Ligand, Receptor, and Cell Type–Dependent Regulation of ABCA1 and ABCG1 Mrna in Prostate Cancer Epithelial Cells

Published OnlineFirst June 16, 2009; DOI: 10.1158/1535-7163.MCT-09-0020 Published Online First on June 16, 2009 as 10.1158/1535-7163.MCT-09-0020

OF1

Ligand, receptor, and cell type–dependent regulation of ABCA1 and ABCG1 mRNA in prostate cancer epithelial cells

Steven E. Trasino,1 Young S. Kim,2 that ABCA1 gene expression is differentially regulated by and Thomas T.Y. Wang1 synthetic and natural LXR ligands, possibly involving kinase mediated signal transduction. [Mol Cancer Ther 2009;8(7): 1 Diet, Genomics, and Immunology Laboratory, Beltsville Human OF1–12] Nutrition Research Center, Agricultural Research Service, U.S. Department of Agriculture, Beltsville, Maryland and 2Nutritional Sciences Research Group, Division of Cancer Introduction Prevention, National Cancer Institute, NIH, Bethesda, Maryland Advanced and metastatic prostate cancer (PCa) remains the most prominent cancer type and the second leading cause of Abstract all cancer deaths among males in the United States (1). Recent evidence suggests that the liver X receptor (LXR) Treatment of recurrent or advanced PCa with hormone ab- is a potential anticancer target in prostate carcinoma. lation therapy typically results in disease regression with There is little characterization, however, of which of the eventual recurrence of PCa with a more aggressive and two LXR isoforms, LXRα or LXRβ, regulates the LXR- untreatable phenotype (2). In the absence of any effective responsive genes ATP-binding cassette subfamily members long-term therapy and with such an overwhelming social A1 (ABCA1)andG1(ABCG1) in transformed prostatic ep- burden, characterization of novel targets for effective che- ithelial cells. In this study, small interfering RNA (siRNA) moprevention or treatment of PCa has been a priority for α β cancer researchers. was used to determine whether LXR or LXR is involved α β in regulating ABCA1 and ABCG1 mRNA expression in Liver X receptors (LXR , LXR ) are nuclear receptor tran- LNCaP and PC-3 cells. Treatment of both cell lines with scription factors that have been recently identified as puta- tive chemotherapeutic targets in models of PCa (3). The the synthetic LXR ligand T0901317 and oxysterols: 25- α β hydroxycholesterol (25HC) and 24(S), 25-epoxycholesterol natural ligands for LXR and LXR are oxysterols, which (24,25EC), resulted in more than a 10-fold increase of are oxygenated cholesterol intermediates such as 25-hydro- ABCA1 and ABCG1 mRNA expression. Transfection of xycholesterol (25HC) and 24(S),25-epoxycholesterol LNCaP cells with siRNA against either LXRβ or LXRα failed (24,25EC) that are derived from both cholesterol catabolism to inhibit T0901317 and 25HC-mediated increase of and synthesis, respectively (4, 5). A highly specific synthetic ABCA1 mRNA. siRNA silencing of LXRβ did, however, in- LXR ligand (T0901317) has also been developed (6). Studies hibit ABCA1 mRNA expression in 24,25EC-treated LNCaP have shown that oxysterols and T0901317 can inhibit cells. In contrast, LXRβ siRNA inhibited T0901317, 25HC, growth of prostate carcinoma cells in culture and repress and 24,25EC induction of ABCA1 mRNA in PC-3 cells and progression of androgen-dependent tumor xenografts to a ABCG1 mRNA in both LNCaP and PC-3 cells. Additional more aggressive androgen-independent phenotype (7). experiments revealed that T0901317 and 25HC induction More recently, it has also been reported that T0901317 can ABCA1 behave as an antiandrogen agent in models of PCa (8). of mRNA expression was significantly inhibited by α β the p38 stress kinase antagonist SB202190 and PKA inhib- LXR and LXR isotypes have been primarily studied for itor H89. Our study is the first to show that LXRβ,butnot their critical role in limiting accumulation of free cholesterol LXRα, is the major regulatory isoform of ABCG1 mRNA in peripheral tissue and macrophages through regulation of expression in LNCaP and PC-3 cells. Our study also reveals reverse cholesterol transporters ATP-binding cassette, subfamily A, member 1 (ABCA1) and subfamily member G1 (ABCG1; refs. 9, 10). LXRα is primarily expressed in liver, adipose, and enterocytes where LXRβ is expressed ubiqui- Received 1/8/09; revised 3/30/09; accepted 4/13/09; published tously (11). Although both LXRα and LXRβ are expressed in OnlineFirst 6/16/09. the human adult prostate (12), it is unclear whether the ob- Grant support: U.S. appropriated funds to USDA project number 1235-51530-052-00 (T.T.Y. Wang and S.E. Trasino), and the National served anticancer effect of oxysterols and LXR ligands in Cancer Institute (Y.S. Kim). PCa models is related to regulation of cholesterol efflux or The costs ofpublication ofthis article were defrayedin part by the if LXR and its ligands take part in yet undetermined signal- payment ofpage charges. This article must thereforebe hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to ing pathways that are prostate specific. The latter could be indicate this fact. true as it has been shown that LXRs are uniquely involved Requests for reprints: Thomas T.Y. Wang, Diet, Genomics, & Immunology in the innate immune response of macrophages and CD4- Laboratory, Beltsville Human Nutrition Research Center, Agricultural positive lymphocytes (13). Research Service, U.S. Department ofAgriculture, Building 307C, Room 132, Beltsville, MD 20705. Phone: 301-504-8459; Fax: 301-504-9062. There is still considerable debate whether selective activa- E-mail: [email protected] tion of LXRα or LXRβ has a differential effect on cholesterol Copyright © 2009 American Association for Cancer Research. homeostasis or whether they exist as functionally redundant doi:10.1158/1535-7163.MCT-09-0020 paralogs (14, 15). Studies using LXRα/β null mice suggest

Mol Cancer Ther 2009;8(7). July 2009

OF2 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

that the regulation of genes in liver and peripheral tissue in- cells per well and grown for 24 h in culture medium, then volved in cholesterol homeostasis is primarily under the treated with various concentrations (0.1–10 μmol/L) of control of LXRα, and activation of LXRβ can partially rescue 25HC, 24,25EC, T0901317, or vehicle (DMSO 0.05% v/v). LXRα null animals from gross peripheral cholesterol accu- After termination of experiments, total RNA was isolated mulation (16). However, outside of its role in cholesterol ef- using the TRIzol reagent (Invitrogen), and reverse tran- flux, the broader biological functions of LXRβ are emerging, scribed to complementary DNA using StrataScript First yet remain unclear (17). Strand complementary DNA Synthesis kit (Stratagene; Unlike ABCG1 mRNA expression, which seems to be ex- ref. 21). Real-time PCR was carried out using a TaqMan clusively under the transcriptional control of LXRα, ABCA1 Universal PCR Master Mix on an ABI Prism 7000 Se- mRNA in a number of cell types is regulated through sig- quence Detection System (Applied Biosystems). The Taq- naling mechanisms independent of both LXR isotypes and Man probes and primers were purchased from Applied its role in cholesterol transport (18). Despite this, measuring Biosystems using inventoried TaqMan gene expression as- ABCA1 mRNA changes is often used as a surrogate marker says: LXRα, (assay ID: Hs00167445), LXRβ (assay ID: for in vitro and in vivo LXR activation (19–21). Hs00180254), ABCG1 (assay ID: Hs00167476), ABCA1 (as- In light of the heterogeneity between LXRα and LXRβ say ID: Hs02576345), CYP3A4 (assay ID: Hs01546612), and signaling in other cell types, and the observed anti-PCa CYP3A5 (assay ID: Hs00241417), MDR1 (assay ID: properties of LXR ligands, we sought to identify the rele- Hs00184500). Human glyceraldehyde-3-phosphate dehy- vant LXR isotypes and ligand behavior of T0901317, drogenase (assay ID: Hs99999905) was used as an endog- 25HC, and 24,25EC in the regulation of ABCA1 and ABCG1 enous control for all gene expression except for basal mRNA in two models of transformed prostatic epithelial mRNA quantitation in Fig. 1, where 18S rRNA (assay cells—LNCaP and PC-3. In this report, we show that LXRβ, ID: Hs99999901) was used as an additional control. The but not LXRα, is the primary regulator of ABCA1 and amplification parameters used were as follows: 50°C for ABCG1 mRNA expression in these in vitro cell line models. 2 min, 95°C for 10 min, followed by 46 cycles of amplifi- We also show in LNCaP cells that ABCA1 mRNA expres- cation at 95°C for 15 s and 60°C for 1 min. Quantitation of sion comes under LXRβ regulation only when cells are ex- mRNA fold changes were derived using the comparative −ΔΔCt posed to the endogenously synthesized oxysterol 24,25EC. CT (2 ) cycle (ΔCt) method (22). Lastly, our data suggests that synthetic LXR ligand Quantitation of Basal mRNA levels within LNCaP and T0901317 and the natural LXR ligand 25HC can regulate PC-3 Cells ABCA1 mRNA expression through LXRβ-independent me- We first assigned the largest ΔCt value of one gene as the chanisms involving the kinase-mediated pathway. control and normalized the lower ΔCt value of the second gene to determine a comparative ΔΔCt value. LXRβ ΔCt values were normalized to LXRα ΔCt values (control), Materials and Methods and ABCG1 ΔCt values were normalized to ABCA1 ΔCt Chemicals and Reagents values (control). Dharmacon ON-TARGETplus SMARTpool siRNA Quantitation of Comparative Basal mRNA Levels reagents targeting LXRα (NM_005693) or LXRβ Between LNCaP and PC-3 Cells (NM_007121) were purchased from Thermo Fisher Scientific. We assigned the higher LNCaP cells ΔCt values of each HiPerFect Transfection Reagent was purchased from Qiagen. gene as control and normalized the lower ΔCt value of The selective inhibitors for p38α/β (SB203580), c-Jun corresponding genes in PC-3 cells to determine a compara- NH2-terminal kinase (JNK; SP600125), extracellular signal- tive ΔΔCt value. ΔCt values for LXRα and LXRβ in PC-3 regulated kinases 1 and 2 (ERK1/2; PD98059), cyclic AMP– cells were normalized to LXRα and LXRβ ΔCt values in dependent-protein kinase A (PKA; H-89), the LXR ligand LNCaP cells, and ΔCt values for ABCA1 and ABCG1 in 24,25EC, and the synthetic PXR ligand (SR12813), were PC-3 cells were normalized to ΔCt values for ABCA1, all purchased from BIOMOL International. The synthetic ABCG1 in LNCaP cells. LXR ligand T0901317 was purchased from Cayman Experiments Using Small Interfering RNA Oligo‐ Chemical Company. 25HC and dimethylsulfoxide (DMSO) nucleotides were purchased from Sigma Co. LNCaP and PC-3 cells were seeded in six-well plates at Cell Culture a density of 0.2 × 106 cells per well and grown for 24 h in The human PCa cell lines LNCaP and PC-3 were pur- culture medium; then cells were switched to serum free chased from the American Type Culture Collection and media in RPMI 1640 (without phenol red) containing maintained in RPMI 1640 (Invitrogen), supplemented with 1% glutamine and 1% penicillin/streptomycin (serum-free 10% fetal bovine serum (FBS; Invitrogen), 1% glutamine, culture medium). Cells were transfected with 5 nmol/L of 100 U/mL penicillin, and 100 μg/mL streptomycin (Invitro- Dharmacon ON-TARGETplus SMARTpool small interfer- gen; culture medium). Cells were incubated in the presence ing RNA (siRNA) oligonucleotides targeting LXRα or of 5% carbon dioxide and air at 37°C. LXRβ using the HiPerFect transfection reagent (Qiagen) RNA Isolation and Real-Time Reverse Transcriptase PCR according to the manufacturer's protocol. Preliminary ex- For gene expression experiments, LNCaP and PC-3 cells periments determined that 5 nmol/L siRNA oligonucleo- were seeded in six-well plates at a density of 0.5 × 106 tides were optimal for 70% to 80% gene silencing without

Mol Cancer Ther 2009;8(7). July 2009

Molecular Cancer Therapeutics OF3

Figure 1. LXRα,LXRβ, ABCA1, and ABCG1 mRNA expression in LNCaP and PC-3 cells. LNCaP and PC-3 cells were seeded at a density of0.2 × 10 6 cells per well and grown for 72 h in RPMI 1640 10% FBS and then subjected to RNA isolation and reverse transcription-PCR (RT-PCR) as described in Materials and Methods. A, relative mRNA levels of LXRα,LXRβ, ABCA1, and ABCG1 in LNCaP cells. Columns, means of comparative fold differences; bars, SE. LXRβ mRNA expression levels were normalized to LXRα mRNA expression and ABCA1 mRNA expression levels normalized to ABCG1 mRNA expression. Data are representative ofthree independent experiments. B, relative mRNA levels of LXRα,LXRβ,ABCA1,andABCG1 in PC-3 cells. Columns, means of comparative fold differences; bars, SE. LXRβ mRNA expression levels were normalized to LXRα mRNA expression and ABCG1 mRNA expression levels normalized to ABCA1 mRNA. Data are representative ofthree independent experiments. C, comparative mRNA levels of LXRα,LXRβ between LNCaP and PC-3 cells. Columns, means of comparative fold differences; bars, SE. PC-3 cell LXRα,LXRβ mRNA expression levels were normalized to LXRα,LXRβ mRNA expression in LNCaP cells. Data are representative ofthree independent experiments. D, comparative mRNA levels of ABCA1 and ABCG1 between LNCaP and PC-3 cells. Columns, means of comparative fold differences; bars, SE. PC-3 cell ABCA1 and ABCG1 mRNA expression levels were normalized to ABCA1 and ABCG1 mRNA expression in LNCaP cells. Data are representative ofthree independent experiments.

any off-target effects. After 48 h of siRNA transfection, the PKA inhibitor H-89, or vehicle (DMSO 0.05% v/v). cells were switched back to culture medium and treated After 60 min, cell culture media was replaced and cells with various LXR ligands as described in Materials and were stimulated with 5 μmol/L T0901317, 25HC, or Methods for an additional 48 h. 24,25EC for an additional 24 h. Kinase Inhibitor Experiments Statistics For kinase inhibition experiments, LNCaP and PC-3 Statistical analysis of data were carried out with the cells were plated in six-well plates as described above GraphPad PRISM program (GraphPad Software, Inc.). Mul- and pretreated for 60 min with 10 μmol/L of one of tiple group data were analyzed using ANOVA or two-way the following: the p38 mitogen-activated protein kinase ANOVA followed by Bonferroni post hoc tests. The unpaired inhibitor SB203580, the JNK MAP kinase inhibitor Student's t test was used to compare experiments between SP600125, the ERK1/2 MAP kinase inhibitor PD98059, two groups. Gene expression results are expressed as means

Mol Cancer Ther 2009;8(7). July 2009

OF4 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

± SE of comparative fold differences. Data are representative normalizing against both 18S rRNA and human glyceralde- of three independent experiments. Values were considered hyde-3-phosphate dehydrogenase. significant at a P value of <0.05. Concentration and Temporal Effects of LXR Ligands on ABCG1 and ABCA1 mRNA Expression To measure the concentration effects of synthetic and nat- Results ural LXR ligands on mRNA expression of ABCA1 and Basal Levels of LXRα,LXRβ, ABCA1,andABCG1 ABCG1, LNCaP and PC-3 cells were treated with increasing mRNA Expression in Prostate Cancer Cells concentrations (0.1–10 μmol/L) of T0901317, 24,25EC, and As shown in Fig. 1A and B, our data revealed that basal 25HC for 24 h. In both LNCaP (Fig. 2A and B) and PC-3 mRNA levels of LXRβ were significantly higher than LXRα cells (Fig. 2C and D), treatments with T0901317, 25HC, in both LNCaP and PC-3 cells, as were basal levels of and 24,25EC resulted in dose responses in ABCA1 and ABCG1 compared with ABCA1. Across the cell types, ABCG1 mRNA expression. LNCaP cells treated with the PC-3 cells expressed significantly higher basal levels of all synthetic LXR ligand T0901317 resulted in a ∼4-to 20-fold genes with an exceptionally higher expression of ABCA1 induction of ABCA1 mRNA compared with ∼4-to 10-fold mRNA (Fig. 1C and D). Similar results were obtained when inductions in 24,25EC and in 25HC-treated cells (Fig. 2A).

Figure 2. Concentration-dependent effects of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA expression. LNCaP and PC-3 cells were seeded at a density of0.5 × 10 6 cells per well and grown for 24 h in RPMI 1640 10% FBS and then treated for 24 h with or without 0.1 to 10 μmol/L T0901317, 24,25EC, or 25HC. After treatments, cells were subjected to RNA isolation and RT-PCR as described in Materials and Methods. A and B, dose effects of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA induction in LNCaP cells. ABCA1, ABCG1 mRNA induction is expressed as means ± SE of fold change relative to control. Data are representative of three independent experiments. *, significantly different than control (P < 0.05). C and D, dose effects of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA induction in PC-3 cells. ABCA1, ABCG1 mRNA induction is expressed as means ± SE of fold change relative to control. Data are representative of three independent experiments. *, significantly different than control (P < 0.05).

Mol Cancer Ther 2009;8(7). July 2009

Molecular Cancer Therapeutics OF5

ABCG1 mRNA fold induction in T0901317-treated LNCaP ever, the dose-dependent effect of T0901317 and all oxyster- cells was ∼9-to 20-foldcompared with ∼2-to 8-foldin ols was linear up to 10 μmol/L. This trend was also oxysterol-treated cells (Fig. 2B). PC-3 cells treated with in- observed for LXR ligand induction of ABCG1 mRNA in creasing concentrations of T0901317 resulted in a ∼12-to LNCaP and PC-3 cells. Fold changes for ABCG1 mRNA 25-fold induction of ABCA1 mRNA compared with 2- with increasing doses of all LXR ligands were similar in to 7-fold inductions in 24,25EC and 25HC-treated cells both LNCaP and PC-3 cells (Fig. 2B and D). (Fig. 2C). In a similar manner to ABCA1, ABCG1 mRNA Temporal changes in ABCA1 and ABCG1 mRNA expres- fold induction in T0901317-treated PC-3 cells was ∼8-to sion were also determined by treating LNCaP and PC-3 15-fold compared with ∼2-to 6-foldin 24,25EC and cells with 5 μmol/L of T0901317, 25HC, and 24,25 EC for 25HC-treated cells (Fig. 2D). Between PC-3 and LNCaP 2 to 24 hours. Figure 3A and B show a significantly earlier cells, we observed a significantly stronger ABCA1 mRNA change in mRNA response of ABCA1 at 2 hours compared induction in PC-3 cells specifically within the T0901317 con- with ABCG1 in LNCaP cells (∼3-fold versus 0-fold). We ob- centration range of 0.1 and 1.0 μmol/L. (∼10-to 20-foldin served the opposite effect in PC-3 cells with the change in PC-3 versus ∼3- to 5-fold in LNCaP; Fig. 2A and C). In PC-3 ABCG1 mRNA increasing dramatically between 2 and 4 cells, the concentration plateau for T0901317 induction of hours by almost 10-fold, where ABCA1 mRNA changes ABCA1 mRNA began at 5 μmol/L. In LNCaP cells, how- were ∼5-fold less in the same time frame (Fig. 3C and D).

Figure 3. Temporal effects of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA expression. LNCaP and PC-3 cells were seeded at a density of0.5 × 10 6 cells per well and grown for 24 h in RPMI 1640 10% FBS and then treated for 2 to 24 h with or without 5 μmol/L ofT0901317, 24,25EC, or 25HC. After treatments, cells were subjected to RNA isolation and RT-PCR as described in Materials and Methods. A and B, temporal effect of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA induction in LNCaP cells. ABCA1, ABCG1 mRNA induction is expressed as means ± SE of fold change relative to control at each time point. Data are representative of three independent experiments. *, significantly different than control (P < 0.05). C and D, temporal effect of T0901317, 24,25EC, and 25HC on ABCA1 and ABCG1 mRNA induction in PC-3 cells. ABCA1, ABCG1 mRNA induction is expressed as means ± SE offoldchange relative to control at each time point. Data are representative ofthree independent experiments. *, significantly different than control (P < 0.05).

Mol Cancer Ther 2009;8(7). July 2009

OF6 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

Figure 4. Effects of LXRα siRNA on ligand-mediated expression of ABCA1 and ABCG1 in LNCaP and PC-3 cells. LNCaP and PC-3 cells were seeded in six- well plates at a density of0.2 × 10 6 cells per well and grown for 24 h in RPMI 1640 10% FBS. After 24 h, cells were switched to serum-free culture medium and transfected with or without LXRα siRNA oligonucleotides for 48 h and then treated with or without 5 μmol/L ofT0901317, 24,25EC, or 25HC foran additional 48 h. After treatments, cells were subjected to RNA isolation and RT-PCR as described in Materials and Methods. A, effects of LXRα siRNA transfection on basal and ligand inducible LXRα and LXRβ mRNA levels in LNCaP cells. Relative mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). B, effects of LXRα siRNA transfection on basal and ligand inducible LXRα and LXRβ mRNA levels in PC-3 cells. Relative mRNA levels are expressed as means ± SE of fold change relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). C, effects of LXRα siRNA transfection on basal and ligand inducible ABCA1 and ABCG1 mRNA levels in LNCaP cells. Relative mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative ofthree independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). D, effects of LXRα siRNA transfection on basal and ligand inducible ABCA1 and ABCG1 mRNA levels in PC-3 cells. Relative mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05).

Ligand and LXR-Specific Regulation of ABCA1 and quired for oxysterol-and T0901317-mediated induction of ABCG1 mRNA Expression Varies Between Cell Types ABCA1 and ABCG1 mRNA using siRNA technology. Our Given that both LXR isoforms are expressed in PCa cells, siRNA transfection conditions resulted in ∼70% to 80% si- we sought to determine whether LXRα or LXRβ are re- lencing of LXRα and LXRβ mRNA expression levels in both

Mol Cancer Ther 2009;8(7). July 2009

Molecular Cancer Therapeutics OF7

LNCaP and PC-3 cells. In both cell lines, all LXR ligands sig- To our surprise, LNCaP cells transfected with either siR- nificantly increased basal levels of LXRβ (Figs. 4A and B, 5A NA against LXRβ or LXRα could not inhibit T0901317 or and B), but had no significant effect on LXRα mRNA levels 25HC induction of ABCA1 mRNA (Figs. 4C and 5C). Use (Figs. 4A and B, 5A and B). of siRNA against LXRβ was able to inhibit ABCA1 mRNA

Figure 5. Effects of LXRβ siRNA interference on ligand-mediated expression of ABCA1 and ABCG1 in LNCaP and PC-3 cells. LNCaP and PC-3 cells were seeded in six-well plates at a density of0.2 × 10 6 cells per well and grown for 24 h in RPMI 1640 10% FBS. After 24 h, cells were switched to serum-free culture medium and transfected with or without LXRβ siRNA oligonucleotides for 48 h and then treated with or without 5 μmol/L ofT0901317, 24,25EC, or 25HC for an additional 48 h. After treatments, cells were subjected to RNA isolation and RT-PCR as described in Materials and Methods. A, effects of LXRβ siRNA transfection on basal and ligand inducible LXRα and LXRβ mRNA levels in LNCaP cells. Relative mRNA levels are expressed as means ± SE offold change relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). B, effects of LXRβ siRNA transfection on basal and ligand inducible LXRα and LXRβ mRNA levels in PC-3 cells. Relative mRNA levels are expressed as means ± SE of fold change relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). C, effects of LXRβ siRNA transfection on basal and ligand inducible ABCA1 and ABCG1 mRNA levels in LNCaP cells. Relative mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative ofthree independent experim ents. Error bars with a different letter indicate significant differences between groups (P < 0.05). D, effects of LXRβ siRNA transfection on basal and ligand inducible ABCA1 and ABCG1 mRNA levels in PC-3 cells. Relative mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05).

Mol Cancer Ther 2009;8(7). July 2009

OF8 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

in LNCaP cells treated with 24,25EC only (Fig. 5C). We did hibitor H-89 (Fig. 6C). In both cell lines, inhibition of observe a significant inhibition of ABCG1 mRNA induction ERK1/2 had no effect on mRNA expression of either by all LXR ligands in LNCaP cells transfected with LXRβ ABCA1 or ABCG1 (Fig. 6A–D). siRNA (Fig. 5C); however, LXRα siRNA had no effect on ABCG1 mRNA induction by any LXR ligand (Fig. 4C). In PC-3 cells, T0901317 and oxysterol regulation of ABCG1 Discussion and ABCA1 mRNA expression was more straightforward. The recent emergence of data suggesting that LXR ligands The use of siRNA against LXRβ inhibited the mRNA ex- activated an anticancer effect in models of prostate car- pression of both ABCG1 and ABCA1 in PC-3 cells exposed cinoma prompted us to examine the regulation of LXR- to T0901317, 25HC, and 24,25EC (Fig. 5D). siRNA against mediated pathways in LNCaP and PC-3 human PCa cells. LXRα alsohadnoeffectonthesetargetsinPC-3cells Our study specifically sought to characterize in these trans- (Fig. 4D). formed epithelial cells, which LXR isoform is involved in There is evidence that T0901317 can also activate the the transcriptional regulation of the LXR responsive genes pregnane X receptor (PXR) in vivo and in vitro (23). To test ABCA1 and ABCG1. We hypothesized that the LXR regula- whether T0901317-induction of ABCA1 mRNA may have tory mechanisms in these models of prostatic carcinoma occurred through PXR, we treated LNCaP and PC-3 cells would be distinct from other cell types and that a cell- for 24 h with multiple concentrations of T0903171 (0.5–5.0 specific response would have to be considered in any future μmol/L) and measured mRNA expression of three PXR studies of the anticancer effect of LXR ligands. responsive genes: cytochrome P450 3A4 (CYP3A4), cyto- We initially determined that relative LXRβ mRNA ex- chrome P450 3A5 (CYP3A5), and ATP-binding cassette, pression is significantly higher than LXRα in both LNCaP subfamily B (MDR), member 1(MDR1). At all concentra- and PC-3 cells. This did not come as a surprise as most tions, treatments with T0901317 had no effect on mRNA expression profiling has determined that LXRα is primarily levels of these genes in either cell line. Moreover, we expressed in liver, adipose, and enterocytes, where LXRβ observed that treatment of LNCaP and PC-3 cells with is expressed ubiquitously (4). Our LXRα and LXRβ basal the specific PXR ligand SR12813 at concentrations be- mRNA profiling does conflict with a previous study that tween 2 to 20 μmol/L for 24 hours failed to affect mRNA reported that LXRα is the dominant isoform in LNCaP expression of ABCA1 or ABCG1. cells and the ratio of LXRα to LXRβ in PC-3 cells is ap- LXR-Independent Regulation of ABCA1 mRNA in proximately one (24). We used a similar normalization LNCaP and PC-3 Cells method as this previous report (24), but found LXRβ to In observing that neither LXRα nor LXRβ seemed to be have higher relative mRNA levels in both LNCaP and involved in T0901317 or 25HC regulation of ABCA1 mRNA PC-3 cells. expression, we interrogated a number of kinase pathways Our basal mRNA measurements revealed disproportion- to determine whether T0901317 or 25HC can regulate ally lower ABCA1 mRNA expression compared with ABCA1 mRNA expression through a non–LXR-mediated ABCG1 in both LNCaP and PC-3 cells. There is evidence pathway involving kinase signal transduction. LNCaP in some cell types that without functional ABCA1, ABCG1 and PC-3 cells were pre treated for 60 minutes with alone is incapable of regulating the cholesterol efflux ma- 10 μmol/L of each of the following specific kinase inhibi- chinery due to the cooperative requirement of ABCA1 and tors: ERK1/2 (PD98059), PKA (H-89), p38α/β (SB203580), apolipoprotein A-1 in formation of pre-HDL particles (25, or JNK (SP600125); they were then treated with and with- 26). Further support for a critical role of ABCA1 in HDL for- out 5 μmol/L of T0901317 and 25HC for an additional mation and cholesterol efflux come from observations of se- 24 hours. Pretreatment of LNCaP cells with the p38 antag- vere HDL deficiency and pathologic cholesterol retention in onist SB203580 caused a ∼45% inhibition of both T0901317 persons with functional mutations of the ABCA1 gene (27). and 25HC induction of ABCA1 mRNA expression (P < Carriers of ABCA1 mutations, such as those seen in Tangier 0.01; Fig. 6A). This was specific to ABCA1 as we did not disease, are largely incapable of normal reverse cholesterol observe any effects on ABCG1 mRNA levels (Fig. 6B). Inhi- transport despite carrying normal copies of the ABCG1 gene bition of p38 in PC-3 cells only affected T0901317-mediated (28). If ABCA1 synergizes with ABCG1 for cholesterol re- induction of ABCA1 mRNA (Fig. 6C); however, unlike moval in LNCaP and PC-3 cells as it does in other cell types LNCaP cells, SB203580 also inhibited T0901317-mediated (25), then our data may have revealed a propensity for im- increase in ABCG1 mRNA levels by ∼36% in PC-3 cells paired cholesterol efflux in these models. This would sup- (Fig. 6D). In contrast, inhibition of JNK kinase in both port a large body of data demonstrating abnormal LNCaP and PC-3 cells resulted in a ∼6-fold induction of prostatic cholesterol retention in men with PCa (29), which basal mRNA levels of ABCA1 (Fig. 6A and C). Moreover, is believed to provide transformed cells a reservoir of sterol JNK inhibition produced an additive effect on T0901317- precursors for de novo synthesis of androgens in the face of and 25HC-induced increase of ABCA1 mRNA in LNCaP castration and other antiandrogen therapies (30–32). It has cells (Fig. 6A). In PC-3 cells, this effect was only observed been shown that androgen signaling exerts a concerted pro- for T0901317 induction of ABCA1 mRNA (Fig. 6C). We also gram to increase prostatic cholesterol synthesis through observed complete inhibition of T0901317 induction of up-regulation of sterol regulatory element-binding protein ABCA1 mRNA in PC-3 cells pretreated with the PKA in- and hydroxymethyl-glutaryl-CoA synthase pathways (33).

Mol Cancer Ther 2009;8(7). July 2009

Molecular Cancer Therapeutics OF9

Figure 6. Effects of p38 and JNK MAP kinase inhibitors on T0901317 and 25HC-mediated induction of ABCA1 mRNA. LNCaP and PC-3 cells were seeded at a density of0.5 × 10 6 cells per well and grown for 24 h in RPMI 1640 10% FBS and then pretreated with or without 10 μmol/L kinase inhibitors for p38 (SB203580), JNK (SP600125), ERK1/2 (PD98059), or PKA (H89) for 60 min followed by treatment with or without 5 μmol/L ofT0901317, 24,25EC, or 25HC for 24 h. After treatments, cells were subjected to RNA isolation and RT-PCR as described in Materials and Methods. A and B, effects ofvarious kinase inhibitors on ABCA1 and ABCG1 mRNA expression in LNCaP cells. Relative ABCA1, ABCG1 mRNA levels are expressed as means ± SE of fold change relative to control. Data are representative of three independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05). C and D, effects of various kinase inhibitors on ABCA1 and ABCG1 mRNA expression in PC-3 cells. Relative ABCA1, ABCG1 mRNA levels are expressed as means ± SE offoldchange relative to control. Data are representative ofthree independent experiments. Error bars with a different letter indicate significant differences between groups (P < 0.05).

A number of epidemiologic studies have already observed Expression Omnibus database3 (36). We found in two sep- that inhibiting endogenous cholesterol synthesis with arate array profiles of normal prostate epithelial cells that agents such as mevastatin decreases individual risk for ABCA1 mRNA expression was approximately twice to developing PCa, and in particular advanced PCa (34, 35). thrice less than ABCG1 [GEO accessions: GDS1746 (37), To further examine whether the ABCA1, ABCG1 mRNA GDS1973 (38)]. We also found expression profile data in ratio profile in these transformed cell lines differs to those found in normal prostate epithelial cells, we did a query of the National Center for Biotechnology Information's Gene 3 http://www.ncbi.nlm.nih.gov/projects/geo

Mol Cancer Ther 2009;8(7). July 2009

OF10 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

GEO accession: GDS1746, which showed that, similar to sion of ABCA1, and that T0901317 can act as an antiandro- our findings, a low ABCA1, ABCG1 mRNA ratio exists gen (24, 43). It is plausible, therefore, that T0901317 in LNCaP cells. Collectively, these data strongly suggested indirectly increased ABCA1 mRNA expression in this model that the ABCA1, ABCG1 mRNA expression ratio is de- through direct antagonism of androgen receptor signaling creased in transformed prostatic epithelial cells and sup- by acting as an antiandrogen, rather than ligand activation ported data demonstrating abnormal prostatic cholesterol of LXRβ. Conversely, the regulation of ABCG1 mRNA by retention in PCa (29). Due to the integral role that LXR synthetic and natural LXR ligands in both LNCaP and and ABCA1 have in cholesterol efflux, it is warranted to PC-3 cells is consistent with reports that ABCG1 mRNA determine whether suppression of LXR/ABCA1 function transcriptional regulation comes exclusively under LXR is another mechanism through which transformed prostat- (44). Given that our siRNA transfections against LXRα ic cells can increase their cholesterol pool. and LXRβ achieved 70% to 80% silencing of these genes, Data from our siRNA experiments supported our original our data strongly support a principal role of LXRβ but hypothesis that LXR biology in these transformed cells not LXRα in the regulation of induction of ABCA1 and would be atypical. We found that the synthetic and natural ABCG1 mRNA by T0901317 and oxysterols in these models. LXR ligands T0901317, 25HC, and 24,25EC showed distinct Ourpharmacologicinhibitorstudiessuggestthat LXR-dependent and independent regulation of ABCA1 T0901317 and 25HC regulation of ABCA1 mRNA expres- mRNA in both LNCaP and PC3 cells. The discovery that sion may involve the kinases-mediated pathway of p38, siRNA against LXRα or LXRβ failed to inhibit T0901317 JNK, and PKA. Both p38 and JNK mediate the cellular stress and 25HC-mediated ABCA1 mRNA induction in LNCaP response and are naturally activated by a variety of stimuli cells was compelling and contrasts with reports from nu- including proinflammatory cytokines, osmotic shock, and merous LXR functional studies demonstrating that UV irradiation (45). There is a report of tumor necrosis fac- T0901317 and oxysterols are potent LXR ligands and tor α induction of ABCA1 mRNA expression in mouse peri- increase ABCA1 mRNA expression through direct ligand toneal macrophages through the stress kinase p38 (46). activation of LXR (39). Although there are reports in macro- Nevertheless, the p38 antagonist used in our experiments, phages and skin fibroblasts that T0901317 can also activate SB203580, has been shown to also activate JNK and the the xenobiotic transcription factor PXR (23), we found that ERK1/2 map kinase pathway (47, 48). The pleiotropic effect LNCaP and PC3 cells treated for 24 hours with multiple of SB203580 raises the possibility that its inhibitory effect on concentrations of T0903171 had no effect on expression of T0901317 and 25HC induction of ABCA1 mRNA could oc- the PXR responsive genes CYP3A4, CYP3A5,andMDR1. cur through uncharacterized targets other than p38. To add We also did not observe any ABCA1 and ABCG1 mRNA complexity to the regulation of ABCA1 mRNA, we also ob- changes in cells treated with multiple concentrations of served that inhibition of JNK with SP600125 raised basal the PXR agonist SR12813. These experiments suggest to us and LXR ligand–inducible levels of ABCA1 mRNA ∼6- that T0901317 induction of ABCA1 and ABCG1 mRNA is and 1.5-fold, respectively, in both LNCaP and PC-3 cells not mediated through PXR in LNCaP or PC3 cells. (Fig. 6A and B). Both p38 and JNK are typically redundant Our experiments revealed that, in comparison with other members of the stress-activated protein kinase pathway; ligands tested, 24,25EC uniquely required LXRβ in regulat- however, it has been shown that they can oppose each other ing both ABCA1 and ABCG1 mRNA expression in LNCaP depending on the downstream target (49). The observed in- cells. 24,25EC is synthesized exclusively in a shunt reaction duction of ABCA1 mRNA by the JNK inhibitor could reflect from the mevalonate de novo cholesterol synthesis pathway a similar antagonism of JNK toward p38 signaling and and is believed to permit cellular sensing of excess choles- could represent be a molecular “on” and “off” of ABCA1 terol biosynthesis (40). The unique ability of 24,25EC in reg- gene regulation. We observed a similar effect on ABCA1 ulating ABCA1 through LXRβ in LNCaP cells could reflect mRNA response in PC-3 cells pretreated with p38 and similar findings in a study that reported that 24,25EC discri- JNK inhibitors supporting a common regulation of ABCA1 minately up-regulated mRNA expression of certain LXR- gene expression by a signal transduction pathway in both responsive genes in THP-1 monocytes (41). LNCaP and PC-3 cells. However, in PC-3 cells, the PKA in- We observed that ABCG1 mRNA regulation by synthetic hibitor H-89 completely reversed T0901317 induction of and natural LXR ligands in LNCaP cells strongly required ABCA1 mRNA. As previously mentioned, in the absence the expression of LXRβ. This was also observed for ABCG1 of a functional androgen receptor and necessity of androgen and ABCA1 mRNA expression in PC3 cells transfected with signaling in PC-3 cells, regulation of ABCA1 mRNA could siRNA against LXRβ. The lack of a consistent involvement reflect mechanisms shown in other cell types such as macro- of LXRβ in ligand induction of ABCA1 mRNA in PC3 and phages where PKA positively regulates ABCA1 activity LNCaP cells likely reflected the numerous molecular dis- (50). In light of the redundancy and cross talk within and tinctions between these two cell lines (42). The most prom- across kinase families, we believe further studies are needed inent difference is the lack of functional androgen receptors to reveal the specific interactions between the various iso- and the need for androgen-driven growth in PC3 cells, forms of each kinase and LXR ligands T0901317 and which is suggestive of a relevant cross-talk between andro- 25HC in the regulation of ABCA1 mRNA in these models. gen and LXR signaling in LNCaP cells. It has been reported Data from our LXRα or LXRβ experiments uniquely show in LNCaP cells that androgens strongly suppress the expres- that the highly specific LXR ligands T0901317 and 25HC do

Mol Cancer Ther 2009;8(7). July 2009

Molecular Cancer Therapeutics OF11

not seem to require LXRα or LXRβ in their regulation of induces expression ofABCA1 but not ABCG1 via an LXR-independent pathway. J Lipid Res 2002;6:952–9. ABCA1 mRNA expression in LNCaP cells. This finding 19. Birrell MA, Catley MC, Hardaker , et al. Novel role for the liver X nu- should serve as a caution for studies using T0901317 or clear receptor in the suppression oflung inflammatory responses. J Biol 25HC to characterize novel LXR targets in PCa or other dis- Chem 2007;44:31882–90. ease models where ABCA1 mRNA is sometimes the only 20. Verschuren L, de Vries-van der Weij J, Zadelaar S, et al. LXR agonist suppresses atherosclerotic lesion growth and promotes lesion regression proxy for T0901317 or oxysterol LXR activation. In light of in ApoE*3Leiden mice: time course and potential mechanisms. J Lipid Res the central role that LXR and ABCA1 have in coordinating 2009;50:301–11. cholesterol efflux and the apparent dysregulation of choles- 21. Delvecchio CJ, Bilan P, Nair P, et al. LXR-induced reverse cholesterol terol removal in PCa, research aimed at understanding die- transport in human airway smooth muscle is mediated exclusively by ABCA1. Am J Physiol Lung Cell Mol Physiol 2008;295:L949–57. tary or pharmacologic activation of the LXR-ABCA1 axis as 22. Livak KJ, Schmittgen TD. Analysis ofrelative gene expression data a novel target for adjuvant therapy of PCa is warranted. using real-time quantitative PCR and the 2ΔΔCT method. Methods 2001;25:402–8. 23. Mitro N, Vargas L, Romeo R, et al. T0901317 is a potent PXR ligand: Disclosure of Potential Conflicts of Interest implications for the biology ascribed to LXR. FEBS Lett 2007;9:1721–6. Fukuchi J, Hiipakka RA, Kokontis JM, et al. Androgenic suppression No potential conflicts of interest were disclosed. 24. ofATP-binding cassette transporter A1 expression in LNCaP human prostate cancer cells. Cancer Res 2004;21:7682–5. References 25. Ikonen E. Cellular cholesterol trafficking and compartmentalization. Nat Rev Mol Cell Biol 2008;2:125–38. 1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;2:71–96. 26. Gelissen IC, Harris M, Rye KA, et al. ABCA1 and ABCG1 synergize to mediate cholesterol export to apoA-I. Arterioscler Thromb Vasc Biol 2006; 2. Arlen PM, Gulley JL, Todd N, et al. Antiandrogen, vaccine and combi- 3:534–40. nation therapy in patients with nonmetastatic hormone refractory prostate cancer. J Urol 2005;2:539–46. 27. Iatan I, Alrasadi K, Ruel I, et al. Effect of ABCA1 mutations on risk for myocardial infarction. Curr Atheroscler Rep 2008;5:413–26. 3. Chuu CP, Kokontis JM, Hiipakka RA, et al. Modulation ofliver X receptor signaling as novel therapy for prostate cancer. J Biomed Sci 28. Mott S, Yu L, Marcil M, et al. Decreased cellular cholesterol efflux is a α 2007;5:543–53. common cause offamilialhypo lipoproteinemia: role ofthe ABCA1 gene mutations. Atherosclerosis 2000;2:457–68. 4. Ulven SM, Dalen KT, Gustafsson JA, et al. LXR is crucial in lipid metabolism. Prostaglandins Leukot Essent Fatty Acids 2005;1:59–63. 29. Hager MH, Solomon KR, Freeman MR. The role ofcholesterol in prostate cancer. Curr Opin Clin Nutr Metab Care 2006;4:379–85. 5. Wong J, Quinn CM, Gelissen IC, et al. Endogenous 24(S),25-epoxycholesterol fine-tunes acute control of cellular cholesterol homeostasis. 30. Mostaghel EA, Nelson PS. Intracrine androgen metabolism in prostate J Biol Chem 2008;2:700–7. cancer progression: mechanisms ofcastration resistance and therapeutic implications. Best Pract Res Clin Endocrinol Metab 2008;2:243–58. 6. Schultz JR, Tu H, Luk A, et al. Role ofLXRs in control oflipogenesis. Genes Dev 2000;22:2831–8. 31. Locke JA, Guns ES, Lubik AA, et al. Androgen levels increase by intratumoral de novo steroidogenesis during progression ofcastration- 7. Chuu CP, Hiipakka RA, Kokontis JM, et al. Inhibition oftumor growth resistant prostate cancer. Cancer Res 2008;15:6407–15. and progression ofLNCaP prostate cancer cells in athymic mice by androgen and liver X receptor agonist. Cancer Res 2006;13:6482–6. 32. Holzbeierlein J, Lal P, LaTulippe E, et al. Gene expression analysis of human prostate carcinoma during hormonal therapy identifies androgen- 8. Chuu CP, Chen RY, Hiipakka RA, et al. The liver X receptor agonist responsive genes and mechanisms oftherapy resistance. Am J Pathol T0901317 acts as androgen receptor antagonist in human prostate cancer 2004;1:217–27. cells. Biochem Biophys Res Commun 2007;2:341–6. 33. Swinnen JV, Ulrix W, Heyns W, et al. Coordinate regulation oflipo- 9. Tontonoz P, MangelsdorfDJ. Liver X receptor signaling pathways in genic gene expression by androgens: evidence for a cascade mechanism cardiovascular disease. Mol Endocrinol 2003;6:985–93. involving sterol regulatory element binding proteins. Proc Natl Acad Sci 10. Levin N, Bischoff ED, Daige CL, et al. Macrophage liver X receptor is U S A 1997;24:12975–80. required for antiatherogenic activity of LXR agonists. Arterioscler Thromb 34. Magura L, Blanchard R, Hope B, et al. Hypercholesterolemia and pros- Vasc Biol 2005;1:135–42. tate cancer: a hospital-based case-control study. Cancer Causes Control 11. Lehmann JM, Kliewer SA, Moore LB, et al. Activation ofthe nuclear 2008;19:1259–66. receptor LXR by oxysterols defines a new hormone response pathway. 35. Hamilton RJ, Freedland SJ. Review ofrecent evidence in support ofa J Biol Chem 1997;6:3137–40. role for statins in the prevention of prostate cancer. Curr Opin Urol 2008; 12. Shmueli O, Horn-Saban S, Chalifa-Caspi V, et al. GeneNote: whole 3:333–9. genome expression profiles in normal human tissues. C R Biol 2003; 36. Barrett T, Edgar R. Mining microarray data at NCBI's Gene Expression 10-1:1067–72. Omnibus (GEO). Methods Mol Biol 2006;338:175–90. 13. Fontaine C, Rigamonti E, Nohara A, et al. Liver X receptor activation 37. Nanni S, Priolo C, Grasselli A, et al. Epithelial-restricted gene profile potentiates the lipopolysaccharide response in human macrophages. Circ ofprimary cultures fromhuman prostate tumors: a molecular approach Res 2007;1:40–9. to predict clinical behavior ofprostate cancer. Mol Cancer Res 2006;2: 14. Lund EG, Peterson LB, Adams AD, et al. Different roles of liver X re- 79–92. ceptor α and β in lipid metabolism: effects of an α-selective and a dual 38. Oudes AJ, Campbell DS, Sorensen CM, et al. Transcriptomes of agonist in mice deficient in each subtype. Biochem Pharmacol 2006;4: human prostate cells. BMC Genomics 2006;25:7–92. 453–63. 39. Janowski BA, Grogan MJ, Jones SA, et al. Structural requirements of 15. Alberti S, Schuster G, Parini P, et al. Hepatic cholesterol metabolism ligands for the oxysterol liver X receptors LXRα and LXRβ. Proc Natl Acad and resistance to dietary cholesterol in LXRβ-deficient mice. J Clin Invest Sci U S A 1999;1:266–71. 2001;5:565–73. 40. Wong J, Quinn CM, Brown AJ. Synthesis ofthe oxysterol, 24(S), 25- β 16. Bradley MN, Hong C, Chen M, et al. Ligand activation ofLXR re- epoxycholesterol, parallels cholesterol production and may protect against verses atherosclerosis and cellular cholesterol overload in mice lacking cellular accumulation ofnewly-synthesized cholesterol. Lipids Health Dis α – LXR and apoE. J Clin Invest 2007;8:2337 46. 2007;5:6–10. β β 17. Kim HJ, Fan X, Gabbi C, et al. Liver X receptor (LXR ): a link be- 41. Beyea MM, Heslop CL, Sawyez CG, et al. Selective up-regulation of β tween -sitosterol and amyotrophic lateral sclerosis-Parkinson's dementia. LXR-regulated genes ABCA1, ABCG1, and APOE in macrophages through – Proc Natl Acad Sci U S A 2008;6:2094 9. increased endogenous synthesis of24(S),25-epoxycholesterol. J Biol 18. Kaplan R, Gan X, Menke JG, et al. Bacterial lipopolysaccharide Chem 2007;8:5207–16.

Mol Cancer Ther 2009;8(7). July 2009

OF12 ABCA1 and ABCG1 mRNA in LNCaP and PC-3 Cells

42. Mitchell S, Abel P, Ware M, et al. Phenotypic and genotypic charac- 47. Henklova P, Vrzal R, Papouskova B, et al. SB203580, a pharmacolog- terization ofcommonly used human prostatic cell lines. BJU Int 2000;7: ical inhibitor ofp38 MAP kinase transduction pathway activates ERK 932–44. and JNK MAP kinases in primary cultures ofhuman hepatocytes. Eur J – 43. Chuu CP, Chen RY, Hiipakka RA, et al. The liver X receptor agonist Pharmacol 2008;1:16 23. T0901317 acts as androgen receptor antagonist in human prostate cancer 48. Muniyappa H, Das KC. Activation ofc-Jun N-terminal kinase (JNK) cells. Biochem Biophys Res Commun 2007;2:341–6. by widely used specific p38 MAPK inhibitors SB202190 and 44. Wang N, Ranalletta M, Matsuura F, et al. LXR-induced redistribution of SB203580: a MLK-3-MKK7-dependent mechanism. Cell Signal 2008; ABCG1 to plasma membrane in macrophages enhances cholesterol mass 4:675–83. efflux to HDL. Arterioscler Thromb Vasc Biol 2006;6:1310–6. 49. Nemoto S, Sheng Z, Lin A. Opposing effects of Jun kinase and p38 45. Loesch M, Chen G. The p38 MAPK stress pathway as a tumor sup- mitogen-activated protein kinases on cardiomyocyte hypertrophy. Mol Cell pressor or more? Front Biosci 2008;13:3581–93. Biol 1998;6:3518–26. 46. Gerbod-Giannone MC, Li Y, Holleboom A, et al. TNFα induces ABCA1 50. See RH, Caday-Malcolm RA, Singaraja RR, et al. Protein kinase A through NF-κB in macrophages and in phagocytes ingesting apoptotic site-specific phosphorylation regulates ATP-binding cassette A1 cells. Proc Natl Acad Sci U S A 2006;9:3112–7. (ABCA1)-mediated phospholipid efflux. J Biol Chem 2002;44:41835–42.

Mol Cancer Ther 2009;8(7). July 2009

Ligand, receptor, and cell type−dependent regulation of ABCA1 and ABCG1 mRNA in prostate cancer epithelial cells

Steven E. Trasino, Young S. Kim and Thomas T.Y. Wang

Mol Cancer Ther Published OnlineFirst June 16, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-09-0020

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/early/2009/06/12/1535-7163.MCT-09-0020. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.