- Email: [email protected]
Cardiac Glycosides Decrease Prostate Specific Antigen Expression by Down-Regulation of Prostate Derived Ets Factor
Cardiac Glycosides Decrease Prostate Specific Antigen Expression by Down-Regulation of Prostate Derived Ets Factor Horng-Heng Juang, Yu-Fen Lin, Phei-Lang Chang and Ke-Hung Tsui* From the Department of Anatomy, Chang Gung University (HHJ, YFL) and Department of Urology (PLC, KHT) and Bioinformation Center (HHJ, PLC, KHT), Chang Gung Memorial Hospital, Kwei-Shan, Taiwan, Republic of China
Abbreviations and Acronyms ATPase ⫽ adenosine triphosphatase BCA ⫽ bicinchoninic acid CD-FCS ⫽ dextran coated charcoal FCS ETS ⫽ E 26 FCS ⫽ fetal calf serum LDH ⫽ lactate dehydrogenase MTS ⫽ 3-(4,5-dimethylthiazol-2yl)-5-(3-carboxymethoxyphenyl)-2(4-sulfophenyl)-2H-tetrazolium PCR ⫽ polymerase chain reaction PDEF ⫽ prostate derived Ets factor PSA ⫽ prostate specific antigen RPMI-PRF ⫽ RPMI 1640 phenol red-free Submitted for publication November 24, 2009. Study received institutional review board approval. Supported by Chang Gung Memorial Hospital Research Grants CMRPD-170472, CMRPD-160133 and CMRPG-360413, and the National Science Council, Taiwan, Republic of China Research Grants 97-2320-B-182-023-MY3 and 98-2314-B-182-042MY3. * Correspondence: Department of Urology, Chang Gung Memory Hospital, 5 Fu-Shing St. Kwei-Shan, Tao-Yuan 333, Taiwan, Republic of China (telephone: 886-3-3281200; FAX: 886-32118112; e-mail: [email protected]).
See Editorial on page 1831.
2158
www.jurology.com
Purpose: While cardiac glycosides are the mainstay of congestive heart failure treatment, early studies showed that pharmacological doses of cardiac glycosides inhibited prostate cancer cell line proliferation. We evaluated the mechanisms of cardiac glycosides, including digoxin, digitoxin and ouabain (Sigma®), on prostate specific antigen gene expression in vitro. Materials and Methods: We cultured LNCaP cells (ATCC®) and used them to determine the effect of cardiac glycosides on prostate derived Ets factor and prostate specific antigen expression. We determined prostate derived Ets factor and prostate specific antigen expression by reverse transcription-polymerase chain reaction, immunoblot, transient gene expression assay or enzyme-linked immunosorbent assay. Results: Noncytotoxic doses (100 nM) of cardiac glycosides for 24 hours inhibited prostate specific antigen secretion by LNCaP cells. Reverse transcriptase-polymerase chain reaction and immunoblot revealed that cardiac glycosides significantly down-regulated prostate specific antigen and prostate derived Ets factor expression. Transient gene expression assays showed that prostate derived Ets factor over expression enhanced prostate specific antigen promoter activity. However, prostate specific antigen and prostate derived Ets factor gene promoter activity was attenuated when LNCaP cells were treated with 100 nM cardiac glycosides. When LNCaP cells were treated with 25 nM digitoxin or digoxin for 60 hours, prostate specific antigen secretion decreased by 30%. Conclusions: Results suggest that cardiac glycoside inhibition of prostate specific antigen gene expression may be caused by the down-regulation of prostate derived Ets factor gene expression. When cells were chronically treated with digoxin or digitoxin at concentrations close to or at therapeutic plasma levels, prostate specific antigen secretion decreased. This phenomenon merits further study to determine whether it occurs in men on cardiac glycoside therapy. Key Words: prostate; prostatic neoplasms; prostate-specific antigen; gene expression; SPDEF protein, human PROSTATE specific antigen assay and digital rectal examination are the standards of prostate cancer screening. Several strong risk factors for prostate cancer include patient age, ethnic group, and family history of prostate cancer.1 Using PSA, prostate volume and digital rectal examination
physicians can determine the patient risk group, and estimate the underlying risk of prostate cancer and its aggressiveness.2 While digitalis-like steroids and related agents are the mainstay of congestive heart failure treatment,3 early studies showed that pharmaco-
0022-5347/10/1845-2158/0 THE JOURNAL OF UROLOGY® © 2010 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 184, 2158-2164, November 2010 Printed in U.S.A. DOI:10.1016/j.juro.2010.06.093
AND
RESEARCH, INC.
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
logical doses of cardiac glycosides inhibited prostate cancer cell line proliferation and promoted apoptosis by the ability to induce sustained Ca⫹2 increases in cells.4,5 Except for the growth effect of cardiac glycosides on the human prostate, previous in vitro study suggested that ouabain induces an increased tension response in human prostate tissue due to noradrenaline release via an effect on the Na⫹ dependent Ca⫹2 influx system.6 PDEF, also termed SPDEF/PSE, is a member of the Ets family of transcription factors that is expressed in abundance in prostate tissue.7 In situ hybridization on prostate tissue frozen sections revealed diffuse strong expression restricted to luminal epithelial cells.8 PDEF bound with high affinity to DNA containing sequences in 2 of the 11 putative ETS binding sites in the PSA promoter/enhancer region.7 Multiplex gene expression analysis showed that the ability of cardiac glycosides to induce apoptosis in PC-3 human prostate carcinoma cells correlated with their inhibition of prostate target gene expression, including PDEF.9 Recent studies indicated that phytoestrogen compounds block PSA gene expression via PDEF down-regulation in prostate carcinoma LNCaP cells.10 To our knowledge no group to date has examined in detail how in elderly patients long-term cardiac glycoside therapy may affect the results of clinical test, such as PSA, to detect prostate cancer. We provide the first evidence that cardiac glycosides at nontoxic doses down-regulate PDEF gene expression, which decreases PSA gene expression in vitro.
MATERIALS AND METHODS Materials, and Cell Lines and Culture We used the prostate cancer cell line LNCaP, and digoxin, digitoxin and ouabain. Stock solutions (10 mM) were made of all drugs by dissolving each in 100% ethanol. We used the BCA protein assay kit (Pierce, Rockford, Illinois). Steroids were removed from FCS (HyClone®) by treatment with dextran coated charcoal (Sigma) (1 gm/500 ml FCS), resulting in CD-FCS. Since the structure of cardiac glycosides is similar to that of steroid hormones and the androgen decreased 1-subunit of Na⫹K⫹-ATPase of LNCaP cells,11 experiments in which cells were treated with cardiac glycosides in RPMI-PRF medium (Invitrogen™) were supplemented with CD-FCS to prevent the effect of exogenous steroid hormone in FCS.
Cell Proliferation Assays We seeded 5,000 cells in each well of a 96-well plate with RPMI 1640 medium (Invitrogen) and 10% FCS, and incubated them for 48 hours. Medium was then changed to RPMI-PRF medium with 5% CD-FCS and various concentrations of digoxin, digitoxin or ouabain, as indicated, for an additional 48 hours. We measured cell proliferation rate in response to drugs using the CellTiter 96® AQueous One Solution cell proliferation MTS assay.
2159
Reverse Transcriptase-PCR Total RNA was isolated with TRIzol® reagent and cDNA was synthesized using the SuperScript® III preamplification system according to manufacturer instructions. The PDEF primers (5=-GACCAGTGAGGAGAGCTGGACCGA-3= and 5=-TGACCTTGGGCTCTGGAAGGTCAG-3=) were used to amplify sequences specific to human PDEF mRNA. Sequences of primers for PSA and -actin were previously described.12 PCR products were separated by 2% agarose gel electrophoresis and visualized by ethidium bromide staining.
Western Blot Cells were treated with digoxin, digitoxin or ouabain for 24 hours and then lysed with lysing buffer composed of 62.5 mM tris (pH 6.8), 2% sodium dodecyl sulfate, 10% glycerol, 5% -mercaptoethanol and 7 M urea. Equal amounts of protein (20 g) were loaded on a 10% sodium dodecyl sulfate-polyacrylamide gel and assayed by enhanced chemiluminescence, as described by the manufacturer (Amersham Biosciences™). Blotting membranes were probed with polyclonal PSA antiserum (Dako, Glostrup, Denmark) (1:200), diluted -actin antiserum (C11, Santa Cruz Biotechnology, Santa Cruz, California) (1:1,000) or diluted PDEF antiserum (1:5,000). Rabbit anti-human PDEF serum was prepared at our laboratory, as described previously.12 Immunoblot of the whole LNCaP cell extract revealed 1 major band at 37 kDa and another weak band at 50 kDa, which represented the glycosylated form of PDEF. We analyzed the intensity of different bands using GeneTools of ChemiGenius (Syngene, Cambridge, United Kingdom).
PSA Enzyme-Linked Immunosorbent Assay LNCaP cells were incubated with 1 ml RPMI-PRF medium with 5% CD-FCS and digoxin, digitoxin or ouabain as indicated in a 6-well plate (2 ⫻ 105 cells per well) for 24 hours. After incubation the conditioned medium supernatant from each well and cell pellets were collected for PSA enzyme-linked immunosorbent assay, as previously described.13 The PSA level in each sample was adjusted by the protein concentration in the whole cell extract, which was measured using the BCA protein assay kit.
LDH Assay LNCaP cells were incubated with RPMI-PRF medium with 5% CD-FCS and 100 nM digoxin, digitoxin or ouabain for 24 hours. LDH activity was measured at 30C as the amount of pyruvate consumed by continuously monitoring the decrease in absorbance due to nicotinamide adenine dinucleotide oxidation at 339 nm. LDH enzymatic activity was adjusted by the protein concentration of the cytosolic extracts, which were determined by BCA protein assay, as previously described.14
Reporter Vector Constructs The reporter vectors pPSAH ⫺41 to ⫺5874), pPSABHE (⫺4801 to ⫺3933 and ⫺41 to ⫺589), pPSAKH (⫺41 to ⫺1557) and pPSABH (⫺41 to ⫺589) containing the 5=flanking region of the human PSA gene were cloned by 5=-deletion or PCR, as previously described.15 DNA fragments (⫺1 to ⫺3280) of the promoter/enhancer of the
2160
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
PDEF gene were isolated from the BAC clone (RP11375E1, Invitrogen), as previously described.12
Transient Transfection and Reporter Assay LNCaP cells were plated on 24-well plates at 1 ⫻ 104 cells per well 1 day before transfection. Cells were transiently transfected using TransFast™ transfection reagent (0.6 g per well) with 1 g per well reporter vector and 0.5 g per well pCMVSPORT-gal (Invitrogen), as described previously.15 Transfected cells were treated with RPMI-PRF medium with 5% CD-FCS and digoxin, digitoxin or ouabain, as indicated, for an additional 24 hours. Luciferase activity was adjusted for transfection efficiency using the normalization control plasmid pCMVSPORT-gal.
Statistical Analysis Results are shown as the mean ⫾ SE of at least 3 independent replications of each experiment. Statistical significance was determined by the paired Student t test using SigmaStat® for Windows®, version 2.03.
RESULTS We evaluated the effects of ouabain, digitoxin and digoxin on proliferation of the androgen dependent prostate cancer cell line LNCaP. In vitro study using MTS assay revealed that 48-hour cardiac glycoside treatment inhibited cell proliferation in a dose dependent pattern in LNCaP cells. Ouabain, digitoxin and digoxin produced cytotoxic effects in LNCaP cells at concentrations more than 250 nM (fig. 1). Ouabain treatment was more cytotoxic than digitoxin or digoxin treatment. Thus, in the following experiments we decreased cardiac glycoside doses to the noncytotoxic dose of 100 nM. Semiquantitative reverse transcriptase-PCR and immunoblot showed that when LNCaP cells were treated with 100 nM ouabain, digitoxin or digoxin, PSA and PDEF gene expression was significantly decreased (fig. 2, A and B). Figure 2, C and D show the results of quantitative analysis of PSA and PDEF expression in 3 independent experiments. When LNCaP cells were co-transfected with PDEF expression vector and the PSA reporter vector pPSABHE containing the basic promoter and enhancer DNA fragments of the PSA gene, transient gene expression assay revealed that forced over expression of PDEF induced 2 to 4-fold up-regulation in PSA reporter activity with or without androgen stimulation (fig. 3, A). PDEF effects are apparently mediated via a region located at enhancer region of the PSA gene according to 5=-deletion reporter assays (fig. 3, B). Using the pPSABHE reporter vector, transient gene expression assays revealed that promoter activity was decreased 40% to 50% when LNCaP cells were treated with ouabain, digitoxin or digoxin at 100 nM for 24 hours (fig. 3, C). The effects of digitoxin on PSA promoter activity also depended on the enhancer region of the PSA gene according to 5=-deletion reporter assays (fig. 3, D).
Figure 1. MTS assay shows mean ⫾ SE percent proliferation by digitalis glycosides in 8 preparations of human prostatic LNCaP cells treated with different concentrations of ouabain (A), digitoxin (B) or digoxin (C) vs control.
Enzyme-linked immunosorbent assay also revealed that when LNCaP cells were treated with 100 nM ouabain, digitoxin or digoxin, PSA secretion decreased 30% to 50% compared to that in the sham treatment group (fig. 4, A). Nonetheless, ouabain, digitoxin and digoxin treatment did not significantly inhibit intracellular LDH activity (fig. 4, B). Results suggest that the effects of cardiac glycoside treatment are not due to the cytotoxic effect of cardiac glycosides. Immunoblot and transient gene expression assays revealed that PSA and PDEF gene expression was decreased 30% to 50% compared to that in the sham treatment group when LNCaP cells were treated with 100 nM digitoxin for 24 hours (fig. 4, C to E). Finally, we determined whether the established chronic low dose regimens used to treat patients with congestive heart failure could also affect PSA
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
2161
Figure 2. Mean ⫾ SE PSA/-actin and PDEF/-actin density induced in LNCaP cells treated with cardiac glycosides for 24 hours vs control (COL). PDEF, -actin and PSA were determined by reverse transcriptase-PCR (A) and immunoblot (B). Di, digoxin. Dit, digitoxin. Ou, oubain. Immunoblot quantitative analysis was done by determining intensity of each band for PDEF (C), PSA (D) and -actin.
gene expression in vitro. When LNCaP cells were treated with 6.25 to 50 nM digoxin or digitoxin for 60 hours, intracellular PSA biosynthesis and PSA secretion were blocked in a dose dependent manner (fig. 5, A to D). LNCaP cell PSA synthesis and secretion were blocked up to 30% when cells were treated with 25 nM digitoxin or digoxin for 60 hours. Immunoblot revealed that PSA and PDEF gene expression was attenuated when cells were chronically treated with low dose digoxin or digitoxin within the range commonly found in the plasma of cardiac patients (fig. 5, C and F).
DISCUSSION The ETS family genes of transcription factors, which regulate a number of biological processes such as cell proliferation, differentiation and invasion, are thought to have an important role in oncogenesis.16 Recent in vitro and in vivo studies indicated that dysregulation of ETS family members through fusion with transmembrane protease serine 2 may be an initiating event in prostate cancer.17–19 The PDEF gene is an ETS family gene that was identified in a subtracted prostate benign hyperplasia cDNA library.7 Subsequent studies showed that loss of PDEF protein expression occurred in progression from benign prostate disease to carcinoma.20 However, contrary results from different independent laboratories frequently showed increased PDEF protein expression in prostate tumors.8 An early study indicated that human PDEF interacts with androgen receptor to enhance expres-
sion of PSA promoter in vitro.7 In that study using a promoter-reporter construct co-transfected with PDEF into simian kidney cells investigators determined that PDEF is an androgen independent transcriptional activator of PSA promoter. PDEF and androgen receptor co-activators/co-repressors may differentially modulate androgen receptor transcriptional activity in the promoter/enhancer region of kallikrein 2 and PSA in cells.21 Several phytochemicals, such as silibinin and tectorigenin, down-regulate PDEF expression and, thus, PSA secretion with or without androgen stimulation in prostate carcinoma LNCaP cells.10 A study of large-scale gene screening showed that a class of steroids related to cardiac glycosides potently inhibited plasma membrane Na⫹K⫹-ATPase, resulting in PDEF inhibition in prostate cancer PC-3 cells.9 Our results demonstrate that digitalis glycosides inhibit PSA gene expression through the down-regulation of gene expression of PDEF in LNCaP cells. While to our knowledge no current report details the precise regulatory pathway of the PDEF gene, studies indicate that modulators of the Ets family of genes include extracellular regulated kinase, JNK, p38, receptor tyrosine kinase, transforming growth factor-, p300, CAMKII and integrin.22 The regulatory mechanisms of digitalis glycosides on PDEF gene expression still needs further investigation. Although the signal transduction pathway from Na⫹K⫹-ATPase is still not well-known, several studies suggest that the signal is transmitted sequentially from Ras, Raf-1, Src/epidermal growth factor
2162
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
Figure 3. Modulation of PSA promoter activity by PDEF in LNCaP cells. Data are shown as mean ⫾ SE percent luciferase activity induced by different treatments in 6 preparations vs control. PSA promoter-luciferase reporter vector pPSABHE and PDEF expression vector co-transfected LNCaP cells were treated with (white bars) or without (black bars) 1 nM R1881 for 24 hours (A). Asterisk indicates statistically significantly difference vs sham transfected cells (p ⬍0.01). Plus sign indicates statistically significantly difference vs R1881 treated, sham transfected cells (p ⬍0.01). Human PDEF in pcDNA3 (white bars) or sham pcDNA3 (black bars) was co-transfected with PSA promoter-luciferase reporter vectors (B). Asterisk indicates p ⬍0.01. PSA promoterluciferase reporter vector transfected LNCaP cells were treated with ouabain (Ou), digitoxin (Dit) or digoxin (Di) (C). PSA promoter-luciferase reporter vectors transfected LNCaP cells were treated with (white box) or without (black box) 100 nM digitoxin for 24 hours (D). Asterisk indicates p ⬍0.01.
receptor, extracellular regulated kinase 1/2 and mitogen-activated protein kinases.23,24 Other studies showed that cardiac glycosides inhibited tumor necrosis factor-␣/nuclear factor-B signaling in Hela and 293 cells.25 Thus, we cannot rule out the possibility that PSA gene expression inhibition by cardiac glycosides may also act through signal pathways other than PDEF. The high incidence of prostate cancer in aging male patients has prompted many urologists to screen this population more vigorously than others. Some investigators question whether PSA screening is an effective screening method for prostate cancer in patients with heart failure who are also on cardiac glycoside therapy. We noted that at the nonlethal concentration of 100 nM digoxin, digitoxin and ouabain exert significant inhibitory effects on PSA expression in prostate cancer cell lines. Nonetheless, an important question is whether similar cardiac glycoside effects on PSA secretion from prostate carcinoma cells in vitro can be expected in patients
treated with the established dose regimens of cardiac glycosides used to treat congestive heart failure. Therapeutic plasma levels of digoxin and digitoxin are considered to be within the range of 0.6 to 1.9 and 13 to 33 nM, respectively.26 Thus, we studied chronic low dose treatment. Digitoxin significantly inhibited PSA secretion at a concentration of 25 nM, which is commonly found in plasma samples of cardiac patients. Based on these data, cardiac glycoside therapy may confound the interpretation of PSA levels and the risk of prostate cancer in elderly patients on long-term cardiac glycoside therapy. To our knowledge the probability of prostate cancer in aging men treated with cardiac glycosides who have lower PSA than those not treated with digitalis compounds is yet to be determined.
CONCLUSIONS Nonlethal or chronic low dose cardiac glycoside treatment decreased PSA gene expression via the
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
2163
Figure 4. Digitoxin regulation of PSA and PDEF expression in LNCaP cells treated with cardiac glycosides for 24 hours. Data are shown as mean ⫾ SE percent PSA or LDH induced by different treatments vs control (COL) in 6 preparations each. Conditioned medium was collected for PSA assay (A) and cell pellets were assayed for LDH assay (B). Di, digoxin. Dit, digitoxin. Ou, oubain. Cells were treated with various concentrations of digitoxin. PDEF, -actin and PSA were determined by immunoblot (C). PSA promoter-luciferase reporter vector transfected (D) and PDEF promoter-luciferase reporter vector pPDEFSH transfected (E) LNCaP cells were treated with different digitoxin concentrations. Asterisk indicates p ⬍0.01.
Figure 5. Effect of chronic digitalis glycoside treatment on PSA secretion by LNCaP cells treated with various concentrations of digoxin (A and B) or digitoxin (D and E) for 60 hours. Conditioned medium (A and D) and cell pellets (B and E) were collected for PSA assay. Data are shown as mean ⫾ SE percent PSA induced by different treatments in 6 preparations each vs control. Asterisk indicates p ⬍0.01. Cells were treated with various concentrations of digoxin (C) or digitoxin (F) for 60 hours. PDEF, -actin and PSA were determined by immunoblot.
2164
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
attenuation of PDEF gene expression in vitro. We suggest that PSA interpretation in patients on cardiac glycosides could result in underestimating
prostate cancer risk and pathological stage. Thus, this phenomenon merits further study to determine whether it occurs in vivo.
REFERENCES 1. Stamey TA, Caldwell M, McNeal JE et al: The prostate specific antigen era in the United States is over for prostate cancer: what happened in the last 20 years? J Urol 2004; 172: 1297. 2. Catalona WJ, Southwick PC, Slawin KM et al: Comparison of percent free PSA, PSA density and age-specific PSA cutoffs for prostate cancer detection and staging. Urology 2000; 56: 255. 3. Schoner W and Scheiner-Bobis G: Endogenous and exogenous cardiac glycosides: their roles in hypertension, salt metabolism, and cell growth. Am J Physiol Cell Physiol 2007; 293: C509. 4. Yen JY, Huang WJ, Kan SF et al: Inhibitory effects of digitalis on the proliferation of androgen dependent and independent prostate cancer cells. J Urol 2001; 166: 1937. 5. Huang YT, Chueh SC, Teng CM et al: Investigation of ouabain-induced anticancer effect in human androgen-independent prostate cancer PC-3 cells. Biochem Pharmacol 2004; 67: 723.
put drug discovery: screening and analysis of compounds affecting genes overexpressed in cancer cells. Mol Cancer Ther 2002; 1: 1293. 10. Thelen P, Peter T, Hunermund A et al: Phytoestrogens from Belamcanda chinensis regulate the expression of steroid receptors and related cofactors in LNCaP prostate cancer cells. BJU Int 2007; 100: 199. 11. Schoner W: Endogenous cardiac glycosides, a new class of steroid hormones. Eur J Biochem 2002; 269: 2440.
19. Tomlins SA, Laxman B, Varambally S et al: Role of the TMPRSS2: EGR gene fusion in prostate cancer. Neoplasia 2008; 10: 177. 20. Gu X, Zerbini LF, Otu HH et al: Reduced PDEF expression increases invasion and expression of mesenchymal genes in prostate cancer cells. Cancer Res 2007; 67: 4219.
12. Tsui KH, Feng TH, Lin CM et al: Curcumin blocks the activating of androgen and interlukin-6 on prostate specific antigen expression in human prostatic carcinoma cells. J Androl 2008; 29: 661.
21. Magklara A, Brown TJ and Diamandis EP: Characterization of androgen receptor and nuclear receptor co-regulator expression in human breast cancer cell lines exhibiting differential regulation of kallikreins 2 and 3. Int J Cancer 2002; 100: 507.
13. Tsiu KH, Chang PL, Lin HT et al: Down-regulation of the prostate specific antigen promoter by p53 in human prostate cancer cells. J Urol 2004; 172: 2305.
22. Hsu T, Trojanowska M and Watson DK: Ets proteins in biological control and cancer. J Cell Biochem 2004; 91: 896.
6. Guh JH, Chueh SC and Teng CM: Effects of ouabain on tension response and [3H]noradrenaline release in human prostate. J Urol 2000; 163: 338.
14. Juang HH: Nitroprusside stimulates mitochondrial aconitase gene expression through the cyclic adenosine 3=,5=-monosphosphate signal transduction pathway in human prostate carcinoma cells. Prostate 2004; 61: 92.
7. Oettgen P, Finger E, Sun Z et al: PDEF, a novel prostate epithelium-specific ets transcription factor, interacts with the androgen receptor and activates prostate-specific antigen gene expression. J Biol Chem 2000; 275: 1216.
15. Tsiu KH, Wu L, Chang PL et al: Identifying the combination of the transcriptional regulatory sequences on prostate specific antigen and human glandular kallikrein genes. J Urol 2004; 172: 2029.
8. Sood AK, Saxena R, Groth J et al: Expression characteristics of prostate-derived Ets factor support a role in breast and prostate cancer progression. Hum Pathol 2007; 38: 1628.
16. Seth A and Waston DK: ETS transcription factors and their emerging roles in human cancer. Eur J Cancer 2005; 41: 2462.
9. Johnson PH, Walker RP, Jones SW et al: Multiplex gene expression analysis for high-through-
18. Cai C, Wang H, Xu Y et al: Reactivation of androgen receptor-regulated TMPRSS2: EGR gene expression in castration-resistant prostate cancer. Cancer Res 2009; 69: 6027.
17. Tomlins SA, Bjartell A, Chinnaiyan AM et al: ETS gene fusions in prostate cancer: from discovery to daily clinical practice. Eur Urol 2009; 56: 275.
23. Kometiani P, Liu L and Askari A: Digitalis-induced signaling by Na⫹/K⫹-ATPase in human breast cancer cells. Mol Pharmacol 2005; 67: 929. 24. Lopez-Lazaro M: Digitoxin as an anticancer agent with selectivity for cancer cells: possible mechanisms involved. Expert Opin Ther Targets 2007; 11: 1043. 25. Yang Q, Huang W, Jozwik C et al: Cardiac glycosides inhibit TNF-␣/NF-B signaling by blocking recruitment of TNF receptor-associated death domain to the TNF receptor. PNAS 2005; 102: 9631. 26. Katzung BG and Parmley WW: Cardiac glycosides and other drugs used in congestive heart failure. In: Basic and Clinical Pharmacology, 8th ed. Edited by BG Katzung. New York: Appleton & Lange 2000; pp 200 –218.
Abbreviations and Acronyms ATPase ⫽ adenosine triphosphatase BCA ⫽ bicinchoninic acid CD-FCS ⫽ dextran coated charcoal FCS ETS ⫽ E 26 FCS ⫽ fetal calf serum LDH ⫽ lactate dehydrogenase MTS ⫽ 3-(4,5-dimethylthiazol-2yl)-5-(3-carboxymethoxyphenyl)-2(4-sulfophenyl)-2H-tetrazolium PCR ⫽ polymerase chain reaction PDEF ⫽ prostate derived Ets factor PSA ⫽ prostate specific antigen RPMI-PRF ⫽ RPMI 1640 phenol red-free Submitted for publication November 24, 2009. Study received institutional review board approval. Supported by Chang Gung Memorial Hospital Research Grants CMRPD-170472, CMRPD-160133 and CMRPG-360413, and the National Science Council, Taiwan, Republic of China Research Grants 97-2320-B-182-023-MY3 and 98-2314-B-182-042MY3. * Correspondence: Department of Urology, Chang Gung Memory Hospital, 5 Fu-Shing St. Kwei-Shan, Tao-Yuan 333, Taiwan, Republic of China (telephone: 886-3-3281200; FAX: 886-32118112; e-mail: [email protected]).
See Editorial on page 1831.
2158
www.jurology.com
Purpose: While cardiac glycosides are the mainstay of congestive heart failure treatment, early studies showed that pharmacological doses of cardiac glycosides inhibited prostate cancer cell line proliferation. We evaluated the mechanisms of cardiac glycosides, including digoxin, digitoxin and ouabain (Sigma®), on prostate specific antigen gene expression in vitro. Materials and Methods: We cultured LNCaP cells (ATCC®) and used them to determine the effect of cardiac glycosides on prostate derived Ets factor and prostate specific antigen expression. We determined prostate derived Ets factor and prostate specific antigen expression by reverse transcription-polymerase chain reaction, immunoblot, transient gene expression assay or enzyme-linked immunosorbent assay. Results: Noncytotoxic doses (100 nM) of cardiac glycosides for 24 hours inhibited prostate specific antigen secretion by LNCaP cells. Reverse transcriptase-polymerase chain reaction and immunoblot revealed that cardiac glycosides significantly down-regulated prostate specific antigen and prostate derived Ets factor expression. Transient gene expression assays showed that prostate derived Ets factor over expression enhanced prostate specific antigen promoter activity. However, prostate specific antigen and prostate derived Ets factor gene promoter activity was attenuated when LNCaP cells were treated with 100 nM cardiac glycosides. When LNCaP cells were treated with 25 nM digitoxin or digoxin for 60 hours, prostate specific antigen secretion decreased by 30%. Conclusions: Results suggest that cardiac glycoside inhibition of prostate specific antigen gene expression may be caused by the down-regulation of prostate derived Ets factor gene expression. When cells were chronically treated with digoxin or digitoxin at concentrations close to or at therapeutic plasma levels, prostate specific antigen secretion decreased. This phenomenon merits further study to determine whether it occurs in men on cardiac glycoside therapy. Key Words: prostate; prostatic neoplasms; prostate-specific antigen; gene expression; SPDEF protein, human PROSTATE specific antigen assay and digital rectal examination are the standards of prostate cancer screening. Several strong risk factors for prostate cancer include patient age, ethnic group, and family history of prostate cancer.1 Using PSA, prostate volume and digital rectal examination
physicians can determine the patient risk group, and estimate the underlying risk of prostate cancer and its aggressiveness.2 While digitalis-like steroids and related agents are the mainstay of congestive heart failure treatment,3 early studies showed that pharmaco-
0022-5347/10/1845-2158/0 THE JOURNAL OF UROLOGY® © 2010 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 184, 2158-2164, November 2010 Printed in U.S.A. DOI:10.1016/j.juro.2010.06.093
AND
RESEARCH, INC.
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
logical doses of cardiac glycosides inhibited prostate cancer cell line proliferation and promoted apoptosis by the ability to induce sustained Ca⫹2 increases in cells.4,5 Except for the growth effect of cardiac glycosides on the human prostate, previous in vitro study suggested that ouabain induces an increased tension response in human prostate tissue due to noradrenaline release via an effect on the Na⫹ dependent Ca⫹2 influx system.6 PDEF, also termed SPDEF/PSE, is a member of the Ets family of transcription factors that is expressed in abundance in prostate tissue.7 In situ hybridization on prostate tissue frozen sections revealed diffuse strong expression restricted to luminal epithelial cells.8 PDEF bound with high affinity to DNA containing sequences in 2 of the 11 putative ETS binding sites in the PSA promoter/enhancer region.7 Multiplex gene expression analysis showed that the ability of cardiac glycosides to induce apoptosis in PC-3 human prostate carcinoma cells correlated with their inhibition of prostate target gene expression, including PDEF.9 Recent studies indicated that phytoestrogen compounds block PSA gene expression via PDEF down-regulation in prostate carcinoma LNCaP cells.10 To our knowledge no group to date has examined in detail how in elderly patients long-term cardiac glycoside therapy may affect the results of clinical test, such as PSA, to detect prostate cancer. We provide the first evidence that cardiac glycosides at nontoxic doses down-regulate PDEF gene expression, which decreases PSA gene expression in vitro.
MATERIALS AND METHODS Materials, and Cell Lines and Culture We used the prostate cancer cell line LNCaP, and digoxin, digitoxin and ouabain. Stock solutions (10 mM) were made of all drugs by dissolving each in 100% ethanol. We used the BCA protein assay kit (Pierce, Rockford, Illinois). Steroids were removed from FCS (HyClone®) by treatment with dextran coated charcoal (Sigma) (1 gm/500 ml FCS), resulting in CD-FCS. Since the structure of cardiac glycosides is similar to that of steroid hormones and the androgen decreased 1-subunit of Na⫹K⫹-ATPase of LNCaP cells,11 experiments in which cells were treated with cardiac glycosides in RPMI-PRF medium (Invitrogen™) were supplemented with CD-FCS to prevent the effect of exogenous steroid hormone in FCS.
Cell Proliferation Assays We seeded 5,000 cells in each well of a 96-well plate with RPMI 1640 medium (Invitrogen) and 10% FCS, and incubated them for 48 hours. Medium was then changed to RPMI-PRF medium with 5% CD-FCS and various concentrations of digoxin, digitoxin or ouabain, as indicated, for an additional 48 hours. We measured cell proliferation rate in response to drugs using the CellTiter 96® AQueous One Solution cell proliferation MTS assay.
2159
Reverse Transcriptase-PCR Total RNA was isolated with TRIzol® reagent and cDNA was synthesized using the SuperScript® III preamplification system according to manufacturer instructions. The PDEF primers (5=-GACCAGTGAGGAGAGCTGGACCGA-3= and 5=-TGACCTTGGGCTCTGGAAGGTCAG-3=) were used to amplify sequences specific to human PDEF mRNA. Sequences of primers for PSA and -actin were previously described.12 PCR products were separated by 2% agarose gel electrophoresis and visualized by ethidium bromide staining.
Western Blot Cells were treated with digoxin, digitoxin or ouabain for 24 hours and then lysed with lysing buffer composed of 62.5 mM tris (pH 6.8), 2% sodium dodecyl sulfate, 10% glycerol, 5% -mercaptoethanol and 7 M urea. Equal amounts of protein (20 g) were loaded on a 10% sodium dodecyl sulfate-polyacrylamide gel and assayed by enhanced chemiluminescence, as described by the manufacturer (Amersham Biosciences™). Blotting membranes were probed with polyclonal PSA antiserum (Dako, Glostrup, Denmark) (1:200), diluted -actin antiserum (C11, Santa Cruz Biotechnology, Santa Cruz, California) (1:1,000) or diluted PDEF antiserum (1:5,000). Rabbit anti-human PDEF serum was prepared at our laboratory, as described previously.12 Immunoblot of the whole LNCaP cell extract revealed 1 major band at 37 kDa and another weak band at 50 kDa, which represented the glycosylated form of PDEF. We analyzed the intensity of different bands using GeneTools of ChemiGenius (Syngene, Cambridge, United Kingdom).
PSA Enzyme-Linked Immunosorbent Assay LNCaP cells were incubated with 1 ml RPMI-PRF medium with 5% CD-FCS and digoxin, digitoxin or ouabain as indicated in a 6-well plate (2 ⫻ 105 cells per well) for 24 hours. After incubation the conditioned medium supernatant from each well and cell pellets were collected for PSA enzyme-linked immunosorbent assay, as previously described.13 The PSA level in each sample was adjusted by the protein concentration in the whole cell extract, which was measured using the BCA protein assay kit.
LDH Assay LNCaP cells were incubated with RPMI-PRF medium with 5% CD-FCS and 100 nM digoxin, digitoxin or ouabain for 24 hours. LDH activity was measured at 30C as the amount of pyruvate consumed by continuously monitoring the decrease in absorbance due to nicotinamide adenine dinucleotide oxidation at 339 nm. LDH enzymatic activity was adjusted by the protein concentration of the cytosolic extracts, which were determined by BCA protein assay, as previously described.14
Reporter Vector Constructs The reporter vectors pPSAH ⫺41 to ⫺5874), pPSABHE (⫺4801 to ⫺3933 and ⫺41 to ⫺589), pPSAKH (⫺41 to ⫺1557) and pPSABH (⫺41 to ⫺589) containing the 5=flanking region of the human PSA gene were cloned by 5=-deletion or PCR, as previously described.15 DNA fragments (⫺1 to ⫺3280) of the promoter/enhancer of the
2160
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
PDEF gene were isolated from the BAC clone (RP11375E1, Invitrogen), as previously described.12
Transient Transfection and Reporter Assay LNCaP cells were plated on 24-well plates at 1 ⫻ 104 cells per well 1 day before transfection. Cells were transiently transfected using TransFast™ transfection reagent (0.6 g per well) with 1 g per well reporter vector and 0.5 g per well pCMVSPORT-gal (Invitrogen), as described previously.15 Transfected cells were treated with RPMI-PRF medium with 5% CD-FCS and digoxin, digitoxin or ouabain, as indicated, for an additional 24 hours. Luciferase activity was adjusted for transfection efficiency using the normalization control plasmid pCMVSPORT-gal.
Statistical Analysis Results are shown as the mean ⫾ SE of at least 3 independent replications of each experiment. Statistical significance was determined by the paired Student t test using SigmaStat® for Windows®, version 2.03.
RESULTS We evaluated the effects of ouabain, digitoxin and digoxin on proliferation of the androgen dependent prostate cancer cell line LNCaP. In vitro study using MTS assay revealed that 48-hour cardiac glycoside treatment inhibited cell proliferation in a dose dependent pattern in LNCaP cells. Ouabain, digitoxin and digoxin produced cytotoxic effects in LNCaP cells at concentrations more than 250 nM (fig. 1). Ouabain treatment was more cytotoxic than digitoxin or digoxin treatment. Thus, in the following experiments we decreased cardiac glycoside doses to the noncytotoxic dose of 100 nM. Semiquantitative reverse transcriptase-PCR and immunoblot showed that when LNCaP cells were treated with 100 nM ouabain, digitoxin or digoxin, PSA and PDEF gene expression was significantly decreased (fig. 2, A and B). Figure 2, C and D show the results of quantitative analysis of PSA and PDEF expression in 3 independent experiments. When LNCaP cells were co-transfected with PDEF expression vector and the PSA reporter vector pPSABHE containing the basic promoter and enhancer DNA fragments of the PSA gene, transient gene expression assay revealed that forced over expression of PDEF induced 2 to 4-fold up-regulation in PSA reporter activity with or without androgen stimulation (fig. 3, A). PDEF effects are apparently mediated via a region located at enhancer region of the PSA gene according to 5=-deletion reporter assays (fig. 3, B). Using the pPSABHE reporter vector, transient gene expression assays revealed that promoter activity was decreased 40% to 50% when LNCaP cells were treated with ouabain, digitoxin or digoxin at 100 nM for 24 hours (fig. 3, C). The effects of digitoxin on PSA promoter activity also depended on the enhancer region of the PSA gene according to 5=-deletion reporter assays (fig. 3, D).
Figure 1. MTS assay shows mean ⫾ SE percent proliferation by digitalis glycosides in 8 preparations of human prostatic LNCaP cells treated with different concentrations of ouabain (A), digitoxin (B) or digoxin (C) vs control.
Enzyme-linked immunosorbent assay also revealed that when LNCaP cells were treated with 100 nM ouabain, digitoxin or digoxin, PSA secretion decreased 30% to 50% compared to that in the sham treatment group (fig. 4, A). Nonetheless, ouabain, digitoxin and digoxin treatment did not significantly inhibit intracellular LDH activity (fig. 4, B). Results suggest that the effects of cardiac glycoside treatment are not due to the cytotoxic effect of cardiac glycosides. Immunoblot and transient gene expression assays revealed that PSA and PDEF gene expression was decreased 30% to 50% compared to that in the sham treatment group when LNCaP cells were treated with 100 nM digitoxin for 24 hours (fig. 4, C to E). Finally, we determined whether the established chronic low dose regimens used to treat patients with congestive heart failure could also affect PSA
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
2161
Figure 2. Mean ⫾ SE PSA/-actin and PDEF/-actin density induced in LNCaP cells treated with cardiac glycosides for 24 hours vs control (COL). PDEF, -actin and PSA were determined by reverse transcriptase-PCR (A) and immunoblot (B). Di, digoxin. Dit, digitoxin. Ou, oubain. Immunoblot quantitative analysis was done by determining intensity of each band for PDEF (C), PSA (D) and -actin.
gene expression in vitro. When LNCaP cells were treated with 6.25 to 50 nM digoxin or digitoxin for 60 hours, intracellular PSA biosynthesis and PSA secretion were blocked in a dose dependent manner (fig. 5, A to D). LNCaP cell PSA synthesis and secretion were blocked up to 30% when cells were treated with 25 nM digitoxin or digoxin for 60 hours. Immunoblot revealed that PSA and PDEF gene expression was attenuated when cells were chronically treated with low dose digoxin or digitoxin within the range commonly found in the plasma of cardiac patients (fig. 5, C and F).
DISCUSSION The ETS family genes of transcription factors, which regulate a number of biological processes such as cell proliferation, differentiation and invasion, are thought to have an important role in oncogenesis.16 Recent in vitro and in vivo studies indicated that dysregulation of ETS family members through fusion with transmembrane protease serine 2 may be an initiating event in prostate cancer.17–19 The PDEF gene is an ETS family gene that was identified in a subtracted prostate benign hyperplasia cDNA library.7 Subsequent studies showed that loss of PDEF protein expression occurred in progression from benign prostate disease to carcinoma.20 However, contrary results from different independent laboratories frequently showed increased PDEF protein expression in prostate tumors.8 An early study indicated that human PDEF interacts with androgen receptor to enhance expres-
sion of PSA promoter in vitro.7 In that study using a promoter-reporter construct co-transfected with PDEF into simian kidney cells investigators determined that PDEF is an androgen independent transcriptional activator of PSA promoter. PDEF and androgen receptor co-activators/co-repressors may differentially modulate androgen receptor transcriptional activity in the promoter/enhancer region of kallikrein 2 and PSA in cells.21 Several phytochemicals, such as silibinin and tectorigenin, down-regulate PDEF expression and, thus, PSA secretion with or without androgen stimulation in prostate carcinoma LNCaP cells.10 A study of large-scale gene screening showed that a class of steroids related to cardiac glycosides potently inhibited plasma membrane Na⫹K⫹-ATPase, resulting in PDEF inhibition in prostate cancer PC-3 cells.9 Our results demonstrate that digitalis glycosides inhibit PSA gene expression through the down-regulation of gene expression of PDEF in LNCaP cells. While to our knowledge no current report details the precise regulatory pathway of the PDEF gene, studies indicate that modulators of the Ets family of genes include extracellular regulated kinase, JNK, p38, receptor tyrosine kinase, transforming growth factor-, p300, CAMKII and integrin.22 The regulatory mechanisms of digitalis glycosides on PDEF gene expression still needs further investigation. Although the signal transduction pathway from Na⫹K⫹-ATPase is still not well-known, several studies suggest that the signal is transmitted sequentially from Ras, Raf-1, Src/epidermal growth factor
2162
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
Figure 3. Modulation of PSA promoter activity by PDEF in LNCaP cells. Data are shown as mean ⫾ SE percent luciferase activity induced by different treatments in 6 preparations vs control. PSA promoter-luciferase reporter vector pPSABHE and PDEF expression vector co-transfected LNCaP cells were treated with (white bars) or without (black bars) 1 nM R1881 for 24 hours (A). Asterisk indicates statistically significantly difference vs sham transfected cells (p ⬍0.01). Plus sign indicates statistically significantly difference vs R1881 treated, sham transfected cells (p ⬍0.01). Human PDEF in pcDNA3 (white bars) or sham pcDNA3 (black bars) was co-transfected with PSA promoter-luciferase reporter vectors (B). Asterisk indicates p ⬍0.01. PSA promoterluciferase reporter vector transfected LNCaP cells were treated with ouabain (Ou), digitoxin (Dit) or digoxin (Di) (C). PSA promoter-luciferase reporter vectors transfected LNCaP cells were treated with (white box) or without (black box) 100 nM digitoxin for 24 hours (D). Asterisk indicates p ⬍0.01.
receptor, extracellular regulated kinase 1/2 and mitogen-activated protein kinases.23,24 Other studies showed that cardiac glycosides inhibited tumor necrosis factor-␣/nuclear factor-B signaling in Hela and 293 cells.25 Thus, we cannot rule out the possibility that PSA gene expression inhibition by cardiac glycosides may also act through signal pathways other than PDEF. The high incidence of prostate cancer in aging male patients has prompted many urologists to screen this population more vigorously than others. Some investigators question whether PSA screening is an effective screening method for prostate cancer in patients with heart failure who are also on cardiac glycoside therapy. We noted that at the nonlethal concentration of 100 nM digoxin, digitoxin and ouabain exert significant inhibitory effects on PSA expression in prostate cancer cell lines. Nonetheless, an important question is whether similar cardiac glycoside effects on PSA secretion from prostate carcinoma cells in vitro can be expected in patients
treated with the established dose regimens of cardiac glycosides used to treat congestive heart failure. Therapeutic plasma levels of digoxin and digitoxin are considered to be within the range of 0.6 to 1.9 and 13 to 33 nM, respectively.26 Thus, we studied chronic low dose treatment. Digitoxin significantly inhibited PSA secretion at a concentration of 25 nM, which is commonly found in plasma samples of cardiac patients. Based on these data, cardiac glycoside therapy may confound the interpretation of PSA levels and the risk of prostate cancer in elderly patients on long-term cardiac glycoside therapy. To our knowledge the probability of prostate cancer in aging men treated with cardiac glycosides who have lower PSA than those not treated with digitalis compounds is yet to be determined.
CONCLUSIONS Nonlethal or chronic low dose cardiac glycoside treatment decreased PSA gene expression via the
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
2163
Figure 4. Digitoxin regulation of PSA and PDEF expression in LNCaP cells treated with cardiac glycosides for 24 hours. Data are shown as mean ⫾ SE percent PSA or LDH induced by different treatments vs control (COL) in 6 preparations each. Conditioned medium was collected for PSA assay (A) and cell pellets were assayed for LDH assay (B). Di, digoxin. Dit, digitoxin. Ou, oubain. Cells were treated with various concentrations of digitoxin. PDEF, -actin and PSA were determined by immunoblot (C). PSA promoter-luciferase reporter vector transfected (D) and PDEF promoter-luciferase reporter vector pPDEFSH transfected (E) LNCaP cells were treated with different digitoxin concentrations. Asterisk indicates p ⬍0.01.
Figure 5. Effect of chronic digitalis glycoside treatment on PSA secretion by LNCaP cells treated with various concentrations of digoxin (A and B) or digitoxin (D and E) for 60 hours. Conditioned medium (A and D) and cell pellets (B and E) were collected for PSA assay. Data are shown as mean ⫾ SE percent PSA induced by different treatments in 6 preparations each vs control. Asterisk indicates p ⬍0.01. Cells were treated with various concentrations of digoxin (C) or digitoxin (F) for 60 hours. PDEF, -actin and PSA were determined by immunoblot.
2164
CARDIAC GLYCOSIDES DECREASE PROSTATE SPECIFIC ANTIGEN EXPRESSION
attenuation of PDEF gene expression in vitro. We suggest that PSA interpretation in patients on cardiac glycosides could result in underestimating
prostate cancer risk and pathological stage. Thus, this phenomenon merits further study to determine whether it occurs in vivo.
REFERENCES 1. Stamey TA, Caldwell M, McNeal JE et al: The prostate specific antigen era in the United States is over for prostate cancer: what happened in the last 20 years? J Urol 2004; 172: 1297. 2. Catalona WJ, Southwick PC, Slawin KM et al: Comparison of percent free PSA, PSA density and age-specific PSA cutoffs for prostate cancer detection and staging. Urology 2000; 56: 255. 3. Schoner W and Scheiner-Bobis G: Endogenous and exogenous cardiac glycosides: their roles in hypertension, salt metabolism, and cell growth. Am J Physiol Cell Physiol 2007; 293: C509. 4. Yen JY, Huang WJ, Kan SF et al: Inhibitory effects of digitalis on the proliferation of androgen dependent and independent prostate cancer cells. J Urol 2001; 166: 1937. 5. Huang YT, Chueh SC, Teng CM et al: Investigation of ouabain-induced anticancer effect in human androgen-independent prostate cancer PC-3 cells. Biochem Pharmacol 2004; 67: 723.
put drug discovery: screening and analysis of compounds affecting genes overexpressed in cancer cells. Mol Cancer Ther 2002; 1: 1293. 10. Thelen P, Peter T, Hunermund A et al: Phytoestrogens from Belamcanda chinensis regulate the expression of steroid receptors and related cofactors in LNCaP prostate cancer cells. BJU Int 2007; 100: 199. 11. Schoner W: Endogenous cardiac glycosides, a new class of steroid hormones. Eur J Biochem 2002; 269: 2440.
19. Tomlins SA, Laxman B, Varambally S et al: Role of the TMPRSS2: EGR gene fusion in prostate cancer. Neoplasia 2008; 10: 177. 20. Gu X, Zerbini LF, Otu HH et al: Reduced PDEF expression increases invasion and expression of mesenchymal genes in prostate cancer cells. Cancer Res 2007; 67: 4219.
12. Tsui KH, Feng TH, Lin CM et al: Curcumin blocks the activating of androgen and interlukin-6 on prostate specific antigen expression in human prostatic carcinoma cells. J Androl 2008; 29: 661.
21. Magklara A, Brown TJ and Diamandis EP: Characterization of androgen receptor and nuclear receptor co-regulator expression in human breast cancer cell lines exhibiting differential regulation of kallikreins 2 and 3. Int J Cancer 2002; 100: 507.
13. Tsiu KH, Chang PL, Lin HT et al: Down-regulation of the prostate specific antigen promoter by p53 in human prostate cancer cells. J Urol 2004; 172: 2305.
22. Hsu T, Trojanowska M and Watson DK: Ets proteins in biological control and cancer. J Cell Biochem 2004; 91: 896.
6. Guh JH, Chueh SC and Teng CM: Effects of ouabain on tension response and [3H]noradrenaline release in human prostate. J Urol 2000; 163: 338.
14. Juang HH: Nitroprusside stimulates mitochondrial aconitase gene expression through the cyclic adenosine 3=,5=-monosphosphate signal transduction pathway in human prostate carcinoma cells. Prostate 2004; 61: 92.
7. Oettgen P, Finger E, Sun Z et al: PDEF, a novel prostate epithelium-specific ets transcription factor, interacts with the androgen receptor and activates prostate-specific antigen gene expression. J Biol Chem 2000; 275: 1216.
15. Tsiu KH, Wu L, Chang PL et al: Identifying the combination of the transcriptional regulatory sequences on prostate specific antigen and human glandular kallikrein genes. J Urol 2004; 172: 2029.
8. Sood AK, Saxena R, Groth J et al: Expression characteristics of prostate-derived Ets factor support a role in breast and prostate cancer progression. Hum Pathol 2007; 38: 1628.
16. Seth A and Waston DK: ETS transcription factors and their emerging roles in human cancer. Eur J Cancer 2005; 41: 2462.
9. Johnson PH, Walker RP, Jones SW et al: Multiplex gene expression analysis for high-through-
18. Cai C, Wang H, Xu Y et al: Reactivation of androgen receptor-regulated TMPRSS2: EGR gene expression in castration-resistant prostate cancer. Cancer Res 2009; 69: 6027.
17. Tomlins SA, Bjartell A, Chinnaiyan AM et al: ETS gene fusions in prostate cancer: from discovery to daily clinical practice. Eur Urol 2009; 56: 275.
23. Kometiani P, Liu L and Askari A: Digitalis-induced signaling by Na⫹/K⫹-ATPase in human breast cancer cells. Mol Pharmacol 2005; 67: 929. 24. Lopez-Lazaro M: Digitoxin as an anticancer agent with selectivity for cancer cells: possible mechanisms involved. Expert Opin Ther Targets 2007; 11: 1043. 25. Yang Q, Huang W, Jozwik C et al: Cardiac glycosides inhibit TNF-␣/NF-B signaling by blocking recruitment of TNF receptor-associated death domain to the TNF receptor. PNAS 2005; 102: 9631. 26. Katzung BG and Parmley WW: Cardiac glycosides and other drugs used in congestive heart failure. In: Basic and Clinical Pharmacology, 8th ed. Edited by BG Katzung. New York: Appleton & Lange 2000; pp 200 –218.