Portal de Periódicos da CAPES

Androgen Receptor Is a New Potential Therapeutic Target for the Treatment of Hepatocellular Carcinoma

2008; Elsevier BV; Volume: 135; Issue: 3 Linguagem: Inglês

10.1053/j.gastro.2008.05.046

1528-0012

Wen‐Lung Ma, Cheng–Lung Hsu, Ming–Heng Wu, Chun–Te Wu, Cheng‐Chia Wu, Jiann–Jyh Lai, Yuh–Shan Jou, Chun‐Wei Chen, Shuyuan Yeh, Chawnshang Chang,

Pharmacogenetics and Drug Metabolism

Background & Aims: Androgen effects on hepatocellular carcinoma (HCC) remain controversial and androgen ablation therapy to treat HCC also leads to inconsistent results. Here we examine androgen receptor (AR) roles in hepatocarcinogenesis using mice lacking AR in hepatocytes. Methods: By using the Cre-Lox conditional knockout mice model injected with carcinogen, we examined the AR roles in hepatocarcinogenesis. We also tested the possible roles of AR in cellular oxidative stress and DNA damage sensing/repairing systems. By using AR degrading compound, ASC-J9, or AR-small interference RNA, we also examined the therapeutic potentials of targeting AR in HCC. Results: We found AR expression was increased in human HCC compared with normal livers. We also found mice lacking hepatic AR developed later and less HCC than their wild-type littermates with comparable serum testosterone in both male and female mice. Addition of functional AR in human HCC cells also resulted in the promotion of cell growth in the absence or presence of 5α-dihydrotestosterone. Mechanistic dissection suggests that AR may promote hepatocarcinogenesis via increased cellular oxidative stress and DNA damage, as well as suppression of p53-mediated DNA damage sensing/repairing system and cell apoptosis. Targeting AR directly via either AR-small interference RNA or ASC-J9 resulted in suppression of HCC in both ex vivo cell lines and in vivo mice models. Conclusions: Our data point to AR, but not androgens, as a potential new and better therapeutic target for the battle of HCC. Background & Aims: Androgen effects on hepatocellular carcinoma (HCC) remain controversial and androgen ablation therapy to treat HCC also leads to inconsistent results. Here we examine androgen receptor (AR) roles in hepatocarcinogenesis using mice lacking AR in hepatocytes. Methods: By using the Cre-Lox conditional knockout mice model injected with carcinogen, we examined the AR roles in hepatocarcinogenesis. We also tested the possible roles of AR in cellular oxidative stress and DNA damage sensing/repairing systems. By using AR degrading compound, ASC-J9, or AR-small interference RNA, we also examined the therapeutic potentials of targeting AR in HCC. Results: We found AR expression was increased in human HCC compared with normal livers. We also found mice lacking hepatic AR developed later and less HCC than their wild-type littermates with comparable serum testosterone in both male and female mice. Addition of functional AR in human HCC cells also resulted in the promotion of cell growth in the absence or presence of 5α-dihydrotestosterone. Mechanistic dissection suggests that AR may promote hepatocarcinogenesis via increased cellular oxidative stress and DNA damage, as well as suppression of p53-mediated DNA damage sensing/repairing system and cell apoptosis. Targeting AR directly via either AR-small interference RNA or ASC-J9 resulted in suppression of HCC in both ex vivo cell lines and in vivo mice models. Conclusions: Our data point to AR, but not androgens, as a potential new and better therapeutic target for the battle of HCC. See editorial on page 738. See editorial on page 738. Although viral infection and/or environmental carcinogens may lead to the development of hepatocellular carcinoma (HCC), the etiology of this liver cancer remains unclear. Early studies suggested that androgens might contribute to the gender difference of HCC incidence and serum testosterone may have a positive linkage to the development of HCC.1Tejura S. Rodgers G.R. Dunion M.H. et al.Sex-steroid receptors in the diethylnitrosamine model of hepatocarcinogenesis: modifications by gonadal ablation and steroid replacement therapy.J Mol Endocrinol. 1989; 3: 229-237Crossref PubMed Scopus (18) Google Scholar However, clinical trials with androgen targeting via androgen ablation therapy yield inconsistent and disappointing outcomes.2Groupe d'Etude et de Traitement du Carcinome He’patocellulaireRandomized trial of leuprorelin and flutamide in male patients with hepatocellular carcinoma treated with tamoxifen.Hepatology. 2004; 40: 1361-1369Crossref PubMed Scopus (70) Google Scholar Androgen effects are mediated mainly through the androgen receptor (AR).3Chang C.S. Kokontis J. Liao S.T. Molecular cloning of human and rat complementary DNA encoding androgen receptors.Science. 1988; 240: 324-326Crossref PubMed Scopus (720) Google Scholar Androgen/AR signals may modulate many biological events via interaction with various AR coregulators.4Heinlein C.A. Chang C. Androgen receptor (AR) coregulators: an overview.Endocr Rev. 2002; 23: 175-200Crossref PubMed Scopus (631) Google Scholar The biological function of androgen/AR in liver and their detailed consequences, however, remain unclear. We generated a conditional knockout AR mouse lacking only the hepatic AR (L-AR–/y) via mating floxed-AR mice with albumin promoter-driven Cre-recombinase (Alb-Cre) transgenic mice.5Lin H.-Y.I.-C.Y. Wang R.-S. Chen Y.-T. et al.Increased hepatic steatosis and insulin resistance in mice lacking hepatic androgen receptor.Hepatology. 2008; 47: 1924-1935Crossref PubMed Scopus (141) Google Scholar Results from these mice in which HCC was induced via injection of N'-N'-diethylnitrosamine (DEN) suggest that the AR, rather than androgens, may play a more dominant role in HCC development. Ten sets of liver tumors (<3 cm) and corresponding normal liver tissues for immunohistochemical staining were obtained from 10 male human patients who received routine liver cancer surgery after informed consent. All of the animal experiments followed the Guidance of the Care and Use of Laboratory Animals of the National Institutes of Health, with approval from the University of Rochester, Department of Laboratory Animal Medicine. The strategy to generate flox-AR gene-targeting mice has been described previously.6Yeh S. Tsai M.Y. Xu Q. et al.Generation and characterization of androgen receptor knockout (ARKO) mice: an in vivo model for the study of androgen functions in selective tissues.Proc Natl Acad Sci U S A. 2002; 99: 13498-13503Crossref PubMed Scopus (498) Google Scholar Briefly, we mated male β-actin promoter-driven Cre or Alb-Cre5Lin H.-Y.I.-C.Y. Wang R.-S. Chen Y.-T. et al.Increased hepatic steatosis and insulin resistance in mice lacking hepatic androgen receptor.Hepatology. 2008; 47: 1924-1935Crossref PubMed Scopus (141) Google Scholar (Cre recombinase under control of albumin promoter; Jackson Laboratories, Bar Harbor, ME; B6.Cg-Tg[Alb-Cre]21Mgn/J) mice with flox-AR/AR heterozygous female mice to produce T/L-AR–/y (T, total knockout in whole body; L, liver-specific knockout) male and T/L-AR–/+ heterozygous female mice. Another mating using T/L-AR–/+ female mice with ARflox/y/L-AR–/y also generated T/L-AR–/–. We genotyped 21-day-old pups from tail snips by polymerase chain reaction, as described previously.6Yeh S. Tsai M.Y. Xu Q. et al.Generation and characterization of androgen receptor knockout (ARKO) mice: an in vivo model for the study of androgen functions in selective tissues.Proc Natl Acad Sci U S A. 2002; 99: 13498-13503Crossref PubMed Scopus (498) Google Scholar We induced HCC in the liver of 12-day-old pups with intraperitoneal injection of a single dose of the HCC initiator, DEN (20 mg/kg/mouse; Sigma-Aldrich, St Louis, MO).1Tejura S. Rodgers G.R. Dunion M.H. et al.Sex-steroid receptors in the diethylnitrosamine model of hepatocarcinogenesis: modifications by gonadal ablation and steroid replacement therapy.J Mol Endocrinol. 1989; 3: 229-237Crossref PubMed Scopus (18) Google Scholar After genotyping the pups we divided into 7 different groups. The groups were as follows: (1) AR+/y, (2) T-AR–/y, (3) L-AR–/y, (4) AR+/+, (5) T-AR–/–, (6) L-AR–/–, and (7) AR+/y-untreated with solvent injection only. Several mice from each group were killed at 20, 24, 28, 32, 36, and 40 weeks after DEN injection. The nude mice used for xenograft experiments were 10-week-old male nude mice (Crl, CD1-Foxn1nu Origin; Charles River, Wilmington, MA) and ASC-J9 was provided by AndroScience Corporation (San Diego, CA). We killed mice at the indicated time points, drew 1 mL of blood by cardiocentesis, and immediately assayed for serum testosterone level using the Coat-A-Count Total Testosterone radioimmunoassay (Diagnostic Automation, Inc, Calabasas, CA). We flash-froze fresh tissues in liquid nitrogen for preservation at –80°C for gene expression assay. We subjected the hepatic major lobe to 10% neutralized buffered formalin (Sigma) for histologic analysis. We fixed the tissues in 10% buffered formalin (Sigma) and embedded them in paraffin. For general histologic inspection, we treated tissue sections with H&E, and then used an ABC kit (Vector Laboratories, Burlingame, CA) to visualize AR, p53, and 8-oxodeoxyguanosine immunostaining by specific antibodies against mouse AR (C-19; Santa Cruz, Biotechnology, Santa Cruz, CA), human AR (441; Dako, Carpinteria, CA), p53 (Ab-3; Calbiochem, Gibbstown, NJ), and 8-oxodeoxyguanosine (sc-12075; Santa Cruz). We performed the terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling (TUNEL) staining assay (Calbiochem) as previously described.7Yeh S. Hu Y.C. Wang P.H. et al.Abnormal mammary gland development and growth retardation in female mice and MCF7 breast cancer cells lacking androgen receptor.J Exp Med. 2003; 198: 1899-1908Crossref PubMed Scopus (88) Google Scholar We injected 5'-bromo-2'-deoxyuridine (BrdU; Sigma) for 4 consecutive days into 55-week-old DEN-induced mice and stained tissue sections with BrdU-specific antibody (Zymed, South San Francisco, CA) as previously described.7Yeh S. Hu Y.C. Wang P.H. et al.Abnormal mammary gland development and growth retardation in female mice and MCF7 breast cancer cells lacking androgen receptor.J Exp Med. 2003; 198: 1899-1908Crossref PubMed Scopus (88) Google Scholar We analyzed the results using the chi-square test and the Fisher exact test for cancer incidence using Sigmaplot software, using an unpaired t test for other experiments, using SD as experimental variation, and considering P values less than .05 to be statistically significant. For other methods, please see the supplementary Materials and Methods section (see supplementary materials online at www.gastrojournal.org). We first showed the expression of AR in livers from HCC patients. As shown in Figure 1A, AR expression was highly expressed in a dysplastic liver nodule. Among 10 HCC patients examined, stronger AR expression was found in tumor than in surrounding nontumor in 7 patients (Figure 1B; upper panel). Some of the AR was stained in the tumor border as shown in Figure 1B (lower panel). Another patient had AR staining in the nontumor part only. We generated AR knockout mice that either lacked hepatic AR (L-AR–/y), and their littermates (L-AR–/+), or lacked AR in the whole body (T-AR–/y), and their littermates (T-AR–/+), via mating loxP site-AR female transgene (ARflox/flox)6Yeh S. Tsai M.Y. Xu Q. et al.Generation and characterization of androgen receptor knockout (ARKO) mice: an in vivo model for the study of androgen functions in selective tissues.Proc Natl Acad Sci U S A. 2002; 99: 13498-13503Crossref PubMed Scopus (498) Google Scholar mice with Alb-Cre5Lin H.-Y.I.-C.Y. Wang R.-S. Chen Y.-T. et al.Increased hepatic steatosis and insulin resistance in mice lacking hepatic androgen receptor.Hepatology. 2008; 47: 1924-1935Crossref PubMed Scopus (141) Google Scholar or β-actin promoter-driven Cre6Yeh S. Tsai M.Y. Xu Q. et al.Generation and characterization of androgen receptor knockout (ARKO) mice: an in vivo model for the study of androgen functions in selective tissues.Proc Natl Acad Sci U S A. 2002; 99: 13498-13503Crossref PubMed Scopus (498) Google Scholar bearing male transgene mice (supplementary Figure 1; see supplementary material online at www.gastrojournal.org). We further confirmed AR expression in nuclei of HCC foci in AR+/y mice, but not in L-AR–/y mice, by immunohistochemical staining of AR (Figure 1C and D). To develop HCC in these mice, we used a single injection of the DEN carcinogen as described in the Materials and Methods section and separated them into 7 groups: (1) AR+/y, (2) L-AR–/y, (3) T-AR–/y, (4) AR+/+, (5) L-AR–/–, (6) T-AR–/–, and (7) untreated male AR+/y mice. We found that the serum testosterone levels remained comparable between 36-week DEN-induced L-AR–/y and AR+/y, and between L-AR–/– and AR+/+, even though the male mice had much higher serum testosterone levels than the female mice. Notably, unlike L-AR–/y, T-AR–/y mice had much lower serum testosterone levels when compared with littermates AR+/y (Figure 1E). We found none of the untreated mice (group 7) developed HCC by 40 weeks of age (data not shown). In contrast, the other 6 groups developed HCC with different incidences (Figure 2A). HCC developed in all the DEN-induced male AR+/y mice examined at 28, 32, 36, and 40 weeks of age, whereas only 25%–60% of DEN-treated female wild-type (AR+/+) mice examined at 28 to approximately 40 weeks of age developed HCC, confirming the gender difference in HCC incidence.1Tejura S. Rodgers G.R. Dunion M.H. et al.Sex-steroid receptors in the diethylnitrosamine model of hepatocarcinogenesis: modifications by gonadal ablation and steroid replacement therapy.J Mol Endocrinol. 1989; 3: 229-237Crossref PubMed Scopus (18) Google Scholar, 8Ostrowski J.L. Ingleton P.M. Underwood J.C. et al.Increased hepatic androgen receptor expression in female rats during diethylnitrosamine liver carcinogenesis A possible correlation with liver tumor development.Gastroenterology. 1988; 94: 1193-1200Abstract PubMed Google Scholar In contrast, L-AR–/y mice developed less HCC as compared with their wild-type littermates, even though they had comparable serum testosterone levels. Similar results also occurred in female mice, showing that L-AR–/– mice developed less HCC with comparable serum testosterone than their wild-type littermates, suggesting that AR, rather than androgens, is crucial for the development of HCC in both male and female mice. Interestingly, HCC incidence in male L-AR–/y and T-AR–/y mice is still higher than in female L-AR–/– and T-AR–/– mice, suggesting that factors other than AR also might contribute to the gender differences in HCC incidence. Because of the multiple origin nature of DEN-induced HCC, we also counted the numbers of tumor foci and found a reduced number of HCC foci in L-AR–/y and T-AR–/y mice compared with AR+/y, with a ratio of AR+/y:L-AR–/y (or AR+/y:T-AR–/y) = 20:6 (Figure 2B). We also weighed the individual DEN-induced HCC livers and found the ratio of liver weight to whole body weight was reduced in L-AR–/y and T-AR–/y mice as compared with their littermate AR+/y mice, suggesting that loss of hepatic AR might result in reduction of HCC tumor mass (Figure 2C). In contrast, the liver weight in non–DEN-injected L-AR–/y or T-AR–/y mice was similar to their littermate AR+/y mice (supplementary Figure 2; see supplementary material online at www.gastrojournal.org), suggesting that loss of hepatic AR has little influence on the steady state of normal liver growth in mice without HCC development. Having shown that loss of hepatic AR resulted in a reduction of HCC incidence, we determined if loss of hepatic AR also might influence HCC progression that could be correlated with lower proliferation and higher apoptosis rates. We first assessed cell proliferation via intraperitoneal administration of BrdU in mice for 4 consecutive days. The mice were killed and liver tumors were dissected, embedded, sectioned, and stained with anti-BrdU antibody. We counted positive stains for proliferating cells and showed the reduction of BrdU (+) staining in both L-AR–/y and T-AR–/y mice as compared with AR+/y mice (Figure 2D). We also used the TUNEL apoptosis assay to measure apoptosis, and found more positive TUNEL staining in L-AR–/y and T-AR–/y as compared with AR+/y mice (Figure 2D), suggesting that loss of hepatic AR might increase cell death in the liver tumor during HCC progression. We then used primary cells isolated from 55-week-old DEN-induced AR+/y mice to examine the androgen 5α-dihydrotestosterone (DHT) effects on cell growth. The results from 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay showed that the cell numbers increased in a dose-dependent manner on DHT treatment (Figure 2E). Together, using various growth and apoptosis assays, our results (Figure 2D and E) showed that loss of hepatic AR might lead to the suppression of HCC progression. To further strengthen our findings from mice studies showing that loss of hepatic AR results in the suppression of HCC growth, we applied human HCC cell lines to examine the AR effects on HCC cell growth (supplementary Figure 3; see supplementary material online at www.gastrojournal.org). By using a cell-counting assay we showed that DHT had little effect on SKpar (parental transfectant) cell growth (Figure 3A, SKpar-EtOH vs SKpar-DHT). In contrast, SKAR3 (stable AR transfectant) increased cell growth (Figure 3A, SKpar-EtOH vs SKAR3-EtOH) in the absence of DHT and addition of 10 nmol/L DHT further increased cell growth (Figure 3A, SKAR3-EtOH vs SKAR3-DHT). These results suggest that both non–androgen-mediated AR and androgen-mediated AR signals might influence HCC cell growth. The addition of functional AR in SKpar cells also resulted in the decreased cell apoptosis in the absence or presence of DHT (Figure 3B), suggesting that AR, rather than androgen, may play more important roles in hepatic cell apoptosis. This conclusion is supported further with the results from the anchorage-independent cell growth assay. By using a soft agar colony formation assay, we found that SKAR3 cells, but not SKpar cells, were able to grow in an anchorage-independent environment in the absence of androgen, suggesting increased AR expression via transfected functional AR results in anchorage-independent cell growth (Figure 3C). The addition of 10 nmol/L DHT showed little influence on the AR-promoted anchorage-independent cell growth. Together, our results in Figure 3 suggest that the AR, rather than androgen, may play a more important role in human HCC cells growth. Reactive oxygen species (ROS) has been linked to hepatocarcinogenesis during chronic inflammatory liver injury, such as hepatitis and cirrhosis.9Pikarsky E. Porat R.M. Stein I. et al.NF-kappaB functions as a tumour promoter in inflammation-associated cancer.Nature. 2004; 431: 461-466Crossref PubMed Scopus (2097) Google Scholar Early reports documented the linkage between DEN-induced HCC in mice with innate immune response and the related cellular oxidative stress.10Maeda S. Kamata H. Luo J.L. et al.IKKbeta couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis.Cell. 2005; 121: 977-990Abstract Full Text Full Text PDF PubMed Scopus (946) Google Scholar We first evaluated the cellular oxidative stress levels via measuring the carbonylated groups,11Lim G.P. Chu T. Yang F. et al.The curry spice curcumin reduces oxidative damage and amyloid pathology in an Alzheimer transgenic mouse.J Neurosci. 2001; 21: 8370-8377Crossref PubMed Google Scholar the oxidized amino acid side chain of protein (Figure 4A,left panels). We found that cellular ROS levels in the liver tumor of 36-week-old L-AR–/y mice were reduced to 30% as compared with those in DEN-induced AR+/y mice (Figure 4A, right panel). To further confirm the effect of androgen/AR signals on the cellular ROS level, we used AR stably transfected SKAR3 cells to examine cellular oxidative stress. We measured the cellular ROS level in SKpar and SKAR3 cells treated with 200 mmol/L H2O2 in the absence or presence of DHT. The results showed that the ROS level in SKAR3 cells is increased upon H2O2 treatment and further enhanced in the presence of 1 nmol/L DHT as compared with those in SKpar cells (Figure 4B). To further dissect how androgen/AR signals may regulate cellular ROS, we examined several key factors that have been linked to ROS and found messenger RNA (mRNA) expression of thioreducin-2 and superoxide dismutase 2 were decreased after adding 10 nmol/L DHT in SKAR3 cells treated with H2O2 (supplementary Figure 4; see supplementary material online at www.gastrojournal.org). In contrast, because there is little functional AR available in SKpar cells, addition of 10 nmol/L DHT failed to suppress the H2O2-induced thioreducin-2 and superoxide dismutase 2 mRNA expression (supplementary Figure 4; see supplementary material online at www.gastrojournal.org). Because chronic inflammation–induced oxidative stress might result in the breakage or damage of chromosomal DNA, we examined the DNA damage status in the mice liver tumor. By staining for the DNA damage marker 8-oxodeoxyguanosine,12Ichinose T. Nobuyuki S. Takano H. et al.Liver carcinogenesis and formation of 8-hydroxy-deoxyguanosine in C3H/HeN mice by oxidized dietary oils containing carcinogenic dicarbonyl compounds.Food Chem Toxicol. 2004; 42: 1795-1803Crossref PubMed Scopus (25) Google Scholar we found that the positive signal was higher in the liver tumors of AR+/y compared with those in L-AR–/y mice at 36 weeks of DEN induction (Figure 4C). These results suggest that reduced cellular oxidative stress in L-AR–/y mice may suppress the DNA damage, which then may lead to fewer gene mutations and delayed HCC development. In the normal liver condition, the increased DNA damage via cellular oxidative stress13Hussain S.P. Schwank J. Staib F. et al.TP53 mutations and hepatocellular carcinoma: insights into the etiology and pathogenesis of liver cancer.Oncogene. 2007; 26: 2166-2176Crossref PubMed Scopus (433) Google Scholar may result in the increase of the p53-mediated DNA damage sensing and repairing system. The p53 activation can suppress the function of the anti-apoptotic molecule, Bcl-2; therefore, triggering an intrinsic cascade for apoptosis.13Hussain S.P. Schwank J. Staib F. et al.TP53 mutations and hepatocellular carcinoma: insights into the etiology and pathogenesis of liver cancer.Oncogene. 2007; 26: 2166-2176Crossref PubMed Scopus (433) Google Scholar Interestingly, we found that loss of hepatic AR not only reduced DNA damage, but also enhanced the p53 expression in both normal and liver tumor of L-AR–/y mice (Figure 5A and B). The p53 down-stream target gene, p21, was up-regulated in L-AR–/y as well (Figure 5B). Comparative results can be observed consistently in human HCC cells (Supplementary Figure 5, Supplementary Figure 6; see supplementary material online at www.gastrojournal.org). Furthermore, enhanced p53 expression also might promote the DNA sensing and repairing system. For example, the expressions of the p53 target gene, Gadd45α14Gramantieri L. Chieco P. Giovannini C. et al.GADD45-alpha expression in cirrhosis and hepatocellular carcinoma: relationship with DNA repair and proliferation.Hum Pathol. 2005; 36: 1154-1162Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar and β,15Qiu W. David D. Zhou B. et al.Down-regulation of growth arrest DNA damage-inducible gene 45beta expression is associated with human hepatocellular carcinoma.Am J Pathol. 2003; 162: 1961-1974Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar DNA damage-repairing executive genes, were increased in liver tumors of L-AR–/y compared with AR–/y mice (Figure 5C). We also found that Gadd45 can be regulated by AR at the transcriptional level (supplementary Figure 5; see supplementary material online at www.gastrojournal.org). The increased DNA damage sensing and repairing system then might result in the reduced DNA damage seen in liver tumors of L-AR–/y mice. Together, results from Figure 5 suggest that loss of hepatic AR may suppress hepatocarcinogenesis via 2 pathways: (1) suppression of ROS-induced cellular oxidative stress and DNA damage, and (2) increased p53 expression that results in the better DNA sensing and repairing system as well as promoting cell apoptosis. Based on the earlier-described findings showing AR might play a pivotal role for HCC progression, we used both ex vivo cells and the in vivo mice model to investigate whether AR can be a therapeutic target for the treatment of HCC. We used 2 therapeutic approaches: (1) transfection with AR-small interference RNA (AR-siRNA), and (2) treatment with the anti-AR compound 5-hydroxy-1,7-bis(3,4-dimethoxyphenyl)-1,4,6-heptatrien-3-one (ASC-J9).16Yang Z. Chang Y.J. Yu I.C. et al.ASC-J9 ameliorates spinal and bulbar muscular atrophy phenotype via degradation of androgen receptor.Nat Med. 2007; 13: 348-353Crossref PubMed Scopus (132) Google Scholar Targeting AR with AR-siRNA. We first established stable sublines of SKAR3 cells transfected with a retrovirus-based vector that expresses AR-siRNA, which effectively knocked down the AR in MCF-7 cells.7Yeh S. Hu Y.C. Wang P.H. et al.Abnormal mammary gland development and growth retardation in female mice and MCF7 breast cancer cells lacking androgen receptor.J Exp Med. 2003; 198: 1899-1908Crossref PubMed Scopus (88) Google Scholar We substantially knocked down the AR expression in SKAR3 cells stably transfected with AR-siRNA (designated SKAR3-si1, 2, or 3) (Figure 6A). In contrast, AR was expressed normally in SKAR3 cells stably transfected with control scramble RNA (designated SKAR3-sc). We then investigated the effect of the AR-siRNA on the AR-mediated transactivation and AR-mediated cell growth in the stable sublines. We treated each stable subline with 1 nmol/L DHT and assessed transactivation by ARE(4)-luciferase promoter assay. We found that addition of 1 nmol/L DHT could induce substantial AR transactivation in SKAR3-sc, but not SKAR3-si1 cells (Figure 6B). By using the MTT growth assay, we also found that knockdown of AR expression via AR-siRNA resulted in the suppression of DHT-induced cell growth (Figure 6C). Targeting AR by treatment with the anti-AR compound ASC-J9. The recently developed anti-AR compound ASC-J9 targets AR via dissociating AR from its coregulators, leading to selective degradation of the AR protein.16Yang Z. Chang Y.J. Yu I.C. et al.ASC-J9 ameliorates spinal and bulbar muscular atrophy phenotype via degradation of androgen receptor.Nat Med. 2007; 13: 348-353Crossref PubMed Scopus (132) Google Scholar We examined the effects of ASC-J9 on HCC progression in both human HCC cells and in the in vivo mice model, and found that addition of 5 μmol/L ASC-J9 to the SKAR3 and SKAR7 cells resulted in the suppression of cell growth in the presence of 10 nmol/L DHT (Figure 6D). Furthermore, addition of 5 μmol/L ASC-J9 also resulted in increased cell apoptosis in the absence or presence of 10 nmol/L DHT (Figure 6E). We further confirmed this suppression effect on the HCC cell growth when we replaced human SKAR3 or SKAR7 cells with primary tumor cells isolated from AR+/y mice livers. We found that addition of 5 μmol/L ASC-J9 suppressed the primary tumor cell growth in the absence or presence of 10 nmol/L DHT (Figure 6F). Furthermore, in the mice inoculated with cells isolated from primary liver tumor of AR+/y mice, we found intraperitoneal injection of ASC-J9 (50 mg/kg/mice twice per week) resulted in the suppression of tumor growth during the course of 17 weeks of treatment (Figure 6G). Together, results from Figure 6 suggest that directly targeting the AR either via AR-siRNA or ASC-J9 could suppress HCC progression. The AR is expressed in the normal liver tissue from both male and female mammalians, but their expression and activation were reported to be increased in the tumor tissue and in the surrounding liver tissue of individuals with HCC.17Nagasue N. Ito A. Yukaya H. et al.Androgen receptors in hepatocellular carcinoma and surrounding parenchyma.Gastroenterology. 1985; 89: 643-647Abstract PubMed Google Scholar Moreover, the expression and activation of AR was reported to be greatly increased in the liver tissue of male and female rodents during chemical-induced liver carcinogenesis.18Eagon P.K. Elm M.S. Epley M.J. et al.Sex steroid metabolism and receptor status in hepatic hyperplasia and cancer in rats.Gastroenterology. 1996; 110: 1199-1207Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar In hepatitis B virus (HBV)-related HCC, pathways involving androgen-AR signaling, such as serum testosterone concentration, or length of AR CAG length (<23 repeats), may affect the risk of HBV-related HCC among male patients.19Yeh S.H. Chiu C.M. Chen C.L. et al.Somatic mutations at the trinucleotide repeats of androgen receptor gene in male hepatocellular carcinoma.Int J Cancer. 2007; 120: 1610-1617Crossref PubMed Scopus (20) Google Scholar The most important conclusion from these in vivo animal studies with mice lacking hepatic AR and ex vivo studies with human HCC cells transfected with either AR-siRNA or functional AR is a clear demonstration that AR might play pivotal roles for HCC development and therefore AR, rather than androgens, might represent a new target for treatment of HCC. The similar findings of the effect of AR on hepatocarcinogenesis also were observed in HBV transgene mice with a subminimum dosage of DEN injection (unpublished results). This conclusion in opposition to the conventional concept using androgen ablation therapy that only targets androgens is based on the following evidence. First, both male and female mice lacking hepatic AR have less HCC incidence with similar serum testosterone compared with the wild-type littermate mice (Figure 1, Figure 2). Second, stably transfected functional AR increased cell growth in the absence of DHT (Figure 3). Third, SKAR3 cells, but not SKpar cells, were able to grow in the absence of androgen in an anchorage-independent environment, and addition of 10 nmol/L DHT resulted in little influence of the AR-pro