Metabolism

Lipid Uptake Is an Androgen-Enhanced LipidSupply Pathway Associated with Prostate CancerDisease Progression and Bone MetastasisKaylyn D. Tousignant1, Anja Rockstroh1, Atefeh Taherian Fard1, Melanie L. Lehman1,Chenwei Wang1, Stephen J. McPherson1, Lisa K. Philp1, Nenad Bartonicek2,3,Marcel E. Dinger2,3, Colleen C. Nelson1, and Martin C. Sadowski1

Abstract

De novo lipogenesis is a well-described androgen receptor(AR)–regulated metabolic pathway that supports prostatecancer tumor growth by providing fuel, membrane material,and steroid hormone precursor. In contrast, our current under-standing of lipid supply from uptake of exogenous lipids andits regulationbyAR is limited, and exogenous lipidsmayplay amuch more significant role in prostate cancer and diseaseprogression than previously thought. By applying advancedautomated quantitative fluorescence microscopy, we providethe most comprehensive functional analysis of lipid uptake incancer cells to date and demonstrate that treatment of AR-positive prostate cancer cell lines with androgens results insignificantly increased cellular uptake of fatty acids, cholester-ol, and low-density lipoprotein particles. Consistent with adirect, regulatory role of AR in this process, androgen-enhanced lipid uptake can be blocked by the AR-antagonistenzalutamide, but is independent of proliferation and cell-cycle progression. This work for the first time comprehensively

delineates the lipid transporter landscape in prostate cancercell lines and patient samples by analysis of transcriptomicsand proteomics data, including the plasma membrane prote-ome. We show that androgen exposure or deprivation regu-lates the expression of multiple lipid transporters in prostatecancer cell lines and tumor xenografts and that mRNA andprotein expression of lipid transporters is enhanced in bonemetastatic disease when compared with primary, localizedprostate cancer. Our findings provide a strong rationale toinvestigate lipid uptake as a therapeutic cotarget in the fightagainst advanced prostate cancer in combination with inhi-bitors of lipogenesis to delay disease progression andmetastasis.

Implications: Prostate cancer exhibits metabolic plasticity inacquiring lipids from uptake and lipogenesis at differentdisease stages, indicating potential therapeutic benefit bycotargeting lipid supply.

IntroductionThe role of lipid metabolism in the incidence and progression

of prostate cancer and several other cancer types has gainednotable attention in an attempt to develop new therapeuticinterventions. Lipids represent a diverse group of compoundsderived from fatty acids and cholesterol that serve an essential roleinmany physiologic and biochemical processes. They function inenergy generation and storage as well as intracellular signaling,protein modification, and precursor for steroid hormone synthe-

sis. Additionally, fatty acids serve as the main building blocks forphospholipids that are incorporated togetherwith free cholesterolinto membranes and are critical for membrane function, cellsignaling, and proliferation.

As a source of lipid supply, uptake of circulating exogenouslipids is sufficient for the requirements of most normal cells, andfollowing development, lipogenic enzymes remain expressed atrelatively low levels apart from a few specific biological processes(surfactant production in the lungs, production of fatty acids formilk lipids during lactation, and steroidogenic activity in tissuesincluding prostate). However, lipogenic pathways, i.e., de novolipogenesis (DNL) of fatty acids and cholesterol, are reactivated orupregulated inmany solid cancer types, including prostate cancer.Enhanced lipogenesis is now acknowledged as a metabolic hall-mark of cancer and is an early metabolic switch in the develop-ment of prostate cancer. It is maintained throughout the progres-sion of prostate cancer and associated with poor prognosis andaggressiveness of disease (1–5). Yet, the contribution and identityof lipid uptake pathways as a supply route of exogenous lipids andtheir role in disease development and progression remain largelyunknown.

Several lipogenic enzymes, including fatty acid synthase(FASN), are found to be overexpressed in prostate cancer(reviewed in ref. 1). Because increased FASN gene copy number,transcriptional activation, or protein expression are common

1Australian Prostate Cancer Research Centre, Queensland, Institute of Healthand Biomedical Innovation, School of Biomedical Sciences, Faculty ofHealth, Queensland University of Technology, Princess Alexandra Hospital,Translational Research Institute, Woolloongabba, Queensland, Australia.2Kinghorn Centre for Clinical Genomics, Garvan Institute of Medical Research,Sydney, Australia. 3St Vincent's Clinical School, UNSW Sydney, Sydney,Australia.

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

Corresponding Author: Martin C. Sadowski, Queensland University of Technol-ogy, Princess Alexandra Hospital, Translational Research Institute, PrincessAlexandra Hospital 199 Ipswich Rd, Level 1, Building 1, Brisbane, QLD 4102,Australia. Phone: 61-7-3443-7281; E-mail: martin.sadowski@qut.edu.au

doi: 10.1158/1541-7786.MCR-18-1147

�2019 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 17(5) May 20191166

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

characteristics of prostate cancer, fatty acid and cholesterol syn-thesis have become an attractive therapeutic target. However, theantineoplastic effects observed by inhibiting lipogenesis can berescued by the addition of exogenous lipids (6, 7), highlightinglipid uptake as a mechanism of clinical resistance to lipogenesisinhibitors and that lipid uptake capacity is sufficient to substitutefor the loss of lipogenesis. Indeed, it was recently reported thatlung cancer cells expressing a strong lipogenic phenotype gener-ated up to 70% of their cellular lipid carbon biomass fromexogenous fatty acids and only 30% from de novo synthesissupplied by glucose and glutamine as carbon sources (8).Although altered cellular lipid metabolism is a hallmark of themalignant phenotype, prostate cancer is unique in that it ischaracterized by a relatively low uptake of glucose and glycolyticrate compared with many solid tumors subscribing to the "War-burg effect" phenotype (9, 10). Concordantly, prostate cancercells showed a dominant uptake of fatty acids over glucose, withthe uptake of palmitic acid measured at �20 times higher thanuptake of glucose in both malignant and benign prostate cancercells (11). Furthermore, exogenous fatty acids are readily oxidizedby prostate cancer, reducing glucose uptake (12). Together, thesestudies demonstrate that exogenous uptake is a significant andpreviously underappreciated supply route of lipids in cancer cellswith a lipogenic phenotype.

Both healthy andmalignant prostate cells rely on androgens fora variety of physiologic processes, including several metabolicsignaling pathways. Androgens, through binding to the androgenreceptor (AR), transcriptionally regulate a multitude of pathways,including proliferation, differentiation, and cell survival of pros-tate cancer (13), with approximately equal numbers of genesactivated and suppressed by androgen-activated AR. Targeting ofthe AR signaling axis is the mainstay treatment strategy foradvanced prostate cancer. Although initially effective in suppres-sing tumor growth, patients inevitably develop castrate-resistantprostate cancer (CRPC), which remains incurable. Importantly,during progression to CRPC, survival and growth of prostatecancer cells remain dependent on AR activity, as demonstratedby treatment-resistance mechanisms involving AR mutation,amplification, and intratumoral steroidogenesis (reviewed inref. 14). Thus, identifying critical pathways regulated by ARmight provide novel therapeutic strategies to combat develop-ment of CRPC. Lipogenesis is a well-described AR-regulatedmetabolic pathway that supports prostate cancer cell growth byproviding fuel, membrane material, and steroid hormone pre-cursor (cholesterol). Androgens stimulate expression of FASNvia activation of sterol regulatory element-binding proteins(SREBP; ref. 15), lipogenic enzymes ACACA and ACLY andcholesterol synthesis enzymes HMGCS1 and HMGCR (3, 16).In contrast, the role and expression of lipid transporters andtheir regulation by AR in prostate cancer remain largely unchar-acterized (11, 17).

Our current understanding of lipid uptake is mostly derivedfrom studies in nonmalignant cells and tissues. The hydrophobicproperties of lipids allow for passive, nonspecific uptake viadiffusion into the cell. Selective, protein-mediated lipid uptakeinvolves receptor-mediated endocytosis of lipid transporters andtheir cognate lipoprotein cargo (18, 19), which contains variouslipid components (phospholipids, cholesterol esters, triacylgly-cerol) that can be internalized via lipoprotein receptors (LDLRand VLDLR) or scavenger receptors (SCARB1 and SCARB2).Various scavenger receptors have alsobeen shown tobe associated

with uptake of modified (acetylated or oxidized) LDL particles,including SCARF1, SCARF2, andCXCL16 (20, 21). Free fatty acidscan be taken up by a family of six fatty acid transport proteins(SLC27A1-6) as well as fatty acid translocase (FAT/CD36) andGOT2/FABPpm (17).

Taken together, it is becoming evident that enhanced lipogen-esis inprostate cancer development andprogression is not the solederegulated lipid supply pathway, and lipid uptakemight play animportant role in biochemical recurrence of prostate cancer. Thiswarranted a comprehensive investigation and delineation of lipiduptake and the lipid transporter landscape in prostate cancer aswell as its regulation by AR.

Materials and MethodsCell culture

The following cell lines were acquired from ATCC in 2010:LNCaP (CVCL_0395), C4-2B (CVCL_4784), and VCaP(CVCL_2235). Fibroblast-free DuCaP cells were a generous giftfromM.Ness (VTT Technical Research Centre of Finland). LNCaPand C4-2B cells were cultured in Roswell Park Memorial Institute(RPMI) medium (Thermo Fisher Scientific) supplemented with5% fetal bovine serum (FBS) until passage 45. DuCaP and VCaPcells were cultured in RPMI supplemented with 10% FBS untilpassages 40 and 45, respectively. The medium was changed every3 to 4 days. All cell lines were incubated at 37�C in 5%CO2. Cellswere passaged at approximately 80% confluency by trypsiniza-tion. Cell lines were genotyped in March 2018 by GenomicsResearch Centre (Brisbane) and routinely tested for Mycoplasmainfection.

For androgen and antiandrogen treatments, cells were seededin regular growth medium for 72 hours before media werereplaced with RPMI supplemented with 5% charcoal-dextranstripped serum (CSS; Sigma-Aldrich). After 48 hours, media werereplaced with fresh 5% CSS RPMI, and cells were treated witheither dihydroxytestosterone (DHT, 10 nmol/L) or syntheticandrogen R1881 (1 nmol/L) for 48 hours to activate AR signaling.AR-antagonist enzalutamide or bicalutamide (Selleck Chemicals)was used at 10 mmol/L.

Measurement of lipid content by quantitative fluorescentmicroscopy

Prior to seeding of LNCaP and C4-2B cells in 5%CSS RPMI at adensity of 4,000 and 3,000 cells/well, respectively, optical 96-wellplates (IBIDI) were coated with 150 mL Poly-l-ornithine (Sigma-Aldrich) as described previously (22). DuCaP andVCaP cells wereseeded in 10% CSS RPMI at a density of 15,000 cells/well. Aftertreatment as indicated, media were removed, cells were washedwith PBS once, fixed with 4% paraformaldehyde for 20 minutesat room temperature, and remaining aldehyde reacted with30 mmol/L glycine in PBS for an additional 30 minutes. NuclearDNA was then stained with 1 mg/mL 40,6-diamidino-2-phenylin-dole (DAPI, Thermo Fisher Scientific) and lipids were stainedwith 0.1 mg/mL Nile Red (Sigma-Aldrich) overnight as describedpreviously (23). Alternatively, free cholesterol was stained with50 mg/mL filipin (Sigma-Aldrich) for 40 minutes. More than500 cells/well were imaged using the InCell 2200 automatedfluorescence microscope system (GE Healthcare Life Sciences).Quantitative analysis of 1,500 cells/treatment (3 wells) wasperformed in at least two independent experiments with CellProfiler Software (Broad Institute; ref. 24).

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1167

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

Measurement of lipid uptake by quantitative fluorescentmicroscopy

Cells were seeded as described above. For quantifyingC16:0 fatty acid uptake, growth medium was exchanged with65 mL/well serum-free RPMI medium supplemented with 0.2%BSA (lipid-free; Sigma-Aldrich) and 5 mmol/L Bodipy-C16:0(Thermo-Fisher), and cells were incubated at 37�C for 1 hour.Cellular uptake of cholesterol was measured as described recent-ly (25). Briefly, medium was exchanged with serum-free 0.2%BSA RPMI media supplemented with 15 mmol/L NBD-cholesterol(22-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino-23,24-bisnor-5-cholen-3b-Ol) (Thermo Fisher Scientific), and cells were incu-bated at 37�C for 2 hours. For quantifying lipoprotein complexuptake, serum-free 0.2% BSA RPMI medium was supplementedwith1,10-dioctadecyl-3,3,30,30-tetramethylindocarbocyanine (DiI)-labeled acetylated LDL (Thermo Fisher Scientific, 15 mg/mL) orDiI-labeled LDL (Thermo Fisher Scientific; 15 mg/mL) and incu-bated at 37�C for 2 hours. After incubation, cells were washed andfixed as described above. Cellular DNA and F-actin was counter-stained with DAPI and Alexa Fluor 647 Phalloidin (Thermo FisherScientific). Image acquisition and quantitative analysis were per-formed as above.

RNA extraction and quantitative real-time PCR (qRT-PCR)Cells were seeded at a density of 9.0� 104 (LNCaP and C4-2B)

or 1.2 � 105 (DuCaP and VCaP) cells/well in 6-well plates(Thermo Scientific). Following completion of treatment, totalRNA was isolated using the RNEasy mini kit (Qiagen) followingthe manufacturer's instructions. RNA concentration was mea-sured using a NanoDrop ND-1000 Spectrophotometer (ThermoScientific), and RNA frozen at �80�C until further use.

Up to 2 mg of total RNA was used to prepare cDNAwith SensiFast cDNA synthesis kit (Bioline) according to themanufacturer's instructions and diluted 1:5. qRT-PCR wasperformed with SYBR-Green Master Mix (Thermo FisherScientific) using the ViiA-7 Real-Time PCR system (AppliedBiosystems). Determination of relative mRNA levels wascalculated using the comparative DDCt method (26), whereexpression levels were normalized relative to that of thehousekeeping gene receptor–like protein 32 (RPL32) for eachtreatment and calculated as fold change relative to the vehiclecontrol treatment. All experiments were performed indepen-dently in triplicate. Primer sequences are listed in Supplemen-tary Materials.

Protein extraction and Western blot analysisProteins for Western blotting were isolated by lysing cells in

radio immunoprecipitation buffer [RIPA, 25 mmol/L Tris, HClpH 7.6, 150 mmol/L NaCl, 1% NP-40, 1% sodium deoxycho-late, 0.1% SDS, one cOmplete EDTA-free Protease InhibitorCocktail tablet (Roche) per 10 mL, phosphatase inhibitors NaF(30 mmol/L), sodium pyrophosphate (20 mmol/L), b-glycero-phosphate (10 mmol/L), and Na vanadate (1 mmol/L)]. Withcells on ice, media were carefully removed, and cells werewashed with PBS. RIPA lysis buffer was added, and cells wereincubated for 5 minutes on ice before collection of proteinlysates. Protein concentration was measured using the PierceBCA Protein Assay kit according to the manufacturer's instruc-tions (Thermo Fisher Scientific).

Twenty micrograms of total protein/lane were separated bySDS-polyacrylamide gel electrophoresis (SDS-PAGE) using

NuPAGE 4-12% Bis-Tris SDS-PAGE Protein Gels (Thermo FisherScientific), and Western blot was completed using the Bolt MiniBlot Module (Thermo Fisher Scientific) according to the manu-facturer's instructions. Membranes were reacted overnight at 4�Cwith primary antibodies raised against LDLR (Abcam, ab52818)and SCARB1 (Abcam, ab217318) at a dilution of 1:1,000 fol-lowed by probing with the appropriate Odyssey fluorophore-labeled secondary antibody and visualization on the LI-COROdyssey imaging system (LI-CORBiotechnology). Protein expres-sion levels were quantified using Image Studio Lite (LI-CORBiotechnology), normalized relative to the indicated housekeep-ing protein, and expressed as fold changes relative to the controltreatment.

Cistrome analysis of AR ChIPseq peaksAR ChIPseq analysis used BED files (hg38) downloaded from

Cistrome (27) for the� bicalutamide-treated vehicle controls forthe ChIPseq data set, GSE49832 (28). The bedtools software tool(version 2.27.0) was used to identify AR ChIPseq peaks enrichedin regions 5KB upstream and also in a 25KB window aroundGencode transcripts (version 21).

RNA-seq analysisFor mRNA-seq, total cellular RNA was extracted using the

Norgen RNA Purification PLUS kit #48400 (Norgen BiotekCorp.) according to the manufacturer's instructions, includingDNase treatment. RNA quality and quantity were determinedon an Agilent 2100 Bioanalyzer (Agilent Technologies) andQubit. 2.0 Fluorometer (Thermo Fisher Scientific Inc.). Librarypreparation and sequencing was done at the Kinghorn Centrefor Clinical Genomics (KCCG, Garvan Institute, Sydney) usingthe Illumina TruSeq Stranded mRNA Sample Prep Kit with aninput of 1 mg total RNA (RIN > 8), followed by paired-endsequencing on an Illumina HiSeq2500 v4.0 (Illumina), multi-plexing 6 samples per lane and yielding about 30M reads persample. Raw data were processed through a custom-designedpipeline. Raw reads were trimmed using "TRIMGALORE" (29),followed by parallel alignments to the genome (hg38) andtranscriptome (Ensembl v77/Gencode v21) using the"STAR" (30) aligner and read quantification with "RSEM" (31).Differential expression between two conditions was calculatedafter between sample TMM normalization (32) using "edgeR"(ref. 33; no replicates: Fisher exact test; replicates: generallinear model) and is defined by an absolute fold change of>1.5 and a false discovery rate (FDR) corrected P < 0.05.Quality control of raw data included sequential mapping tothe ERCC spike-in controls, rRNA, and a comprehensive set ofpathogen genomes as well as detection and quantification of 30

bias. Heat maps were generated with a hierarchical clusteringalgorithm using completed linkage and Euclidean distancemeasures.

Microarray gene-expression profiling and analysisRNA from LNCaP tumor xenograft models of CRPC was

collected as described previously (34). RNA was preparedfor microarray profiling as described previously using acustom 180K Agilent oligo microarray (VPCv3, ID032034,GEO:GPL16604) (35). Probes significantly different betweentwo groups were identified with an adjusted P � 0.05, andan average absolute fold change of �1.5 (adjusted for an FDRof 5%).

Tousignant et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1168

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

Statistical analysisStatistical analyses were performed with GraphPad Prism 7.0

(GraphPad Software) and R Studio (RStudio). Data reported andappropriate statistical tests are included in figure legends.

ResultsAndrogens strongly increased cellular lipid content in AR-positive prostate cancer cells

Previous analysis by cellular Oil Red O staining and lipidchromatography of cellular extracts have demonstrated that

androgens strongly enhance lipogenesis and cellular lipid contentin prostate cancer cells, predominantly that of neutral lipids(triacylglycerols and cholesterol esters) stored in lipid dropletsand phospholipids and free cholesterol present in mem-branes (16, 36). Consistent with these findings, our quantitativefluorescence microscopy (qFM) assay (23) of Nile Red-stainedAR-positive prostate cancer cell lines LNCaP, C4-2B, VCaP, andDuCaP showed that synthetic androgen R1881 significantlyincreased cellular phospholipid and neutral lipid content as wellas lipid droplet number (Fig. 1). This stimulatory effect of andro-gen was also observed with DHT (Supplementary Fig. S1) and

Figure 1.

Androgens strongly increased lipid content of AR-positive prostate cancer cell lines. A, LNCaP, C4-2B, VCaP, and DuCaP cells were grown in CSS for 48 hoursand treated with 1 nmol/L R1881 or vehicle (Ctrl) for 48 hours. Fixed cells were stained with fluorescent lipid stain Nile Red, and cellular mean fluorescentintensities (MFI) of phospholipid content (top left) and neutral lipid content (top right) as well as mean cellular number of lipid droplets (bottom left) and meantotal cellular area of lipid droplets (bottom right) were measured by quantitative fluorescence microscopy (qFM). Representative 40� images of LNCaP cell areshown (blue, DNA, yellow, lipid droplets containing neutral lipids, scale bar, 20 mm). B, LNCaP cells were grown as described in A and treated with the indicatedandrogens in the presence or absence of enzalutamide (Enz, 10 mmol/L). Fixed cells were stained with filipin to label-free, unesterified cholesterol, and MFI ofcellular-free cholesterol was measured by qFM (n� 3,000 cells from 3 wells; significance calculated relative to vehicle (Ctrl): ns, not significant; , P < 0.001; orvehicle-treated LNCaP cells: #, P < 0.001, representative of 3 independent experiments). Representative 40� images of LNCaP cell are shown (blue, DNA, green,free cholesterol; scale bar, 10 mm).

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1169

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

Figure 2.

Androgens strongly increased lipid uptake.A, Tomeasure fatty acid uptake, indicated cell lines were grown in CSS for 48 hours and treated with 1 nmol/L R1881 inthe presence or absence of Enz (10 mmol/L) or vehicle (Ctrl) for 48 hours. Before fixation, cells were incubated with Bodipy-C16:0 for 1 hour, and lipid uptake wasmeasured by qFM (n� 3,000 cells from 3wells, mean� SD; one-way ANOVAwith Dunnett multiple comparisons test relative to cell line–specific control (Ctrl),ns, not significant; , P < 0.001, representative of 2 independent experiments). (Continued on the following page.)

Tousignant et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1170

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

mibolerone (data not shown) and blocked in the presence ofenzalutamide (Supplementary Fig. S1). Furthermore, qFM offilipin-stained LNCaP cells confirmed that androgens also increas-ed cellular levels of free, unesterified cholesterol (Fig. 1B), whichwas also blocked by enzalutamide. Although androgen-enhancedlipogenesis is a well-characterized fuel source for increased cel-lular lipid content, the role of lipid uptake in this process is stillpoorly understood.

Fatty acid, cholesterol, and lipoprotein uptake are increased byandrogens

To directly measure the stimulatory effect of androgens onlipid uptake, a series of lipid uptake assays (Fig. 2) was usedbased on qFM of fluorophore-labeled lipid probes (Bodipy-C16:0, NBD-cholesterol, DiI-LDL, and DiI-acetylated LDL). Asshown in Fig. 2A, androgen treatment (R1881) of four AR-positive prostate cancer cell lines (LNCaP, C4-2B, DuCaP, andVCaP) for 48 hours significantly increased uptake of Bodipy-C16:0. This effect was significantly blocked when cells werecotreated with enzalutamide. Notably, androgens alsoincreased uptake of Bodipy-C12:0 but not Bodipy-C5:0 (datanot shown and Supplementary Fig. S2), suggesting that cellularuptake of short-chain fatty acids is not androgen regulated.Similar to fatty acid uptake, androgens also significantlyincreased uptake of NBD cholesterol in AR-positive prostatecancer cells (Fig. 2B), and enzalutamide significantly sup-pressed this effect. Representative images of LNCaP cells showthat Bodipy-C16:0 and NBD cholesterol were readily incorpo-rated into lipid droplets (Fig. 2A and B).

The majority of serum lipids are transported as lipoproteinparticles (chylomicrons, VLDL, LDL, HDL), containing a com-plex mixture of apolipoproteins, phospholipids, cholesterol,and triacylglycerols which are taken up into cells by receptor-mediated endocytosis through cognate lipoprotein receptorssuch as the LDL receptor (LDLR) for LDL and scavenger receptorSCARB1 for acetylated LDL/HDL. Notably, in contrast to thecovalent Bodipy and NBD fluorophore tags on C16:0 andcholesterol, the DiI label is a noncovalently bound dye infusedinto the lipoprotein particles that dissociates after cellularuptake and lysosomal processing. As shown in Fig. 2C, andro-gens significantly enhanced uptake of DiI-complexed LDL andacetylated LDL in LNCaP cells in a dose-dependent manner,indicating a potential role for their cognate receptors LDLRand SCARB1.

Androgen-enhanced lipid uptake is independent of cell-cycleprogression and proliferation

Androgen-mediated activation of AR promotes G0/G1 to Sphase progression of the cell cycle and proliferation in prostatecancer cells (reviewed in refs. 13, 37, 38). Because proliferationrequires substantial membrane biogenesis for daughter cell

generation, it was possible that androgen-enhanced lipiduptake was not mediated directly through AR signaling butindirectly as a result of androgen-stimulated proliferation. Toaddress this possibility, LNCaP cells were synchronized inG0/G1 (>95% of cell population; Supplementary Fig. S3A) byincubation in CSS medium for 48 hours and treated for another24 hours with three different cell-cycle inhibitors, which uponandrogen (DHT) treatment-induced reentry into the cell cyclecaused arrest in G0/G1 (tunicamycin), S phase (hydroxyurea),or G2/M (nocodazole; Supplementary Fig. S3A). As shownin Fig. 3, lipid uptake of Bodipy-C16:0 and NBD cholesterolwas significantly and to a similar magnitude enhanced byandrogen in the presence of all three cell-cycle inhibitors whencompared with control. Flow cytometry of DNA content con-firmed cell-cycle arrest (Supplementary Fig. S3B) by the inhi-bitors, and IncuCyte cell confluence analysis demonstratedgrowth inhibition (Supplementary Fig. S3C), respectively.Thus, androgen regulation of lipid uptake is directly mediatedby AR throughout the cell cycle and is independent of cell-cycleprogression and proliferation. Notably, a time course experi-ment of DHT-treated G0/G1 synchronized LNCaP cells in thepresence of tunicamycin (Supplementary Fig. S3B; FACS)confirmed that the androgen-enhanced expression of classicAR-regulated genes KLK3/PSA (Supplementary Fig. S3D),TMPRSS2, and FKBP5 (data not shown) remained unaffectedunder the experimental conditions.

Delineating the lipid transporter landscape in prostate cancerAlthough the role of genes involved in DNL (e.g., ACLY,

ACACA, FASN, HMGCR) is well described in prostate cancer, andtheir overexpression is associated with tumor development, dis-ease progression, aggressiveness, and poor prognosis (reviewed inrefs. 1, 2), very little is known about the expression and functionalimportance of lipid transporters in prostate cancer, their regula-tion by androgens, and their clinical relevance. To delineate thelipid transporter landscape in prostate cancer, a panel of 44candidate lipid transporters was generated based on previouswork describing their lipid transport function in various humantissues (19, 39–44).

Transcriptomic analysis by RNA-seq revealed that 41 candidatelipid transporters were expressed in five prostate cancer cell lines(LNCaP, DuCaP, VCaP, PC-3, and Du145) under normal cultureconditions (a selection of 36 candidates is shown in Fig. 4A).Importantly, lipid transporters LDLR, SCARB1, SCARB2, andGOT2/FABPpm were robustly expressed at levels comparable withlipogenic genes HMGCR and FASN (Fig. 4A), whereas seventransporters, including CD36 and SLC27A6, displayed FPKMvalues <1 in the majority of cell lines. In addition, mRNA expres-sion of these 41 lipid transporters was independently detected inLNCaP and Du145 cells and six additional prostate cancer celllines (CWR22RV1, EF1, H660, LASCPC-01, NB120914, and

(Continued.) Representative 40� images of DuCaP cell are shown (blue, DNA; red, F-actin; green, lipid droplets containing C16:0-Bodipy; scale bar, 20 mm).B, Tomeasure cholesterol uptake, LNCaP cells were grown in CSS for 48 hours and treated with either 1 nmol/L R1881 or 10 nmol/L DHT in the presence orabsence of Enz (10 mmol/L). Before fixation, cells were incubated with NBD cholesterol for 2 hours and cellular levels were measured by qFM [n� 3,000 cellsfrom 3 wells, mean� SD; one-way ANOVAwith Dunnett multiple comparisons test relative to cell line–specific vehicle (Ctrl), or unpaired t test betweenandrogen treatment alone or in combination with enzalutamide; ns, not significant; , P < 0.0001, representative of 2 independent experiments].Representative 40� images of LNCaP cells are shown (blue, DNA; red, F-actin; green, lipid droplets containing NBD-cholesterol; scale bar, 20 mm). C, To measurelipoprotein uptake, LNCaP cells were grown in CSS for 48 hours and treated with increasing concentrations of DHT or 1 nmol/L R1881. Before fixation, cells wereincubated with DiI-LDL or DiI-acLDL for 2 hours, and lipoprotein uptake was measured by qFM [n� 3,000 cells from 3 wells; mean� SD; one-way ANOVAwithDunnett multiple comparisons test relative to control (Ctrl) in each respective cell line; , P < 0.0001, representative of 3 independent experiments].

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1171

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

NE1_3; ref. 45); John Lee, August 29, 2018), verifying mRNAexpression of these transporters in a total of nine prostate cancercell lines. Comparison of this list of lipid transporters with therecently delineated plasmamembrane proteome of eight prostatecancer cell lines, including LNCaP, Du145, and CWR22Rv1[ref. 45; John Lee, August 29, 2018) and previous work in LNCaPand CWR22Rv1 cells (17), confirmed the surface expression ofLDLR, GOT2, LRPAP1, LRP8, and SCARB2. In addition, ourproteomics analysis confirmed the exclusive expression ofSCARB1, SCARB2, LRPAP1, SLCA27A1, and SLC27A2 in themembrane fraction of LNCaP cells, whereas GOT2 was alsopresent in the soluble fraction (data not shown), which is con-sistent with its mitochondrial function. Western blot analysisdemonstrated the expression of LDLR and SCARB1 in cell lysatesof seven malignant and two nonmalignant prostate cell lines(Fig. 4B). Subsequently, expression of these lipid transporters inprostate cancer patient samples and clinical relevance was inves-tigated by analyzing published tumor transcriptome data setswith Oncomine. Comparison of primary, localized prostate can-cer versus normal prostate gland revealed that mRNA levels ofonly a few lipid transporters were significantly (P < 0.05) upre-gulated in primary prostate cancer, and no lipid transporter wassignificantly downregulated (Supplementary Fig. S4A–S4C).However, data mining of the reported proteome analysis ofprimary prostate cancer versus neighboring nonmalignant tis-sue (46) revealed that expression of 21 lipid transporters waslower in primary prostate cancer, whereas protein expression ofboth DNL enzymes FASN and HMGCR was increased by severalmagnitudes (Supplementary Fig. S4D). Although a measurabledegree of discordance betweenmRNA and protein levels has beenpreviously noted in integrated transcriptome and proteome stud-ies of prostate cancer (46, 47), the proteomics data suggested that

lipid uptake is reduced and DNL is increased in primary prostatecancer when compared with normal prostate gland. In contrast,mRNA levels of several lipid transporters were significantly upre-gulated in metastatic tumor samples compared with primary site(48), including SLC27A1, SLC27A3, SCARB1, and LDLR (Fig. 4C).Concordantly, analysis of the proteome comparison of localizedprostate cancer versus bone metastasis (49) demonstrated thatexpression of 16 lipid transporters and FASN was higher in bonemetastases (Fig. 4E), suggesting that tumor lipid supply frombothuptake and DNL was increased. The lipoprotein transportersLDLR and SCARB1 were further investigated across other prostatecancer patient cohorts in Oncomine, including the Varamb-ally (50) and La Tulippe (51) data sets. Both lipid transportermRNAs were found to be significantly upregulated in samplesfrom prostate cancer metastases when compared with primarytumors (Fig. 4E). Together, independently published data andourown analyses confirmed the mRNA, protein, and plasma mem-brane expression of several lipid transporters in prostate cancercell lines and patient-derived tumor samples. Importantly, ouranalysis demonstrated that this route of lipid supply is clinicallysignificant during disease progression and is associated withmetastasis to the bone.

Androgens regulate the expression of several lipid transportersAs shown above, androgens strongly enhance lipid uptake in

AR-positive prostate cancer cell lines.However, our current under-standing of the AR regulation of lipid transporters is very limit-ed (17). We initiated a comprehensive analysis of androgen-regulated lipid transporters by searching for AR binding siteswithin a 25-kb window of the gene sequence and a 5-kb windowupstream of the protein start codon of 45 candidate lipid trans-porters in the reported AR ChIPseq data set of LNCaP cells treated

Figure 3.

Androgen-enhanced lipid uptake is independent of cell-cycle progression and proliferation. LNCaP cells were synchronized in G0–G1 by androgen deprivation(CSS for 48 hours) followed by treatment with tunicamycin (1 mg/mL), hydroxyurea (1 mol/L), or nocodazole (25 mg/mL) for another 24 hours, placing cell-cycleblocks in G0–G1, S phase, and mitosis, respectively. Cell-cycle reentry and progression to the respective cell-cycle block was stimulated by DHT (10 nmol/L). After24 hours, cholesterol (NBD-cholesterol; left) and fatty acid uptake (Bodipy-C16:0; right) was measured by qFM [n� 3,000 cells from 3 wells; mean� SD; one-way ANOVAwith Dunnett multiple comparisons test relative to cell line–specific vehicle (Ctrl), or unpaired t test between androgen treatment alone or incombination with cell-cycle inhibitor; ns, not significant; , P < 0.0001; , P < 0.01, representative of 2 independent experiments].

Tousignant et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1172

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

with AR-antagonist bicalutamide (28). As shown in Fig. 5A, 19and 27 lipid transporters showed enrichment of AR ChIPseqpeaks in the 5-kb and 25-kb windows, respectively, which wasreduced in the presence of bicalutamide. Consistent with itsreported androgen regulation (17), AR ChIPseq peaks weredetected in the 25-kb window of the GOT2 gene, which wereabsent after bicalutamide treatment. For comparison, AR-regulated lipogenesis genes ACACA, FASN, and HMGCR (52)also showed reduced enrichment of AR ChIP peaks with bicalu-tamide. Notably, lipid transporter genesmight contain additionalAR ChIP peaks outside the cutoff of 25 kb. Alternatively, theabsence of AR ChIP peaks might indicate that they are indirectlyregulated by androgen-activated transcription factors, e.g., sterolelement-binding proteins 1 and 2 (SREPB1/2). Indeed, the LDLRgene lacks AR ChIP peaks but contains flanking sterol regulatoryelements and is positively regulated by SREBP1/2 (53, 54).

Next, mRNA transcript levels of our panel of 44 candidate lipidtransporters were measured by RNA-seq in three AR-positive,androgen-sensitive prostate cancer cell lines (LNCaP, DuCaP, andVCaP) under conditions identical to the lipid content and uptakestudies shown above (androgen deprivation in CSS for 48 hoursand treatment with either vehicle or DHT (10 nmol/L) for 48hours). As a control, AR function was blocked with enzalutamidein the presence and absence ofDHT. As shown in Fig. 5B, RNA-seqanalysis demonstrated that expression of 36 lipid transportergenes was altered by androgen treatment in LNCaP cells. Cho-lesterol efflux pump ABCA1 and scavenger receptor SCARF1mRNA was significantly reduced by androgens, a response thatwas antagonized by enzalutamide. In contrast, DHT significantlyincreased the expression of fatty acid transporters (GOT2,SLC27A3, SLC27A4, SLC27A5, and CD36) and lipoprotein trans-porters (LDLR, LRP8, and SCARB1), which was also blocked byenzalutamide. Receptor-mediated endocytosis of lipoprotein par-ticles throughLDLR, VLDLR, SCARB1, SCARB2, andLDL receptor-related proteins (LRP1-12 and LRPAP1) converges in lysosomesfor lipolysis and release of free cholesterol and fatty acids into thecytoplasm through their respective efflux pumps. Consistent withthis,mRNA for lysosomal cholesterol efflux transporterNPC1wasalso increased by DHT. Similar effects of DHT regulation of lipidtransporter expression were observed in DuCaP and VCaP cells,with the exception of SLC25A5, LRP8, and SCARB1, which wererepressed by DHT (Supplementary Fig. S5A). qRT-PCR showedsignificantly increased mRNA expression of the lipoprotein trans-port receptors LDLR (P < 0.0001) and VLDLR (P < 0.0001) inLNCaP cells treated with 1 nmol/L R1881 (Fig. 5C, top panel).Although R1881 also enhanced expression of SCARB1 andSLC27A4, there was no significant change in expression (P >0.05), although both showed similar trends to LDLR and VLDLR(Fig. 5C, bottom). Cotreatment with enzalutamide (10 mmol/L)blocked the increase in lipid transporter expression (Fig. 5C).Analysis of DuCaP cells revealed similar results, demonstratingthat the mRNA expression of the majority of tested lipid trans-porters was significantly enhanced by androgens (SupplementaryFig. S5B). Western blot analysis demonstrated that LNCaP cellsexposed to 10 nmol/L DHT showed an almost 2-fold increase inLDLR protein expression, which was suppressed to levels similarto vehicle control when cotreated with enzalutamide (Fig. 5D,left). A similar trend of increased protein expression in cellstreated with DHT was observed for SCARB1 (Fig. 5D, right).Cellular localization of LDLR protein in response to R1881(1 nmol/L) using immunofluorescent microscopy showed that

androgen treatment resulted in significantly increased expres-sion of LDLR at the cellular periphery (plasma membrane;Fig. 5E), which was blocked by enzalutamide (10 mmol/L),confirming that AR signaling enhanced the abundance of LDLRprotein at the cell surface. Finally, review of our previouslyreported longitudinal LNCaP xenograft study (34) revealedthat mRNA levels of LDLR, VLDLR, SCARB1, SLC27A5, andSLC27A6 were reduced 7 days after castration (nadir) whencompared with castration-na€�ve tumors (intact; Fig. 5F), whichis consistent with their positive AR regulation of expression inLNCaP cells in vitro shown above.

DiscussionIncreased activation of de novo lipogenesis is a well-established

metabolic phenotype in prostate cancer and other types of solidcancer; however, therapeutic inhibition of DNL alone has so farhad only limited clinical success as therapy against neoplasticdisease. TargetingDNL inpreclinical cancermodels, including ourown work in prostate cancer, demonstrated that inhibition ofDNL leading to lipid starvation can be efficiently rescued byexogenous lipids (55). Furthermore, obesity has been associatedwith more aggressive disease at diagnosis and higher rate ofrecurrence in prostate cancer patients (reviewed in refs. 56, 57).Thus, exogenous lipids may play a much more significant role inprostate cancer and other types of cancers than previouslyacknowledged. Indeed, recent estimates derived from studies inlung cancer cells with a similar lipogenic phenotype as prostatecancer cells suggested that 70% of lipid carbon biomass is derivedfrom exogenous lipids and only 30% from DNL (8). Althoughandrogens are known to activate DNL in prostate cancer (58),little is known about lipid uptake in this context.

In this study, we evaluated the effect of androgen treatment onlipid content (free cholesterol, neutral and phospholipids andlipid droplets) and lipid uptake of several lipid probes (C5:0,C12:0, and C16:0 fatty acids, cholesterol, LDL, and acetylatedLDL) in a panel of prostate cancer cells. By applying cutting-edge automated quantitative fluorescence microscopy and imageanalysis, we provide the functionally most comprehensive anal-ysis of lipid uptake in prostate cancer cells to date. Our workdemonstrates that androgen significantly enhanced cellularuptake of LDL particles as well as free fatty acids and cholesteroland their subcellular storage in lipid droplets. Consistent withthis, we showed a concordant increase in cellular phospholipids(membrane), neutral lipids (cholesterol-FA esters and TAGsstored in lipid droplets), and free cholesterol (Fig. 1A and B),which is a major component of cell membranes and essential formembrane structure and functional organization as well as aprecursor for steroidogenesis reviewed in ref. 59. Although ourwork did not delineate the relative contributions of variousanabolic and catabolic lipid metabolism processes to the netincrease in cellular lipid content in response to androgen treat-ment, e.g., enhanced lipid uptake and lipogenesis (58) versus fattyacid oxidation, phospholipid degradation, steroidogenesis, andlipid efflux, it nevertheless shows that androgens caused a strongand expansive increase in lipid uptake across various lipid species.Our ongoing work suggests that lipid uptake has a higher supplycapacity than DNL in prostate cancer cells (data not shown),which is consistent with the ability of exogenous lipids to effi-ciently recueDNL inhibition (55) and recent work estimating that70% of carbon lipid biomass is derived from exogenous lipids in

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1173

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

lung cancer cells expressing the lipogenic phenotype (8). Criti-cally, we demonstrated that androgen-enhanced lipid uptake isdirectly mediated by AR signaling and independent of its stim-ulatory effect on cell-cycle progression and proliferation (13, 37),i.e., androgen-enhanced fatty acid and cholesterol uptakeremained unaffected in prostate cancer cells arrested in G0/G1,S phase, or G2/M and in the absence of cell growth. This suggeststhat AR-regulated lipid uptake is maintained throughout the cellcycle, is not part of a cell-cycle–specific AR subnetwork (60) and isnot indirectly caused by lipid biomass demand of daughter cellgeneration.

Importantly, this work for the first time comprehensivelyelucidated the lipid transporter landscape in prostate cancer.Recent integrative omics studies of prostate cancer patient sam-ples highlighted a measurable degree of discordance betweengenomics, epigenetics, transcriptomics, and proteomics, i.e., thatgene copy number, DNA methylation, and mRNA levels did notreliably predict proteomic changes (46, 47). In addition, theplasma membrane localization of most candidate lipid transpor-ters remains to be confirmed in prostate cancer (17, 61) despiterecent progress in overcoming technical limitations challengingthe comprehensive delineation of the surface proteome of pros-tate cancer cells (45). By comparing transcriptomic andproteomicanalyses of cell lines, tumor xenografts, and patient samples, ourwork has conclusively demonstrated robust mRNA expression of34 lipid transporters in multiple prostate cancer cell lines andexpression of six lipid transporter proteins in the membranefraction of LNCaP cells, of which plasma membrane expressionwas independently confirmed for LDLR, GOT2, LRPAP1, LRP8,and SCARB2 in eight prostate cancer cell lines (17, 45); John Lee,August 29, 2018]. Our data mining of previously reported pros-tate cancer tumor proteomes (normal gland vs. primary prostatecancer and primary prostate cancer vs. bone metastasis; refs. 46,49) demonstrated that the expression of the lipid transporterlandscape substantially changes during prostate cancer progres-sion from localized disease (21 lipid transporters downregulated¼ low lipid uptake) to bone metastatic disease (16 lipid trans-porters upregulated ¼ high lipid uptake). For comparison, theenhanced expression of lipogenic enzymes suggested that lipidsynthesiswasupregulated throughout prostate cancer progressionfrom primary to metastatic disease. Our Oncomine analysisrevealed a similar trend in the mRNA expression of lipid trans-porters in three prostate cancer patient sample cohorts reportedpreviously (Grasso 2012, Varambally 2005, and La Tulippe2002). If, and to what extent, the extremely lipid-rich environ-ment of the bone marrow (50%–70% adiposity in adult men;ref. 62) is associatedwith enhanced lipid uptake in prostate cancerbone metastases remains to be investigated, including the possi-bility that the increased incidence of prostate cancermetastases tobone is linked to high levels of adiposity and specific lipid specieswithin bone marrow, which provide increased stimulus for moreaggressive growth andprotumorgenic lipid signaling ofmetastaticprostate cancer. Of the 22 bone metastasis proteomes that wereanalyzed, 16 were from patients after long-term ADT and classi-fied as CRPC, with one short-term ADT and five hormone-na€�vecases, yet all shared the same general features, including enhancedlipid transport and fatty acid oxidation (49), suggesting thatcastration-resistant bone metastases rely on similar mechanismsfor growth as hormone-na€�vemetastatic bone tumors. Contrary toabove reports, an integrated transcriptomics and lipidomics studyhighlighted increased mRNA levels of SCARB1, GOT2, and

SLC27As 2, 4, and 5 as well as polyunsaturated fatty acid (PUFA)accumulation in 20 paired localized primary tumors comparedwith matched adjacent nonmalignant prostate tissue (63).Although PUFA synthesis from essential fatty acids a-linolenicacid and linoleic acids remained transcriptionally unchanged, theauthors proposed that increased phospholipid uptake throughSCARB1 caused intratumoral PUFA enrichment in localized

Figure 4.

Delineation of the lipid transporter landscape in prostate cancer (PCa;A)mRNA expression levels (mean FPKM, fragments per kilobase million; n¼ 2)of the indicated candidate lipid transporters and 2 lipogenic genes (FASN andHMGCR) were measured by RNA-seq in the five indicated prostate cancer celllines grown in their respective maintenance media. B,Western blotconfirmed the protein expression of LDLR and SCARB1 in the seven indicatedprostate cancer cell lines and in two nonmalignant prostate cell lines (RWPE-1 and BHP-1) grown in their respective maintenance media. GAPDH was usedas a loading control. A representative blot of two independent experiments isshown. (Continued on the following page.)

Tousignant et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1174

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

prostate cancer; however, this hypothesis still awaits experimentalconfirmation. The reason for discordance between both studiesregarding lipid uptake in localized prostate cancer is unclear, but itis noteworthy that the activity of lipid transporters is also regu-lated through changes in their subcellular localization, highlight-ing the need for an integrated analysis of the cell-surface proteomeand tumor lipidome in prostate cancer. We conclude that LDLR,GOT2, LRPAP1, LRP8, SCARB1, and SCARB2 are high-confidence

lipid transporters that are associated with prostate cancer diseaseprogression and bone metastasis, but further work is needed tofully delineate the lipid transporter proteome at the plasmamembrane in prostate cancer.

We have provided the most comprehensive functional analysisof lipid uptake in prostate cancer cells to date and demonstratedthat androgens strongly enhanced lipid uptake of fatty acids,cholesterol, and lipoprotein particles LDL and acetylated LDL in

Figure 4.

(Continued. ) C,Oncomine analysis of candidate lipid transporters in the Grasso data set (48) comparing gene expression of localized, primary prostate cancerversus metastatic prostate cancer. D, Protein expression analysis of indicated 18 lipid transporters and two lipogenesis enzymes (FASN and HMGCR) in pairedpatient samples of localized primary tumor and (blue) and bone metastasis (red) in the Iglesias-Gato proteome data set (49). E, Gene expression of LDLR (top)and SCARB1 (bottom) was compared in primary vs. metastatic prostate cancer in the Grasso, Varambally, and LaTulippe cohorts (mean� SD; unpaired t test; ns,not significant; , P < 0.0001; , P < 0.001; , P < 0.01; , P < 0.05).

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1175

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

Figure 5.

Androgens regulated the expression of lipid transporters. A, AR ChIPseq peak enrichment analysis of 42 lipid transporter genes and six lipogenesisgenes (ACLY, ACSS2, ACACA, FASN, HMGCS1, and HMGCR) in the Ramos-Montoya data set of LNCaP cells treated with bicalutamide (BIC) comparedwith vehicle control (VEH; 28). The number of peaks is highlighted by the bubble size and the enrichment score by the gray scale. B, LNCaP cellswere grown in CSS for 48 hours and treated with 10 nmol/L DHT in the absence or presence of Enz (10 mmol/L) or vehicle (Ctrl) for 48 hours.mRNA expression of indicated lipid transporters was analyzed by RNA-seq, and heat maps were generated with a hierarchical clustering algorithmusing completed linkage and Euclidean distance measures and scaled by z score (red, positive z score; blue, negative z score). C, mRNA expression ofindicated lipid transporters was measured by qRT-PCR in LNCaP cells grown for 48 hours in CSS followed by treatment with 1 nmol/L R1881in the presence or absence of Enz (10 mmol/L) for an additional 48 hours [n ¼ 3; mean � SD; one-way ANOVA with Dunnett multiple comparisonstest relative to vehicle (Ctrl ns, not significant; , P < 0.0001)]. D, LNCaP cells were grown in CSS for 48 hours and treated with 10 nmol/L DHT inthe presence or absence of Enz (10 mmol/L). Protein expression was measured by Western blot analysis and quantitated by densitometry analysis, andtotal protein levels were normalized to loading control [gamma tubulin; mean � SD; one-way ANOVA with Dunnett multiple comparisons test relativeto vehicle control (CSS); a representative blot of three independent experiments is shown]. E, Cells were treated as described above.After fixation, cells were incubated with LDLR primary antibody for 24 hours and counterstained with appropriate secondary antibody.Protein expression was measured by qFM (blue, DAPI; red, LDLR). (Continued on the following page.)

Tousignant et al.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1176

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

AR-positive prostate cancer cell lines (summarized in Fig. 6).Previous work indicated that expression of GOT2/FABPpm isenhanced by androgens and increases the cellular uptake ofmedium- and long-chain fatty acids in LNCaP and CWR22Rv1prostate cancer cells (17). Our comprehensive analyses of ARbinding sites (ChIPseqpeaks), RNA-seq (of threeDHT-treatedAR-positive prostate cancer cell lines), qRT-PCR, Western blot, andDNA microarray of LNCaP tumor xenograft (34) revealed thatan equal number of lipid transporters are activated and sup-pressed by androgens. AR-negative malignant and nonmalignantprostate cell lines (PC-3, Du145, and BHP-1) show avid lipiduptake (data not shown) and expression of transporters (Fig. 4Aand B). Thus, it is likely that other signaling pathways regulatelipid supply in these cell lines. Furthermore, after using theindependently confirmed plasma membrane expression (45) asa high-confidence filter, we conclude that LRPAP1 and SCARB2are androgen-suppressed and LRP8, SCARB1, LDLR, and GOT2are androgen-enhanced surface lipid transporters in prostatecancer cells. Interestingly, GOT2 (mitochondrial aspartate ami-notransferase) is better known for its role in amino acid metab-olism, the cytoplasm–mitochondria malate–aspartate shuttle,and the urea and tricarboxylic acid cycles. This suggests themoonlighting of metabolic enzymes in other subcellular com-partments (64), e.g., the plasma membrane (17, 61) and, strik-ingly, with additional substrate specificities and catalytic activi-ty (65). Thus, there is a possibility that future studies will discoveradditional proteins involved in lipid uptake due to their plasmamembrane expression.

Targeting cholesterol homeostasis in prostate cancer as a ther-apeutic strategy to delay development of CRPC has recentlyreceived increasing attention (66–68). Cholesterol is a precursorof steroid hormone synthesis, and we previously showed thatprogression to CRPC is associated with increased intratumoralsteroidogenesis of androgens (34). Hypercholesterolemia hasbeen reported to enhance LNCaP tumor xenograft growthand intratumoral androgen synthesis (69), and monotherapyagainst dietary cholesterol adsorption in the intestine with eze-timibe (67) or de novo cholesterol synthesis with simvastatin (66)reduced LNCaP tumor xenograft growth, androgen steroidogen-esis, and delayed development of CRPC. Furthermore, targetingcholesterol uptake via SCARB1 antagonism with ITX5061reduced HDL uptake (but not LDL) in LNCaP, VCaP, andCRW22Rv1 cells and sensitized CWR22Rv1 tumor orthografts toADT (68). Comparatively, the same study showed that ITX5061conferred stronger growth inhibition than simvastatin in LNCaPand CWR22Rv1 cells (68) under hormone-deprived conditions,suggesting that cholesterol uptake via SCARB1 is a significantsupply route in this prostate cancer model. Due to the coexpres-sion of multiple lipoprotein transporters (LDLR, VLDLR,SCARB1, and LRP1-12) in conjunction with increased cholesterolsynthesis in prostate cancer, novel cotargeting strategies antago-nizing this cholesterol supply redundancy might have profoundsynergies in extending the efficacy of ADT and delaying thedevelopment of CRPC. Such cotreatment strategies could includesimvastatin in combination with specific inhibitors of lipid pro-cessing in the lysosome, which is a critical hub for lipid uptake

Figure 6.

AR regulates lipid uptake and lipogenesis.Schematic representation of cellular supplypathways of cholesterol and fatty acids inprostate cancer cells (transporter-mediateduptake, lipogenesis, passive diffusion,tunneling nanotubes). Lipid transporters andlipogenic enzymes whose expression isincreased or decreased by androgens arehighlighted in red and blue, respectively. Lipidtransporters without confirmed surfaceexpression in prostate cancer are marked bylighter shades of red and blue. Only lipidtransporters with confirmed mRNA andprotein expression in cell lines and patientsamples are shown.

Figure 5.

(Continued. ) F,mRNA expression analysis of indicated lipid transporters and lipogenic genes in paired LNCaP tumor xenografts before (intact) and 7 days aftercastration (nadir) of our previously reported longitudinal LNCaP tumor progression data set (34); , P < 0.05; , P < 0.01; , P < 0.001.

Androgen Regulation of Lipid Uptake in Prostate Cancer

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1177

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

through endocytosis, including phagocytosis, pinocytosis, andreceptor-mediated endocytosis. The latter pathway is used by allmajor lipoprotein receptors, including LDLR, VLDLR, SCARB1,and the LRPs 1–12, andwas the focus of a recently started phase I/II clinical trial (NCT03513211). Strategies of cotargeting lipiduptake and synthesis with repurposed drugs are currently underinvestigation by our group and show very promising and potentantineoplastic synergies in preclinical models of advanced pros-tate cancer.

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

Authors' ContributionsConception and design: K.D. Tousignant, L.K. Philp, C.C. Nelson,M.C. SadowskiDevelopment of methodology: M.C. SadowskiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K.D. Tousignant, A. Rockstroh, S.J. McPherson,M.E. Dinger, C.C. Nelson, M.C. Sadowski

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): K.D. Tousignant, A. Rockstroh, A. Taherian Fard,M.L. Lehman, C. Wang, L.K. Philp, M.C. SadowskiWriting, review, and/or revision of the manuscript: K.D. Tousignant,M.L. Lehman, S.J. McPherson, L.K. Philp, C.C. Nelson, M.C. SadowskiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): L.K. Philp, N. Bartonicek, M.C. SadowskiStudy supervision: L.K. Philp, M.E. Dinger, C.C. Nelson, M.C. Sadowski

AcknowledgmentsThis study was supported by the Movember Foundation and the Prostate

Cancer Foundation of Australia through a Movember Revolutionary TeamAward (K.D. Tousignant, A. Rockstroh, A. Taherian Fard, M.L. Lehman,C. Wang, S.J. McPherson, L. Philp, C.C. Nelson, and M.C. Sadowski).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received October 24, 2018; revised January 3, 2019; accepted February 21,2019; published first February 26, 2019.

References1. Swinnen JV, Brusselmans K, Verhoeven G. Increased lipogenesis in

cancer cells: new players, novel targets. Curr Opin Clin NutrMetab Care 2006;9:358–65.

2. Menendez JA, Lupu R. Fatty acid synthase and the lipogenicphenotype in cancer pathogenesis. Nat Rev Cancer 2007;7:763–77.

3. Fritz V, Benfodda Z, Rodier G, Henriquet C, Iborra F, Avanc�es C, et al.Abrogation of de novo lipogenesis by stearoyl-CoA desaturase 1 inhibitioninterferes with oncogenic signaling and blocks prostate cancer progressionin Mice. Mol Cancer Ther 2010;9:1740–54.

4. Deep G, Schlaepfer I. Aberrant lipidmetabolism promotes prostate cancer:role in cell survival under hypoxia and extracellular vesicles biogenesis. Int JMol Sci 2016;17:1061.

5. Flavin R, Zadra G, Loda M. Metabolic alterations and targeted therapies inprostate cancer. J Pathol 2011;223:283–94.

6. Kuemmerle NB, Rysman E, Lombardo PS, Flanagan AJ, Lipe BC,WellsWA,et al. Lipoprotein lipase links dietary fat to solid tumor cell proliferation.Mol Cancer Ther 2011;10:427–36.

7. Griffiths B, Lewis CA, Bensaad K, Ros S, Zhang Q, Ferber EC, et al. Sterolregulatory element binding protein-dependent regulation of lipidsynthesis supports cell survival and tumor growth. Cancer Metab2013;1:3.

8. Hosios AM, Hecht VC, Danai LV, Johnson MO, Rathmell JC, SteinhauserML, et al. Amino acids rather than glucose account for the majority of cellmass in proliferating mammalian cells. Developmental Cell 2016;36:540–9.

9. Effert PJ, Bares R, Handt S, Wolff JM, B€ull U, Jakse G. Metabolic imaging ofuntreated prostate cancer by positron emission tomography with sup 18fluorine-labeled deoxyglucose. The J Urol 1996;155:994–8.

10. Zadra G, Photopoulos C, Loda M. The fat side of prostate cancer.Biochim Biophys Acta 2013;1831:1518–32.

11. Liu Y, Zuckier LS, Ghesani NV. Dominant uptake of fatty acid over glucoseby prostate cells: a potential new diagnostic and therapeutic approach.Anticancer Res 2010;30:369–74.

12. Schlaepfer IR, Glod�e LM, Hitz CA, Pac CT, Boyle KE, Maroni P, et al.Inhibition of lipid oxidation increases glucose metabolism and enhances2-Deoxy-2-[(18)F]Fluoro-D-glucose uptake in prostate cancer mousexenografts. Mol Imaging Biol 2015;17:529–38.

13. Lonergan PE, Tindall DJ. Androgen receptor signaling in prostate cancerdevelopment and progression. J Carcinogen 2011;10:20.

14. Dutt SS, Gao AC. Molecular mechanisms of castration-resistant prostatecancer progression. Fut Oncol 2009;5:1403–13.

15. HeemersH, Verrijdt G,Organe S, Claessens F, HeynsW, VerhoevenG, et al.Identification of an androgen response element in intron 8 of the sterol

regulatory element-bindingprotein cleavage-activatingprotein gene allow-ing direct regulation by the androgen receptor. J Biol Chem 2004;279:30880–7.

16. Swinnen JV, Van Veldhoven PP, Esquenet M, Heyns W, Verhoeven G.Androgens markedly stimulate the accumulation of neutral lipids in thehuman prostatic adenocarcinoma cell line LNCaP. Endocrinology 1996;137:4468–74.

17. Pinthus JH, Lu JP, Bidaisee LA, LinH, Bryskine I, Gupta RS, et al. Androgen-dependent regulation of medium and long chain fatty acids uptake inprostate cancer. Prostate 2007;67:1330–8.

18. Sahoo S, Aurich MK, Jonsson JJ, Thiele I. Membrane transporters in ahuman genome-scale metabolic knowledgebase and their implications fordisease. Front Physiol 2014;5:91.

19. Doege H, Stahl A. Protein-mediated fatty acid uptake: novel insights fromin vivo models. Physiology 2006;21:259–68.

20. Tamura Y, Osuga J, Adachi H, Tozawa R, Takanezawa Y, Ohashi K, et al.Scavenger receptor expressed by endothelial cells I (SREC-I) mediates theuptake of acetylated low density lipoproteins by macrophages stimulatedwith lipopolysaccharide. J Biol Chem 2004;279:30938–44.

21. Miller YI, Choi SH, Fang L, Tsimikas S. Lipoprotein modification andmacrophage uptake: role of pathologic cholesterol transport in atherogen-esis. Subcell Biochem 2010;51:229–51.

22. LiberioMS, SadowskiMC, Soekmadji C, Davis RA, Nelson CC. Differentialeffects of tissue culture coating substrates on prostate cancer cell adherence,morphology and behavior. PLoS One 2014;9:e112122.

23. Levrier C, Sadowski MC, Nelson CC, Healy PC, Davis RA. DenhaminolsA-H, dihydro-beta-agarofurans from the endemic Australian rainforestplant Denhamia celastroides. J Nat Prod 2015;78:111–9.

24. Kamentsky L, Jones TR, Fraser A, Bray MA, Logan DJ, Madden KL,et al. Improved structure, function and compatibility for CellProfiler:modular high-throughput image analysis software. Bioinformatics2011;27:1179–80.

25. Egbewande FA, Sadowski MC, Levrier C, Tousignant KD, White JM, CosterMJ, et al. Identification of gibberellic acid derivatives that deregulate cho-lesterol metabolism in prostate cancer cells. J Nat Prod 2018;81:838–45.

26. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparativeC(T) method. Nat Protoc 2008;3:1101–8.

27. Mei S, QinQ,WuQ, SunH, Zheng R, Zang C, et al. CistromeData Browser:a data portal for ChIP-Seq and chromatin accessibility data in human andmouse. Nucleic Acids Res 2017;45:D658–62.

28. RamosMontoya A, Lamb AD, Russell R, Carroll T, Jurmeister S, Galeano-Dalmau N, et al. HES6 drives a critical AR transcriptional programme toinduce castrationresistant prostate cancer through activation of anE2F1mediated cell cycle network. EMBO Mol Med 2014;6:651–61.

Mol Cancer Res; 17(5) May 2019 Molecular Cancer Research1178

Tousignant et al.

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

29. Krueger F. TrimGalore. 2012; Available from: http://www.bioinformatics.babraham.ac.uk/projects/trim_galore/.

30. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR:ultrafast universal RNA-seq aligner. Bioinformatics 2013;29:15–21.

31. Li B,DeweyCN. RSEM: accurate transcript quantification fromRNA-Seq datawith or without a reference genome. BMC Bioinformatics 2011;12:323.

32. RobinsonMD, Oshlack A. A scaling normalizationmethod for differentialexpression analysis of RNA-seq data. Genome Biol 2010;11:R25.

33. RobinsonMD,McCarthyDJ, SmythGK. edgeR: a Bioconductor package fordifferential expression analysis of digital gene expression data. Bioinfor-matics 2010;26:139–40.

34. Locke JA, Guns ES, Lubik AA, Adomat HH, Hendy SC, Wood CA, et al.Androgen levels increase by intratumoral de novo steroidogenesis duringprogression of castration-resistant prostate cancer. Cancer Res 2008;68:6407–15.

35. Sieh S, Taubenberger AV, Rizzi SC, Sadowski M, LehmanML, Rockstroh A,et al. Phenotypic characterization of prostate cancer LNCaP cells culturedwithin a bioengineered microenvironment. PloS One 2012;7:e40217.

36. Swinnen JV, Esquenet M, Goossens K, Heyns W, Verhoeven G. Androgensstimulate fatty acid synthase in the human prostate cancer cell line LNCaP.Cancer Res 1997;57:1086–90.

37. Heinlein CA, Chang C. Androgen receptor in prostate cancer.Endocrine Rev 2004;25:276–308.

38. Balk SP, Knudsen KE. AR, the cell cycle, and prostate cancer. Nucl ReceptorSignal 2008;6:e001.

39. Anderson CM, Stahl A. SLC27 fatty acid transport proteins. Mol AspectsMed 2013;34:516–28.

40. Goldstein JL, Anderson RG, Brown MS. Receptor-mediated endocytosisand the cellular uptake of low density lipoprotein. Ciba Found Symp1982:77–95.

41. Go GW, Mani A. Low-density lipoprotein receptor (LDLR) family orches-trates cholesterol homeostasis. Yale J Biol Med 2012;85:19–28.

42. Wang MD, Kiss RS, Franklin V, McBride HM, Whitman SC, Marcel YL.Different cellular traffic of LDL-cholesterol and acetylated LDL-cholesterolleads to distinct reverse cholesterol transport pathways. J Lipid Res 2007;48:633–45.

43. Kennedy BE, Charman M, Karten B. Niemann-pick type C2 protein con-tributes to the transport of endosomal cholesterol to mitochondria with-out interacting with NPC1. J Lipid Res 2012;53:2632–42.

44. Nath A, ChanC. Genetic alterations in fatty acid transport andmetabolismgenes are associated with metastatic progression and poor prognosis ofhuman cancers. Sci Rep 2016;6:18669.

45. Lee JK, Bangayan NJ, Chai T, Smith BA, Pariva TE, Yun S, et al. Systemicsurfaceome profiling identifies target antigens for immune-based therapyin subtypes of advanced prostate cancer. Proc Nat Acad Sci U S A 2018;115:E4473–82.

46. Iglesias-Gato D, Wikstr€om P, Tyanova S, Lavallee C, Thysell E, Carlsson J,et al. The proteome primary prostate cancer. Eur Urol 2016;69:942–52.

47. Latonen L, Afyounian E, Jylh€a A, N€attinen J, Aapola U, Annala M, et al.Integrative proteomics in prostate cancer uncovers robustness againstgenomic and transcriptomic aberrations during disease progression.Nat Commun 2018;9:1176.

48. GrassoCS,WuYM,RobinsonDR, CaoX,Dhanasekaran SM,KhanAP, et al.The mutational landscape of lethal castration-resistant prostate cancer.Nature 2012;487.

49. Iglesias-GatoD, Thysell E, Tyanova S, Crnalic S, Santos A, Lima TS, et al. Theproteome of prostate cancer bone metastasis reveals heterogeneity withprognostic implications. Clin Cancer Res 2018;24:5433–44.

50. Varambally S, Yu J, Laxman B, Rhodes DR, Mehra R, Tomlins SA, et al.Integrative genomic and proteomic analysis of prostate cancer revealssignatures of metastatic progression. Cancer Cell 2005;8:393–406.

51. LaTulippe E, Satagopan J, Smith A, Scher H, Scardino P, Reuter V, et al.Comprehensive gene expression analysis of prostate cancer reveals distinct

transcriptional programs associated with metastatic disease. Cancer Res2002;62:4499–506.

52. Swinnen JV, Ulrix W, Heyns W, Verhoeven G. Coordinate regulation oflipogenic gene expression by androgens: evidence for a cascademechanisminvolving sterol regulatory element binding proteins. Proc Natl Acad SciU S A 1997;94:12975–80.

53. Yokoyama C, Wang X, Briggs MR, Admon A, Wu J, Hua X, et al. SREBP-1, a basic-helix-loop-helix-leucine zipper protein that controls tran-scription of the low density lipoprotein receptor gene. Cell 1993;75:187–97.

54. Streicher R, Kotzka J, M€uller-Wieland D, Siemeister G, Munck M, Avci H,et al. SREBP-1 mediates activation of the low density lipoprotein receptorpromoter by insulin and insulin-like growth factor-I. J Biol Chem 1996;271:7128–33.

55. SadowskiMC, Pouwer RH,Gunter JH, Lubik AA,QuinnRJ,NelsonCC. Thefatty acid synthase inhibitor triclosan: repurposing an anti-microbial agentfor targeting prostate cancer. Oncotarget 2014;5:9362–81.

56. Balaban S, Lee LS, Schreuder M, Hoy AJ. Obesity and cancer progression:is there a role of fatty acid metabolism? BioMed Res Int 2015;2015:274585.

57. Taylor RA, Lo J, Ascui N, Watt MJ. Linking obesogenic dysregulationto prostate cancer progression. Endocrine Connections 2015;4:R68–80.

58. Swinnen JV, Van Veldhoven PP, Esquenet M, Heyns W, Verhoeven G.Androgens markedly stimulate the accumulation of neutral lipids in thehuman prostatic adenocarcinoma cell line LNCaP. Endocrinology 1996;137:4468–74.

59. SubczynskiWK, Pasenkiewicz-GierulaM,Widomska J,Mainali L, RaguzM.High cholesterol/low cholesterol: effects in biological membranes: areview. Cell Biochem Biophys 2017;75:369–85.

60. McNair C, Urbanucci A, Comstock CE, Augello MA, Goodwin JF, Launch-bury R, et al. Cell-cycle coupled expansion of AR activity promotes cancerprogression. Oncogene 2017;36:1655–68.

61. StumpDD,ZhouSL, Berk PD.Comparisonof plasmamembrane FABPandmitochondrial isoform of aspartate aminotransferase from rat liver. Am JPhysiol Gastrointest Liver Physiol 1993;265:G894–902.

62. Blebea JS, HouseniM, TorigianDA, FanC,Mavi A, Zhuge Y, et al. Structuraland functional imaging of normal bone marrow and evaluation of its age-related changes. Sem Nuclear Med 2007;37:185–94.

63. Li J, Ren S, PiaoHL,Wang F, Yin P, XuC, et al. Integration of lipidomics andtranscriptomics unravels aberrant lipidmetabolismanddefines cholesteryloleate as potential biomarker of prostate cancer. Scientific Rep 2016;6:20984.

64. Boukouris AE, Zervopoulos SD,Michelakis ED.Metabolic enzymesmoon-lighting in the nucleus: metabolic regulation of gene transcription.Trends Biochem Sci 2016;41:712–30.

65. Bradbury MW, Stump D, Guarnieri F, Berk PD. Molecular modelingand functional confirmation of a predicted fatty acid binding siteof mitochondrial aspartate aminotransferase. J Mol Biol 2011;412:412–22.

66. Gordon JA, Midha A, Szeitz A, Ghaffari M, Adomat HH, Guo Y, et al. Oralsimvastatin administration delays castration-resistant progression andreduces intratumoral steroidogenesis of LNCaP prostate cancer xenografts.Prostate Cancer Prostatic Dis 2016;19:21–7.

67. Allott EH, Masko EM, Freedland AR, Macias E, Pelton K, Solomon KR, et al.Serum cholesterol levels and tumor growth in a PTEN-null transgenicmousemodel of prostate cancer. Prostate Cancer Prostatic Dis 2018;21:196–203.

68. Patel R, Fleming J, Mui E, Loveridge C, Repiscak P, Blomme A, et al.Sprouty2 loss-induced IL6 drives castration-resistant prostate cancerthrough scavenger receptor B1. EMBO Mol Med 2018;10pii:e8347.

69. Mostaghel EA, Solomon KR, Pelton K, Freeman MR, Montgomery RB.Impact of circulating cholesterol levels on growth and intratumoral andro-gen concentration of prostate tumors. PLoS One 2012;7:e30062.

www.aacrjournals.org Mol Cancer Res; 17(5) May 2019 1179

Androgen Regulation of Lipid Uptake in Prostate Cancer

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147

2019;17:1166-1179. Published OnlineFirst February 26, 2019.Mol Cancer Res   Kaylyn D. Tousignant, Anja Rockstroh, Atefeh Taherian Fard, et al.   MetastasisAssociated with Prostate Cancer Disease Progression and Bone Lipid Uptake Is an Androgen-Enhanced Lipid Supply Pathway

  Updated version

  10.1158/1541-7786.MCR-18-1147doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://mcr.aacrjournals.org/content/suppl/2019/02/26/1541-7786.MCR-18-1147.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://mcr.aacrjournals.org/content/17/5/1166.full#ref-list-1

This article cites 66 articles, 17 of which you can access for free at:

  Citing articles

  http://mcr.aacrjournals.org/content/17/5/1166.full#related-urls

This article has been cited by 1 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

  .pubs@aacr.org

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/17/5/1166To request permission to re-use all or part of this article, use this link

on March 9, 2021. © 2019 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst February 26, 2019; DOI: 10.1158/1541-7786.MCR-18-1147