaP-SIK2-KI were similar to the genes differentially expressedafter Forskolin treatment, which suggests that the impact of thekinase-dead mutant on gene transcription probably involvestranscription factors other than CREB1. Consistent with this wasthe very different network of genes identified following over-expression of kinase-dead SIK2 and depletion of the kinase (veryfew overlaps between both conditions). This was further sup-ported by the motif analysis that revealed that transcriptionfactors, such as E2F, HSF, IRF, NF-kB, were enriched after over-expression of the kinase-dead protein.

Knockdown of SIK2 and overexpression of the kinase-deadmutant both affect cell-cycle progression. Consequently, thedifferential expression of cell cycle/E2F–associated genes maynot be a direct consequence of the effect of targeting SIK2 ontranscription, but rather a by-product of the impact on cell-cycleprogression. Equally the overexpression of stress response geneswhen the cells were transfected with the kinase-dead mutant maybe due to the stress associated with ectopic expression of highlevels of a non-native protein in the cells rather than a directregulatory impact of the mutant on a particular transcriptionfactors or group of transcription factors. Clearly dissecting causeversus effect within these gene expression datasets will requiresignificant additional studies that are beyond the scope of thisarticle. In particular, given that SIK2 and TORCs undergo nuclear–

cytoplasmic translocation, it is of interest in future work to maptheir associated proteomes in these compartments. A recentmethod, rapid immunoprecipitation mass spectrometry ofendogenous proteins (RIME) also provides the possibility toenrich CREB1 or indeed TORCs/SIK2 together with chromatinand begin to define the chromatin-associated protein complexes(44). This method has revealed new coregulators for estrogenreceptor–a and may uncover important differences in the tran-scription factors associating with the kinase-dead or wild-typekinase (44). These points are of fundamental importance alsobecause of the rapid progress that is being made to develop drugsthat inhibit SIK2 kinase activity, based on the previous article inovarian cancer, with the aim of arresting cell-cycle progression. Inprostate cancer, enhanced CREB1 activity has been reported topromote prostate cancer progression and cell survival (45). Inpart, this has been suggested to occur because CREB1 importantlyalso enhances the expression of a number of antiapoptotic genesincluding Bcl-2 in other experimental models (46, 47). In thissetting, it is alsoworth noting that CREB1 activity is also enhancedby other pathways and enzymes that have been associated withprogression to castrate-resistant disease including CAMKII acti-vation (48). This is important neuronal survival and memorypotentiation but the kinase also has a role in promoting cellprogression in conditions of androgen deprivation (49). Conse-quently, in prostate cancer, the possibility exists that while inhi-biting SIK2 may inhibit cell-cycle progression by the same mech-anism thatwas reported in ovarian cancer, theremay be a counter-acting pressure towards survival and ultimately cancer progres-sion through enhanced CREB1 activation.

The impact of SIK2 on cellular phenotypes, as with manykinases, can be through their kinase activity but also throughprotein–protein interactions (50). By overexpressing thekinase-dead mutant, we are potentially sequestering SIK2 inter-acting proteins in complexes that are no longer dynamicallyregulated by SIK2 kinase activity whereas with SIK2 knock-down, we are disrupting processes that require both SIK2-mediated kinase and SIK2-mediated protein–protein interac-tions. The impact of the SIK2 kinase-dead mutant on networkof genes and pathways involved in cell growth and survivalunder certain stress conditions indicates that SIK2 protein–protein interactions are important regulators of viability. SIK2inhibitors are under development to treat ovarian cancer basedon the role of SIK2 as a regulator of mitosis. We have tested onesuch inhibitor with low nanomolar IC50 values in activityassays and preclinical models, ARN-3236 (22). In reporterassays, this drug induced CREB activation in a similar mannerto the kinase-dead mutant (SIK2-KI; Supplementary Fig. S6A)underscoring the importance of considering the impact onCREB signaling in evaluating responses to these novel agents.In conclusion, our study argues for further dissection of themechanism of action of SIK2 in prostate cancer, particularlygiven the metabolic features of localized disease and the cell-cycle dysregulation associated with metastatic progression.

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

Authors' ContributionsConception and design: H. Bon, A.A. Ahmed, I.G. MillsDevelopment of methodology: H. Bon, A.A. Ahmed, I.G. Mills






Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): H. Bon, K. Wadhwa, A. Schreiner, M. Osborne,R. HoffmannAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):H.Bon, K.Wadhwa, A. Schreiner, T. Carroll,M. Visser,R. Hoffmann, A.A. Ahmed, I.G. MillsWriting, review, and/or revision of the manuscript: H. Bon, A. Schreiner,M. Osborne, A. Ramos-Montoya, R. Hoffmann, A.A. Ahmed, D.E. Neal,I.G. MillsAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): H. Bon, H. Ross-AdamsStudy supervision: A. Ramos-Montoya, D.E. Neal, I.G. MillsOther (group leader for the Neal laboratory): D.E. Neal

AcknowledgmentsThe authors thank study volunteers for their participation and to staff at

the Welcome Trust Clinical Research Facility, Addenbrooke's Clinical ResearchCentre, and Cambridge for their help in conducting the study. The authors

also acknowledge the support of the NIHR Cambridge Biomedical ResearchCentre, the DOHHTA, and the NCRI/MRC for help with the bio-repository. Allexpression array data have been deposited onGene ExpressionOmnibus (GEO;GSE45711).

Grant SupportThis study was financially supported by Cancer Research UK; NIHR Cam-

bridge Biomedical Research Centre, the DOH HTA (ProtecT grant); and theNCRI/MRC (ProMPT grant). H. Bonwas supported by a BBSRC/CASE industrialPhD studentship.

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 April 15, 2013; revised November 4, 2014; accepted December 2,2014; published OnlineFirst December 29, 2014.

References1. Available from:http://www.cancerresearchuk.org/cancer-info/cancerstats/.2. Nørgaard M, Jensen A�, Jacobsen JB, Cetin K, Fryzek JP, Sørensen HT.

Skeletal related events, bone metastasis and survival of prostate cancer: apopulation based cohort study in Denmark (1999 to 2007). J Urol 2010;184:162–7.

3. Kirby M, Hirst C, Crawford E. Characterising the castration-resistantprostate cancer population: a systematic review. Int J Clin Pract 2011;65:1180–92.

4. Kahn B, Alquier T, Carling D, Hardie DG. AMP-activated protein kinase:ancient energy gauge provides clues to modern understanding of metab-olism. Cell Metab 2005;1:15–25.

5. Muraoka M, Fukushima A, Viengchareun S, Lomb�es M, Kishi F,Miyauchi A, et al. Involvement of SIK2/TORC2 signaling cascade inthe regulation of insulin-induced PGC-1 and UCP-1 gene expressionin brown adipocytes. Am J Physiol Endocrinol Metab 2009;296:E1430–9.

6. Koo S-H, Flechner L, Qi L, Zhang X, Screaton RA, Jeffries S, et al. The CREBcoactivator TORC2 is a key regulator of fasting glucose metabolism.Nature 2005;437:1109–11.

7. Dentin R, Liu Y, Koo SH, Hedrick S, Vargas T, Heredia J, et al. Insulinmodulates gluconeogenesis by inhibition of the coactivator TORC2.Nature 2007;449:366–9.

8. Wang B, Goode J, Best J, Meltzer J, Schilman PE, Chen J, et al. The insulin-regulated CREB coactivator TORCpromotes stress resistance in drosophila.Cell Metab 2008;7:434–44.

9. Ryu D, Oh KJ, Jo HY, Hedrick S, Kim YN, Hwang YJ, et al. TORC2 regulateshepatic insulin signaling via amammalian phosphatidic acid phosphatase,LIPIN1. Cell Metab 2009;9:240–51.

10. Screaton RA, Conkright MD, Katoh Y, Best JL, Canettieri G, Jeffries S, et al.The CREB coactivator TORC2 functions as a calcium- and cAMP-sensitivecoincidence detector. Cell 2004;119:61–74.

11. Sasaki T, Takemori H, Yagita Y, Terasaki Y, Uebi T, Horike N, et al. SIK2 is akey regulator for neuronal survival after ischemia via TORC1-CREB.Neuron 2011;69:106–19.

12. Lonze B, Ginty D. Function and regulation of CREB family transcriptionfactors in the nervous system. Neuron 2002;35:605–23.

13. Conkright MD, Canettieri G, Screaton R, Guzman E,Miraglia L, HogeneschJB, et al. TORCs. Mol Cell 2003;12:413–23.

14. Katoh Y, Takemori H, Lin XZ, Tamura M, Muraoka M, Satoh T, et al.Silencing the constitutive active transcription factor CREB by the LKB1-SIKsignaling cascade. FEBS J 2006;273:2730–48.

15. Katoh M, Katoh M. Identification and characterization of humanKIAA1391 and mouse Kiaa1391 genes encoding novel RhoGAP familyproteins with RA domain and ANXL repeats. Int J Oncol 2003;23:1471–6.

16. Imielinski M, Berger AH, Hammerman PS, Hernandez B, Pugh TJ, Hodis E,et al. Mapping the hallmarks of lung adenocarcinoma with massivelyparallel sequencing. Cell 2012;150:1107–20.

17. Wehr MC, Holder MV, Gailite I, Saunders RE, Maile TM, Ciirdaeva E, et al.Salt-inducible kinases regulate growth through the Hippo signalling path-way in Drosophila. Nat Cell Biol 2013;15:61–71.

18. Nagel S, Leich E, Quentmeier H, Meyer C, Kaufmann M, Zaborski M, et al.Amplification at 11q23 targets protein kinase SIK2 in diffuse large B-celllymphoma. Leuk Lymphoma 2010;51:881–91.

19. Ahmed AA, Lu Z, Jennings NB, Etemadmoghadam D, Capalbo L, JacamoRO, et al. SIK2 is a centrosome kinase required for bipolar mitotic spindleformation that provides a potential target for therapy in ovarian cancer.Cancer Cell 2010;18:109–21.

20. Donovan J, Hamdy F, Neal D, Peters T, Oliver S, Brindle L, et al. Prostatetesting for cancer and treatment (ProtecT) feasibility study. Health TechnolAssess 2003;7:1–88.

21. Ghevaria N, Visser M, Hoffmann R. Quality control for a large-scale studyusing protein arrays and protein beads to measure immune response inserum and plasma. Proteomics 2012;12:2802–7.

22. Alfredi A, Zhang S, Mao W, Wang Y, Takemori H, Lu Z, et al. Highlypotent and orally available SIK2 inhibitors block growth of humanovarian cancer cells in culture and xenografts. In: Proceedings of the105th Annual Meeting of the American Association for Cancer Research;2014 Apr 5–9; San Diego, CA. Philadelphia (PA): AACR; 2014. Abstractnr: 749.

23. Wersto R, Chrest FJ, Leary JF,Morris C, Stetler-StevensonMA,Gabrielson E.Doublet discrimination in DNA cell-cycle analysis. Cytometry 2001;46:296–306.

24. Dunning M, Smith ML, Ritchie ME, Tavar�e S. beadarray: R classesand methods for Illumina bead-based data. Bioinformatics 2007;23:2183–4.

25. Zweig M, Campbell G. Receiver-operating characteristic (ROC) plots: afundamental evaluation tool in clinical medicine. Clin Chem 1993;39:561–77.

26. Horike N, Takemori H, Katoh Y, Doi J, Min L, Asano T, et al. Adipose-specific expression, phosphorylation of Ser794 in insulin receptor sub-strate-1, and activation in diabetic animals of salt-inducible kinase-2. J BiolChem 2003;278:18440–7.

27. Wiese C, Zheng Y. Microtubule nucleation: gamma-tubulin and beyond.J Cell Sci 2006;119(Pt 20):4143–53.

28. Souza N, Dohadwalla A, Reden J. Forskolin: a labdane diterpenoid withantihypertensive, positive inotropic, platelet aggregation inhibitory,and adenylate cyclase activating properties. Med Res Rev 1983;3:201–19.

29. Ravnskjaer K, Kester H, Liu Y, Zhang X, LeeD, Yates JR III, et al. Cooperativeinteractions betweenCBPandTORC2confer selectivity toCREB target geneexpression. EMBO J 2007;26:2880–9.

30. Zhang X, Odom DT, Koo SH, Conkright MD, Canettieri G, Best J, et al.Genome-wide analysis of cAMP-response element binding protein occu-pancy, phosphorylation, and target gene activation in human tissues. ProcNatl Acad Sci U S A 2005;102:4459–64.

Bon et al.





31. Yang JY, Hung MC. A new fork for clinical application: targeting forkheadtranscription factors in cancer. Clin Cancer Res 2009;15:752–7.

32. Nunes FD, de Almeida FC, Tucci R, de Sousa SC. Homeobox genes: amolecular link between development and cancer. Pesqui Odontol Bras2003;17:94–8.

33. Muta T. IkappaB-zeta: an inducible regulator of nuclear factor-kappaB.Vitam Horm 2006;74:301–16.

34. Hee SW, Tsai SH, Chang YC, Chang CJ, Yu IS, Lee PC, et al. The role ofnocturnin in early adipogenesis and modulation of systemic insulinresistance in human. Obesity 2012;20:1558–65.

35. Segatto O, Anastasi S, Alema S. Regulation of epidermal growth factorreceptor signalling by inducible feedback inhibitors. J Cell Sci 2011;124(Pt 11):1785–93.

36. den Hertog J, Blanchetot C, Buist A, Overvoorde J, van der Sar A, TertoolenLG. Receptor protein-tyrosine phosphatase signalling in development. Int JDev Biol 1999;43:723–33.

37. Carmona-Saez P, Chagoyen M, Tirado F, Carazo JM, Pascual-Montano A.GENECODIS: a web-based tool for finding significant concurrent annota-tions in gene lists. Genome Biol 2007;8:R3.

38. Nogales-Cadenas R, Carmona-Saez P, Vazquez M, Vicente C, Yang X,Tirado F, et al. GeneCodis: interpreting gene lists through enrichmentanalysis and integration of diverse biological information. Nucleic AcidsRes 2009;37(Web Server issue):W317–22.

39. Cancer Genome Atlas Research NetworkWeinstein JN, Collisson EA, MillsGB, Shaw KR, Ozenberger BA, et al. The Cancer Genome Atlas Pan-Canceranalysis project. Nat Genet 2013;45:1113–20.

40. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S,et al. The Cancer Cell Line Encyclopedia enables predictive modelling ofanticancer drug sensitivity. Nature 2012;483:603–7.

41. Bright NJ, Thornton C, Carling D. The regulation and function of mam-malian AMPK-related kinases. Acta Physiol 2009;196:15–26.

42. Carling D, Thornton C, Woods A, Sanders MJ. AMP-activated proteinkinase: new regulation, new roles? Biochem J 2012;445:11–27.

43. Johansson M, Persson JL. Cancer therapy: targeting cell cycle regulators.Anticancer Agents Med Chem 2008;8:723–31.

44. Mohammed H, D'Santos C, Serandour AA, Ali HR, Brown GD, Atkins A,et al. Endogenous purification reveals GREB1 as a key estrogen receptorregulatory factor. Cell Rep 2013;3:342–9.

45. Wu D, Zhau HE, Huang WC, Iqbal S, Habib FK, Sartor O, et al. cAMP-responsive element-binding protein regulates vascular endothelial growthfactor expression: implication in human prostate cancer bone metastasis.Oncogene 2007;26:5070–7.

46. Pugazhenthi S, Nesterova A, Sable C, Heidenreich KA, Boxer LM,Heasley LE, et al. Akt/protein kinase B up-regulates Bcl-2 expressionthrough cAMP-response element-binding protein. J Biol Chem 2000;275:10761–6.

47. Yune TY, Park HG, Lee JY, Oh TH. Estrogen-induced Bcl-2 expression afterspinal cord injury is mediated through phosphoinositide-3-kinase/Akt-dependent CREB activation. J Neurotrauma 2008;25:1121–31.

48. Wei G, Chen YB, Chen DF, Lai XP, Liu DH, Deng RD, et al. beta-Asaroneinhibits neuronal apoptosis via the CaMKII/CREB/Bcl-2 signaling pathwayin an in vitro model and AbetaPP/PS1 mice. J Alzheimers Dis 2013;33:863–80.

49. RokhlinOW, TaghiyevAF, Bayer KU, BumcrotD, Koteliansk VE, Glover RA,et al. Calcium/calmodulin-dependent kinase II plays an important role inprostate cancer cell survival. Cancer Biol Ther 2007;6:732–42.

50. Taylor SS, Kornev AP. Protein kinases: evolution of dynamic regulatoryproteins. Trends Biochem Sci 2011;36:65–77.






2015;13:620-635. Published OnlineFirst December 29, 2014.Mol Cancer Res Hélène Bon, Karan Wadhwa, Alexander Schreiner, et al. Transcription in Prostate CancerSalt-Inducible Kinase 2 Regulates Mitotic Progression and

Updated version

10.1158/1541-7786.MCR-13-0182-Tdoi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2015/01/06/1541-7786.MCR-13-0182-T.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/13/4/620.full#ref-list-1

This article cites 48 articles, 9 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/4/620.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/13/4/620To 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-13-0182-T

http://mcr.aacrjournals.org/content/suppl/2015/01/06/1541-7786.MCR-13-0182-T.DC1

http://mcr.aacrjournals.org/content/13/4/620.full#ref-list-1

http://mcr.aacrjournals.org/content/13/4/620.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/4/620


Documents

Salt-Inducible Kinase 2 Regulates Mitotic Progression and Transcription in Prostate … · Cell Cycle and Senescence Salt-Inducible Kinase 2 Regulates Mitotic Progression and Transcription