Bafilomycin Induces the p21-Mediated Growth Inhibition of Cancer Cells under Hypoxic Conditions by Expressing Hypoxia-Inducible Factor-1 □S

Bafilomycin A1, a macrolide antibiotic isolated from Streptomyces species, has been used as an inhibitor of vacuolar H ATPase (V-ATPase). Bafilomycin has been also evaluated as a potential anticancer agent because it inhibits cell proliferation and tumor growth. Although these anticancer effects of bafilomycin are considered to be attributable to the intracellular acidosis by V-ATPase inhibition, the exact mechanism remains unclear. In the present study, we tested the possibility that bafilomycin targets a tumor-promoting factor, hypoxia-inducible factor-1 (HIF-1 ). Bafilomycin A1 and its analog, concanamycin A, were found to up-regulate HIF-1 in eight human cancer cell-lines, and this effect is attributed to inhibited degradation of HIF-1 protein. Furthermore, the HIF-1 induction by bafilomycin was augmented by hypoxia, which caused a robust induction of p21 and cell cycle arrest in cancer cells. The cell cycle inhibition was shown only in cancer cells expressing both HIF-1 and p21. In HIF-1 ( / ) or HIF-1 ( / ) fibrosarcomas grafted in nude mice, bafilomycin showed the HIF-1 dependent anticancer effect. Based on these results, the exorbitant expression of HIF-1 is likely to contribute to the anticancer action of bafilomycin. Bafilomycin A1, a macrolide antibiotic isolated from Streptomyces species, has been used as an inhibitor of vacuolar H ATPase (V-ATPase). It binds to a pocket formed by V0 sector subunit c (ATP6V0C) of the V-ATPase complex and inhibits H translocation by preventing the rotation of the ATP6V0C multimer, which causes the accumulation of H in the cytoplasm (Bowman et al., 2004). Biologically, it has been found that bafilomycin induces cell growth inhibition (Ohkuma et al., 1993) and apoptosis (Kinoshita et al., 1996, Nakashima et al., 2003). In vivo, bafilomycin also inhibits the growth of xenografted tumors (Ohta et al., 1998) and thus could be evaluated as a potential anticancer agent. These biological effects of bafilomycin are considered to be attributable to the intracellular acidosis by V-ATPase inhibition. There is the question of whether the acidosis is only one mechanism underlying the anticancer activity of bafilomycin; generally, the agents that have in vivo anticancer activities inhibit some cancer-specific events, but the acidosis by bafilomycin occurs in normal cells as well. Therefore, we hypothesized that bafilomycin targets a cancer-specific molecule. Hypoxia-inducible factor-1 (HIF-1 ) is a basic-helix-loophelix protein of the PAS family (Wang and Semenza, 1995). It plays a key role in cellular adaptation to hypoxia by upregulating 60 or more genes essential for angiogenesis and cell survival. To date, the roles of HIF-1 in tumor progression have been extensively investigated, and its overexpression is frequently found in various human tumors (Zhong et al., 1999). HIF-1 levels in tumors are positively related with tumor hypervascularity, aggressiveness, and poor prognosis (Birner et al., 2000, Zagzag et al., 2000). Moreover, xenograft studies have disclosed that HIF-1 is essential for tumor This work was supported by a grant from the Korean Ministry of Health and Welfare Research Fund 2005 (A050479). J.-H.L. and J.-W.P. contributed equally to this work. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.106.028076. □S The online version of this article (available at http://molpharm. aspetjournals.org) contains supplemental material. ABBREVIATIONS: V-ATPase, vacuolar H ATPase; HIF-1 , hypoxia-inducible factor 1 ; BM, bafilomycin A1; pVHL, Von Hippel-Lindau protein; FIH, factor-inhibiting hypoxia-inducible factor; VEGF, vascular endothelial factor; PCR, polymerase chain reaction; RT-PCR, reverse transcription polymerase chain reaction; MEF, mouse embryonic fibroblast; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling; -gal, -galactoside; MG132, N-benzoyloxycarbonyl (Z)-Leu-Leu-leucinal. 0026-895X/06/7006-1856–1865$20.00 MOLECULAR PHARMACOLOGY Vol. 70, No. 6 Copyright © 2006 The American Society for Pharmacology and Experimental Therapeutics 28076/3151282 Mol Pharmacol 70:1856–1865, 2006 Printed in U.S.A. 1856 http://molpharm.aspetjournals.org/content/suppl/2006/08/29/mol.106.028076.DC1 Supplemental material to this article can be found at: at A PE T Jornals on Jauary 0, 2018 m oharm .aspeurnals.org D ow nladed from growth and angiogenesis (Ryan et al., 2000). Therefore, HIF-1 is indicated as an aggravating factor in cancer diseases. However, in stark contrast with the above, HIF-1 overexpression in hypoxic cancer cells has been suggested to inhibit cell division by inducing p21, p27, or p53 and to promote apoptosis by inducing p53, Nip3, Noxa, or HGTD-P (Goda et al., 2003; Bacon and Harris, 2004). Thus, HIF-1 seems to be a double-edged sword from the viewpoint of tumors subjected to hypoxia, and thus, its expression is probably controlled at optimal levels in growing tumors. In this respect, both HIF-1 suppression and overexpression could inhibit the growth of hypoxic tumors. In the present study, we address the mechanism underlying the anticancer action of bafilomycin. Bafilomycin A1 and its analog, concanamycin A, were found to induce HIF-1 in eight human cancer cell-lines. The HIF-1 induction by bafilomycin was augmented by hypoxia, which caused a robust induction of p21 and cell cycle arrest in cancer cells. In fibrosarcoma xenografts, bafilomycin showed the HIF-1 dependent anticancer effect. The exorbitant expression of HIF-1 is likely to contribute to the anticancer action of bafilomycin. Materials and Methods Materials. Bafilomycin A1 and MG132 were purchased from Alexis Biochemicals (Lausen, Switzerland), and concanamycin A, cycloheximide, propidium iodide, and other chemicals were from Sigma-Aldrich (St. Louis, MO). Culture media and fetal calf serum were purchased from GIBCO/BRL (Grand Island, NY). Anti-HIF-1 antiserum was generated in rabbits against a bacterially expressed fragment encompassing amino acids 418 to 698 of human HIF-1 , as described previously (Chun et al., 2001). p53, p27, mouse p21, and -actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Human p21, hemagglutinin, Von Hippel-Lindau protein (pVHL), and Flag antibodies were purchased from Cell Signaling (Beverly, MA), Roche (Basel, Switzerland), PharMingen (San Diego, CA), and Sigma-Aldrich, respectively. Cell Culture. PC3 (prostate), Caki-1 (kidney), SiHa (uterine cervix), and Hep3B (liver) cancer cell lines were obtained from American Type Culture Collection (Manassas, VA). Four lung cancer cell lines, HCC1171 (epithelial origin), HCC2108 (epithelial), HCC1195 (squamous), and HCC2279 (squamous), were obtained from the Korean Cell Line Bank (Seoul, Korea), and an HIF-1 -null mouse embryonic fibroblast (MEF) cell line was provided as described previously (Ryan et al., 2000). HCT116(p21 / ) and HCT116(p21 / ) were generous gifts from Dr. Deug Y Shin (Dankook University College of Medicine, Cheonan, Korea). Lung cancer cells were cultured in RPMI 1640 medium and other cells in Dulbecco’s modified Eagle’s medium, supplemented with 10% fetal calf serum, in a 5% CO2 humidified atmosphere at 37°C. O2 levels in the chamber were either 20% (normoxic) or 1% (hypoxic). Bafilomycin A1 or concanamycin A were administered to medium 5 min before normoxic or hypoxic incubation for 4 h. Immunoblotting and Immunoprecipitation. To quantify protein levels, total cell lysates were prepared as described previously (Chun et al., 2001). Proteins were separated on SDS-polyacrylamide gels and transferred to an Immobilon-P membrane (Millipore, Bedford, MA). Membranes were blocked with 5% nonfat milk and incubated overnight at 4°C with a primary antibody diluted 1:1000. Membranes were incubated with a horseradish peroxidase-conjugated secondary antibody (diluted 1:5000) for 2 h, and antigenantibody complexes were visualized using an Enhanced Chemiluminescence Plus kit (Amersham Biosciences Corporation, Piscataway, NJ). -Actin protein was used as an internal standard. For coimmunoprecipitation of HIF-1 and pVHL, cell lysates (150 g of protein) were incubated with 10 l of anti-HIF-1 antiserum or preimmune rabbit serum at 4°C for 4 h. The immune complex was further incubated with protein A/G-Sepharose beads (Amersham Biosciences) at 4°C for 4 h. The antigen-bead complexes obtained were Fig. 1. Bafilomycin and concanamycin induce HIF-1 . A, dose-response of HIF-1 expression. Eight cancer cell lines were incubated with BM or concanamycin A (CM) for 4 h and harvested. The concentrations of BM and CM used were 10, 5, 2, 1, and 0.2 nM. Protein levels of HIF-1 were evaluated by immunoblotting. B, time course of HIF-1 expression. PC3 cells were incubated with 10 nM BM for the times indicated above the figure. C, synergistic induction of HIF-1 by bafilomycin and hypoxia. PC3 cells were incubated under normoxic or hypoxic conditions in the absence or in the presence of 10 nM BM for 4 h. HIF-1 levels were evaluated by immunoblotting. HIF-1 -Dependent Anticancer Effect of Bafilomycin 1857 at A PE T Jornals on Jauary 0, 2018 m oharm .aspeurnals.org D ow nladed from washed extensively with lysis buffer. Immunocomplexes were eluted by boiling for 3 min in a sample buffer containing 2% SDS and 10 mM dithiothreitol, subjected to SDS-polyacrylamide gel electrophoresis, and then immunoblotted using anti-HIF-1 or anti-pVHL antibody. Semiquantitative RT-PCR. To quantify mRNA levels, we used a highly sensitive, semiquantitative RT-PCR method, as described previously (Yeo et al., 2003). Total RNAs were isolated from cultured cells or rat kidney using Trizol (GIBCO/BRL). One microgram of RNA was reverse-transcribed, and the cDNA obtained was amplified over 18 PCR cycles in a reaction mixture containing 5 Ci [ -P]dCTP and 250 nM concentration of each primer set. PCR products were electrophoresed on a 4% polyacrylamide gel, and dried gels were autoradiographed. Primers for human HIF-1 , vascular endothelial growth factor (VEGF), phosphoglycerate kinase 1, enolase 1, and -actin were constructed as described previously (Yeo et al., 2003). Construction and Assay of a p21 Reporter Gene. To make a p21 promoter-driven luciferase reporter gene, the DNA coding human p21 promoter region ( 966 to 2) containing a c-Myc binding site was cloned using PCR and then inserted into the KpnI and BglII sites of the pGL3 promoter plasmid (Promega, Madison, WI). PC3 cells were cotransfected with 0.5 g each of reporter gene and plasmid cytomegalovirus-gal using Lipofectamine (Invitrogen, Carlsbad, CA). After being allowed to stabilize for 48 h, cells were lysed to determine luciferase and -gal activities. Luciferase activities were analyzed using a Lumat LB9507 luminometer (Berthold Technologies, Bad Wildbad, Germany), and -gal assays were performed to normalize transfection efficiencies. Chromatin Immunoprecipitation. Cells were fixed with formaldehyde, and soluble chromatin samples were immunoprecipitated with anti-c-Myc or anti-HIF-1 at 4°C overnight (Jung et al., 2005). DNA isolated from immunoprecipitated material was amplified by semiquantitative PCR with [ -P]dCTP. The PCR primer sequences used were 5 -GATTTGTGGCTCACTTCGT-3 and 5 -GCTCCACAAGGAACTGACT-3 , which produced a 320-base pair fragment, including the c-Myc binding site of the p21 gene. PCR products were electrophoresed in a 4% polyacrylamide gel, and dried gels were autoradiographed. Cell Cycle Analysis. PC3 cells were plated in 10-cm culture dishes at concentrations determined to yield 70 to 80% confluence within 24 h. Cells were then incubated under normoxic or hypoxic Fig. 2. Bafilomycin stabilizes HIF-1 protein. A, HIF-1 mRNA levels. Total RNAs were extracted from PC3 cells treated with 10, 5, 2, 1, or 0.2 nM BM, and HIF-1 and -actin mRNAs were analyzed by semiquantitative RT-PCR. B, HIF-1 stability. PC3 cells were incubated in the absence ( BM) or presence ( BM) of 10 nM bafilomycin A1 for 4 h and then treated with 60 g/ml cycloheximide (CHX). After 5, 10, 30, 60, or 120 min, the cell lysates were analyzed by immunoblotting using anti-HIF-1 and anti-actin antibodies (top). Band intensities were quantified using a Microcomputer Imaging Device model 4 (MCID-M4) image analysis system and are plotted at the bottom. Halflives (t1/2) were calculated from the slopes of first-order decay curves. Each point represents the mean value of three separate experiments. C, bafilomycin dissociates pVHL from HIF-1 . PC3 cells were treated with BM and MG132 (10 M) for 6 h. HIF-1 was immunoprecipitated using anti-HIF-1 antibody or nonimmunized rabbit serum (Pre), and the coprecipitation of pVHL with HIF-1 was analyzed with anti-pVHL antibody. D, HIF-1 expression under alkaline conditions. A HEPES buffer (50 mM, pH 7.2–8.0) was added to buffer-free Dulbecco’s modified Eagle’s medium supplemented with 5 mM NaHCO3 and 10% fetal calf serum. Cells were incubated in the alkaline media for 30 min and treated with 10 nM BM for 4 h. MG, treatment of 10 M MG132 proteasome inhibitor. E, sublocalization of pVHL and HIF-1 . After PC3 cells were treated with BM (10, 2, and 1 nM) for 4 h, cells were harvested and fractionated to total (T), cytosolic (C), and nuclear (N) fractions. HIF-1 and pVHL proteins were analyzed in each fraction using anti-HIF-1 and antipVHL antibodies. 1858 Lim et al. at A PE T Jornals on Jauary 0, 2018 m oharm .aspeurnals.org D ow nladed from conditions in the absence or presence of bafilomycin A1 for 16 h. Cells were then harvested and resuspended in 200 l of phosphate-buffered saline and fixed in 75% ethanol for 30 min on ice. After washing with phosphate-buffered saline, cells were labeled with propidium iodide (0.05 mg/ml) in the presence of RNase A (0.5 mg/ml) and then incubated in the dark for 30 min. DNA contents were analyzed using a Becton Dickinson FACStar flow cytometer (BD Biosciences, San Jose, CA). Propidium iodide was excited using an argon laser at 488 nm and detected at 630 nm. Animals and Tumor Grafts. Male nude (BALB/cAnNCrj/ ) mice were purchased from Charles River Japan Inc. (Shin-Yokohama, Japan). Animals were housed in a specific pathogen-free room under controlled temperature and humidity. All animal procedures were performed according to the procedures described in the Seoul National University Laboratory Animal Maintenance Manual. Mice aged 7 to 8 weeks were injected subcutaneously in the flank with 5 10 viable cells of HIF-1 ( / ) or HIF-1 ( / ) MEF. Tumor volumes were measured with a caliper and calculated using the formula Volume a b/2, where a was the width at the widest point of the tumor, and b was the maximal width perpendicular to a. Tumor Histology and TUNEL Staining. Tumors were removed 4 days after injecting bafilomycin A1, fixed with formalin, and embedded in paraffin. Serial sections (6 m thick) were cut from each paraffin block and stained with hematoxylin and eosin (H&E). Necrosis was identified at a magnification of 40 and examined using a Sony XC-77 CCD camera and a Microcomputer Imaging Device Fig. 3. Bafilomycin induces p21 expression. A, two different roles of HIF-1 . B, PC3 cells were incubated under normoxic or hypoxic conditions for 16 h in the absence or in the presence of 10 nM of BM. Expression of genes related to angiogenesis and glycolysis. mRNA levels of VEGF, phosphoglycerate kinase 1 (PGK), enolase 1 (Eno), and -actin were determined by semiquantitative RT-PCR. C, normoxic control; H, hypoxia; BM, BM treatment; H BM, BM treatment under hypoxic conditions. C, synergistic induction of p21 by BM and hypoxia. In PC3 cells, p27 and -actin levels were analyzed by immunoblotting. Band intensities of p21 were quantified using the MCID-M4 image analysis system and are plotted on the bottom. Each point represents the mean and standard deviation of four separate experiments. , P 0.05 versus the normoxic control. D, bafilomycin-induced p21 expression in various cell lines. Four cancer cell lines were incubated with various concentrations (10, 5, 2, 1, or 0.2 nM) of BM for 4 h and then harvested for p21, p53, and -actin immunoblotting. E, bafilomycin activates the p21 promoter. PC3 cells were transfected with a p21 promoter reporter plasmid, and treated with BM for 16 h. Luciferase activities are quoted as relative values versus the control value and are plotted as the means S.D. of four experiments. , P 0.05 versus the control. F, bafilomycin-induced HIF-1 dissociates cMyc from p21 promoter. After PC3 cells were treated with 10 nM bafilomycin for 16 h, c-Myc or HIF-1 binding to p21 promoter was analyzed using chromatin immunoprecipitation. Proximal promoter DNAs of p21 were amplified by 32 cycles of PCR, electrophoresed, and autoradiographed. HIF-1 -Dependent Anticancer Effect of Bafilomycin 1859 at A PE T Jornals on Jauary 0, 2018 m oharm .aspeurnals.org D ow nladed from model 4 (MCID-M4) image analysis system (Sony, Tokyo, Japan). The extent of necrosis was determined in four different cross-sections per tumor by dividing the total cross-sectional necrotic area by the total cross-sectional area. An ApopTag in situ apoptosis detection kit of Oncor (Gaithersburg, MD) was used to evaluate apoptotic death. Serial sections were dewaxed, treated with proteinase K, incubated with equilibration buffer for 10 min, and then incubated with working-strength TdT enzyme solution at 37°C for 2 h. The reaction was terminated by incubation in working-strength stop/wash buffer for 30 min at 37°C. Sections were then incubated with antidigoxigenin peroxidase and then with diaminobenzidine and 0.01% H2O2 for 5 min at room temperature. Finally, they were lightly counterstained with H&E and examined under an optical microscope. Statistical Analysis. All data were analyzed using Microsoft Excel 2000 (Microsoft, Redmond, WA), and results are expressed as means and S.D. The Mann-Whitney U test (SPSS 10.0 for Windows; SPSS Inc., Chicago, IL) was used to compare reporter activities or protein levels in cultured cells. Tumor volumes of the control and the bafilomycin-treated groups were compared using analysis of variance followed by Duncan’s multiple range test. Differences were considered statistically significant at the P 0.05 level. All statistical tests were two-sided.