Bone Morphogenetic Protein-7 Inhibits Telomerase Activity, Telomere Maintenance, and Cervical Tumor Growth

Research Article

Lucy Cassar, He Li, Alexander Ruvantha Pinto, Craig Nicholls, Sharyn Bayne, and Jun-Ping Liu

Department of Immunology, Central Eastern Clinical School, Monash University, Melbourne, Australia

Abstract crucial to cancer cell proliferation toward immortalization. Telomere maintenance is critical in tumor cell immortaliza- Containing human telomerase reverse transcriptase (hTERT), tion. Here, we report that the cytokine bone morphogenetic human RNA template component, and its binding protein dyskerin, protein-7 (BMP7) inhibits telomerase activity that is required telomerase catalyses telomeric DNA synthesis and protects for telomere maintenance in cervical cancer cells. Application telomere ends (21–23). As the catalyst of telomerase complex, of human recombinant BMP7 triggers a repression of the hTERT is rate limiting and expressed specifically in most cancers to human telomerase reverse transcriptase (hTERT) gene, short- maintain short telomeres. Ectopic expression of hTERT lengthens ening of telomeres, and hTERT repression–dependent cervical telomeric DNA, stabilizes telomeres, and mediates cell immortal- cancer cell death. Continuous treatment of mouse xenograft ization (24, 25). Inhibition of hTERT by gene silencing, dominant- tumors with BMP7, or silencing the hTERT gene, results in negative gene expression, or substrate competitive inhibitors sustained inhibition of telomerase activity, shortening of accelerates telomere shortening and thereby triggers cancer cell senescence and apoptosis in cell cultures and tumors (24, 25). telomeres, and tumor growth arrest. Overexpression of hTERT lengthens telomeres and blocks BMP7-induced tumor growth To date, little is known of how hTERT is regulated in cells by arrest. Thus, BMP7 negatively regulates telomere maintenance, extracellular cues. Considerable evidence indicates that when inducing cervical tumor growth arrest by a mechanism of normal cells differentiate in most tissues and organs, hTERT gene expression becomes repressed, conferring a limited length of inducing hTERTgene repression. [Cancer Res 2008;68(22):9157–66] telomeres and therefore proliferative potential on cells (26, 27). Repression of the hTERT gene occurs at the gene transcription level Introduction under a concerted regulation by a number of transcription factors Bone morphogenetic proteins (BMPs) operate through autocrine and repressors (28–30). We hypothesize that extracellular factors and paracrine mechanisms to regulate cell proliferation, differen- regulating cell differentiation exert an effect on hTERT gene tiation, and apoptosis during development (1, 2). Similar to other expression, telomerase activity, and telomere maintenance through transforming growth factor-h (TGF-h) family members, BMPs bind intracellular signaling. Studies from our laboratory and others cell membrane type I and II receptors of serine/threonine kinases show that TGF-h can induce an hTERT gene repression (30–35). In eliciting intracellular signaling via Smad1, Smad5, Smad8 and the present study, we have examined effects of the BMP family on Smad9 proteins (3, 4), and the Lim kinase (5, 6), and mitogen- telomerase activity and telomere maintenance in cervical cancer activated protein kinase pathways (7, 8). BMPs are present cells. We show that BMP7 induces sustained telomerase inhibition in cancers including breast, prostate, and colon cancers (9–11) and telomere shortening in vitro and in vivo, resulting in cancer cell where BMPs are implicated in regulating cancer cell proliferation apoptosis by a mechanism involving the repression of the hTERT (8, 12–14). BMP7 (osteogenic protein-1) is a potent inducer of cell gene and telomerase activity. differentiation required for vertebrate development, deletion of which is postnatal-lethal with multiple defects in differentiated Materials and Methods tissues (15). Recent studies show that BMP7 maintains epithelial and endothelial phenotypes against epithelial-mesenchymal tran- Cytokines, gene expression plasmids, and antibodies. Human sition (16, 17). BMP7 inhibits cancer cell proliferation causing recombinant BMP 2, 4, 5, 6, and 7 were from R & D systems. Plasmid pEGFP-C1, pEGFP-C1-hTERT, and pEGFP-C1-hTERT shRNA were produced apoptosis of myeloma cells (12) and prostate cancer cells (13, 18), in this laboratory. For hTERT shRNA, the oligonucleotide (5¶-AATT- and inhibits breast (19) and prostate (16) cancer metastasis. CAAAAAGGGTCTTTCTACCAGAGGTGCTTCTCTTGAAATCATCTCTGG- h However, apart from BMP7 antagonizing TGF- –mediated proin- TAGCAAGACC-3) was annealed and cloned to the EcoR1 site downstream vasive and protumorigenic effects (16, 17, 19), the exact of the U6 promoter. All plasmids were verified by DNA sequencing. The mechanisms in gene expression underlying these inhibitory actions primary antibodies of mouse anti–c-myc, mouse anti-p53, and mouse anti- of BMP7 in cancer remain unknown. p21 were from Santa Cruz Biotechnology, rabbit anti-p16 were from Cell Cancer cells hold unlimited proliferative potential as the cells Signaling Technology, mouse anti-actin were from Chemicon, and maintain their telomeres (ends of chromosomes) when they divide. horseradish peroxidase–coupled second antibodies were from Dako. Most cancer cells maintain telomeres by mechanisms requiring the Cell culture and treatment. Human cervical cancer HeLa cells were j ribonucleoprotein complex telomerase (20). Telomerase is thus grown in 5% CO2 atmosphere at 37 C in DMEM (Invitrogen) containing 10% fetal bovine serum (FBS) in 6-well plastic plates or 10-cm dishes (Nunc). Recombinant BMP proteins at concentrations of 0.1, 0.3, 3, 10, and 30ng/mL were added to cell cultures for various periods of time as Requests for reprints: Jun-Ping Liu or He Li, Department of Immunology, Central indicated in individual experiments, in which the serum concentration was Eastern Clinical School, Alfred Medical Research and Education Precinct, Commercial 0.5% in the cell culture medium. Cells were lysed in ice cold CAHPS lysis Road, Prahran, Victoria 3181, Australia. Phone: 61-3-99030715; Fax: 61-3-99030120; buffer (0.5% 3-[(Cholamidopropyl)dimethylammonio]-1-propanesulfonate, E-mail: [email protected] or [email protected]. I2008 American Association for Cancer Research. 10mmol/L Tris (pH 7.5), 1 mmol/L MgCl 2, 1 mmol/L EGTA, 0.1 mmol/L doi:10.1158/0008-5472.CAN-08-1323 benzamidine, 5 mmol/L h-mercaptoethanol, and 10% glycerol), in the www.aacrjournals.org 9157 Cancer Res 2008; 68: (22).November 15, 2008

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2008 American Association for Cancer Research. Cancer Research presence of complete mini protease inhibitors (Roche Diagnostic). Clarified synthesized telomeric DNAs were observed after PCR using specific cell lysates were normalized for total protein concentrations by the telomeric DNA primers and [a-32P]ATP (Amersham Biosciences), poly- Bradford protein assay (Bio-Rad). To reverse BMP7 inhibition of telomerase, acrylamide slab gel electrophoresis, and autoradiography. As an internal green fluorescent protein (GFP)-hTERT was expressed by transfection with control of the PCR and loading, the primers NT (ATCGCTTCTC- pEGFP-C2-hTERT plasmid, with pEGPF-C2 plasmid as control. Cell GGCCTTTT) and TSNT (AATCCGTCGAGCAGAGTTAAAAGGCCGA- transfection was conducted using Lipofectamine 2000 (Invitrogen) accord- GAAGCGAT) were included in the reaction. ing to the manufacturer’s instruction. After 24 h of transfection, cells Telomere length analysis. For telomeres in cultured cells, metaphase positive for GFP were isolated by fluorescence-activated cell sorting, spreads from cycling HeLa cells that were treated with or without BMP7 cultured, and incubated with BMP2, 4, 5, 6, or 7 for different periods of time (30ng/mL, 15 h) thrice per week for 2 wk were generated using standard and analyzed in Western blotting, semiquantitative reverse transcription- laboratory protocols. The slides of metaphase cells or tumor sections were PCR (RT-PCR) for gene expressions, telomerase activity assay, and cell death fixed in 4% formaldehyde before treatment with acidified 1% pepsin analysis, as indicated in individual experiments. solution, and hybridized with the probe solution [0.3 Ag/mL Cy3-conjugated

FACS. GFP-transfected cells were sorted using FACSAria flow cytometer [CCCTAA]3 PNA probe (Panagene), 70% formamide, 20 mmol/L Tris-HCl (BD Biosciences). The GFP-positive cells were reseeded into 6-well plates at (pH 7.0), 1% bovine serum albumin (BSA)]. Washing was conducted in PBS/ a density of 0.2 Â 106 cells/mL in DMEM plus 10% FBS including Tween 20with one high stringent wash at 57 jC. DNA was counterstained Gentamicin antibiotics (Pfizer). For apoptosis analysis, after treatment of with 4¶,6-diamidino-2-phenylindole, visualized, and captured using Nikon cells as indicated in individual experiments, cells were dislodged from tissue Eclipse TE2000 microscope, Plan Fluor Â40objective, DS-5MC CCD culture plates using 0.5M EDTA for 5 min at 37jC and incubated with camera, and NIS-Elements F 2.20software (Nikon). Telomere images were annexin-V-FLUOS conjugate (Roche Diagnostic) and propidium iodide captured with a Plan Fluor Â100 oil emersion objective, and individual (PI; Sigma) for 15 min at room temperature in an incubation buffer that telomere fluorescence was integrated using spot IOD analysis in the TFL- facilitates binding per the manufacturers’ instructions. The cells were then Telo 2.2 program (gift from Dr. Peter Lansdorp, Vancouver, Canada; ref. 37). analyzed using FL-2 and FL-3 channel, respectively. Percentages of stained Images from at least 13 metaphase spreads from each data point were apoptotic cells were determined using a FACSCalibur flow cytometer quantified before assembly of data in a standard spreadsheet program. At (BD Biosciences). An acquisition gate was set to include f20,000 of the least 50cell nuclei from each condition were analyzed. centrally located cells for each sample acquisition using linear forward Tumor cell growth analysis in soft agar. Tumor cell colonies were scatter versus linear side scatter. This acquisition strategy resulted in grown in soft agar to assess the effects of BMP7 on tumor growth in vitro. f40,000 ungated events being included for each sample analysis. Dot-plot Two milliliters of mixture of 2Â DMEM containing 20% FBS were used to integration was determinate from the background fluorescence using make 0.5% agar (base layer) with or without BMP7 in a 35-mm culture unstained HeLa cells. This integration cursor placement remained dish or 6-well plate. On top of the base layer, a mixture of serum unchanged when stained HeLa cells were analyzed. supplemented medium and 0.35% agar (2 mL) containing 2,500 HeLa cells Preparation of protein extracts and Western blotting. These in the presence or absence of BMP7 (100 ng/mL or as indicated; top layer) procedures were described previously (30). Briefly, tumor cells were treated was added. The dishes were kept in tissue culture incubator at 37jC and j by CHAP lysis buffer, including protein inhibitors (Roche Diagnostic), at 4 C 5% CO2 for 14 d to allow for colony formation and growth. All assays were and homogenized immediately. The lysates were incubated for 10min on performed in triplicates. The colony assay results were photographed or ice and then centrifuged for 45 min at 5,000 Â rpm, then centrifuged again scanned after the plates were stained with methylthiazolyldiphenyl- for 30 min at 13,000 rpm. The supernatants were collected and stored at tetrazolium (MTT). À80jC. Extracted proteins (f25 Ag) in SDS sample buffer were boiled and Tumor xenografting and treatment procedure. HeLa cells were electrophoresed on a 10% SDS-polyacrylamide gel and electroblotted onto xenografted in female nude mice (BALB/c Nude; Animal Resources Centre a pre-wet nitrocellulose membrane (Bio-Rad Laboratories). The blotted ARC). Cultured monolayer cells were detached by trypsinization, washed in nitrocellulose membrane was blocked in PBS containing 5% skim milk and PBS, and counted. Approximately 10 Â 106 cells were resuspended in 0.1 mL 0.02% Tween 20, and probed with antibodies as indicated in individual PBS and inoculated s.c. in the flank of the mice. Twenty-four hours after experiments at 4jC overnight. The blots were developed using enhanced inoculation, BMPs and other reagents prescribed as indicated in individual chemiluminescence Western blotting detection system (Amersham Bio- experiments were administered into the xenografts on every second day. sciences). The development of xenograft tumors was measured with Vernier callipers RNA isolation and gene expression analysis. As previously described before each treatment. The animals were sacrificed by carbon dioxide (30), total RNA was extracted by TRIzol reagent (Invitrogen) according to asphyxiation when tumor growth in the controls reached 20mm in the manufacturer’s instructions. The RNA (1 Ag) was transcribed by diameter. Immediately after culling, the tumors were removed, measured, oligo(dT) and ThermoScript Reverse Transcriptase (Invitrogen) in a volume snap-frozen in liquid nitrogen, and stored at À80jC for further analyses. of total 21 AL. Contaminants were removed from the samples by Rnase H The mice were maintained under specific pathogen-free conditions at treatment, according to manufacturer’s instruction (Invitrogen). Linear constant temperature (f22jC) and humidity (f40%). Sterilized food and amplification for semiquantitative PCR was performed using ThermoScript water were given ad libitum. RT-PCR kit following the manufacturer’s instruction (Invitrogen), for Tumor tissue sectioning and staining. Freshly dissected tumor 25 to 35 cycles of 30s at 72 jC as the optimized annealing temperature samples were embedded in ornithine carbamyl transferase compound and 64jC extension. Primers specific for different genes were as follows: and frozen using isopentane cooled with liquid nitrogen. Five-micrometer hTERT (5¶-CCACCTTGACAAAGTACAG-3¶) and (5¶-CGTCCAGACTCCGCTT- fresh-frozen sections were prepared and used to stain for the Ki67 CAT-3¶), and Actin (5¶-GCTCGTCGTCGACAACGGCTC-3¶) and (5¶-CAAA- antigen, a marker of cell proliferation. Sections were fixed in 4% CATGATCTGGGTCSTCTTCTC-3¶). Actin was used as control. PCR paraformaldehyde/10% sucrose, washed once in PBS, and blocked with produces were mixed with 6Â loading dye (30% glycerol, 0.5 mol/L EDTA, 5% BSA (Fraction V; Sigma) for 30min. Sections were incubated with and Bromophenol Blue) and run on 1.5% Agarose gel, visualized in the polyclonal rabbit anti-Ki67 antibodies (ab15580; Abcam) at 4jC overnight presence of ethidium bromide, photographed in a gel 1000 UV documen- at a 1:400 dilution. Primary antibody was detected with Cy3-conjugated tation system (Bio-Rad), and analyzed by densitometry. goat anti-rabbit antibodies (Chemicon) for 90min at a 1:200dilution. Telomerase activity assay. Telomerase activity was determined by Nuclei were stained with 0.5 Ag/AL Hoechst 33258 (Sigma). The staining TRAP as described previously (36). Briefly, cells treated with different was visualized by fluorescence microscopy. Five random images within reagents were washed and lysed by detaching and passing the cells though a the tumor area were analyzed in 3 sections, f5 Am apart, for each 261/2G needle attached to a 1-mL syringe in prechilled TRAP lysis buffer. treatment group. Equal amounts of telomerase protein extract (0.4 Ag) were incubated with Statistical analysis. Data were analyzed using Student’s t tests and a telomeric DNA substrate and deoxynucleotide triphosphate, and newly P value of <0.05 was considered statistically significant.

Cancer Res 2008; 68: (22).November 15, 2008 9158 www.aacrjournals.org

Results To verify that the inhibition of telomerase activity in HeLa and BMP7 induces telomerase inhibition and telomere shorten- breast cancer cells was at the levels of gene expression of hTERT, ing in cultured cancer cells. To investigate a potential role of the we measured hTERT mRNA by semiquantitative RT-PCR. As BMP family in telomerase activity, we examined effects of BMP2, shown in Fig. 1D, BMP7 induced a significant down-regulation of BMP4, BMP5, BMP6, and BMP7 by spiking the medium of cancer hTERT gene expression in both HeLa and PMC42 cells. cell cultures with purified recombinant human cytokines. In addition, we noted that BMP7 also induced a significant Incubation of human cervical cancer HeLa cells with different down-regulation of c-myc gene expression (Fig. 1D). The f concentrations of BMP7 for 48 hours resulted in significant inhibition of c-myc gene expression was 80% of control, and inhibition of telomerase activity (Fig. 1A). The maximal inhibition the inhibition of hTERT gene expression was f70% of the control was f85% inhibition of telomerase activity, achieved with the (Fig. 1D). median inhibitory concentration (IC50) of BMP7 of 8 F 0.6 ng/mL To determine the effect of BMP7 on telomeres, we treated HeLa and the maximal inhibitory concentration of 35 F 1.4 ng/mL cells with 30ng/mL of BMP7 for 15 hours each time on every second (n =3;Fig.1B). Incubations of the cells with BMP2, BMP4, BMP5, day for 2 weeks. The pulsatile treatments of cultured HeLa cells with or BMP6 showed no significant inhibitory effect on telomerase BMP7 resulted in a sustained telomerase inhibition and significant activity (data not shown). In the time course studies, BMP7 shortening of telomeres as measured by quantitative fluorescence (10ng/mL) induced telomerase inhibition in 24 hours of the in situ hybridization (Q-FISH), using a specific telomeric DNA probe treatment for f3days(Fig.1B). To determine the inhibitory in cell metaphase spreads. As shown in Fig. 2A, BMP7 treatments of effect on telomerase in another cancer cell type, we tested human cultured HeLa cells caused a marked shift of the telomere breast cancer PMC42 cells with different concentrations of BMP7. fluorescence peak compared with the control. The size of telomeres As shown in Fig. 1C, BMP7 induced significant inhibition of in the control cells was ranged with a major peak of >100 telomerase activity in a dose-dependent manner in PMC42 cells. frequencies between f300 and f1,200 fluorescence unit. In

Figure 1. BMP7 induces inhibition of telomerase activity in cultured cancer cells. A, concentration- dependent inhibition of telomerase activity. HeLa cells were incubated with different concentrations of BMP7 as indicated for 24 h in cell cultures, and then subjected to telomerase activity analysis. Telomerase activity was determined by measuring newly synthesized telomeric DNA as described in the Materials and Methods. Telomeric DNA ladders and an internal loading control are indicated. Lanes 1 and 2, interexperimental positive (P) and negative (N) controls, respectively. B, quantitative data of BMP7 concentration- and time-dependent inhibition of telomerase activity in HeLa cells. Cells were treated for 24 h in the concentration-dependent experiments, or treated with BMP7 at 30 ng/mL for various time periods as indicated in the time course studies. Telomeric DNAs resolved gel electrophoresis and autoradiography were scanned and presented as percentages of the values in nontreated controls. Columns, mean from three determinations; bars, SD. C, effects of BMP7 on telomerase activity in PMC42 cells. PMC42 cells were incubated with BMP7 for 24 h at the indicated concentrations followed by telomerase activity analysis. Telomere DNA produced and internal control are indicated. D, semiquantitative RT-PCR analysis of hTERT and c-myc gene expressions. HeLa and PMC42 cells treated with BMP7 (30 ng/mL, 24 h) were extracted for mRNA and semiquantitative RT-PCR as described in the Materials and Methods. The levels of hTERT and c-myc (top) were quantified as ratios to actin; columns, mean from three similar experiments; bars, SD. *, significant difference compared with nontreated control (P < 0.01).

www.aacrjournals.org 9159 Cancer Res 2008; 68: (22).November 15, 2008

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2008 American Association for Cancer Research. Cancer Research contrast, the telomeres in the BMP7-treated cells were shown as was f50% on each day of the controls treated with diluent, BMP2, a major peak with a relatively narrow range exhibiting between BMP4, BMP5, or BMP6, showing a complete arrest of cell 100 and 1,100 fluorescence unit. The highest frequency of telomere population doubling by the presence of BMP7 (Fig. 3A). Treatment fluorescence in control cells occurred at 800 to 900 fluorescence of the cells with different concentrations of BMP7 for 24 hours units, whereas the highest frequency of telomere fluorescence in showed a concentration-dependent reduction of cell numbers BMP7-treated cells occurred at 500 to 600 fluorescence units (data not shown). To determine if BMP7 induces cancer cell showing significantly shortened telomeres compared with the apoptosis, we examined HeLa treated with or without BMP7 for the control (Fig. 2A). Analyzing the mean telomere length in control apoptotic markers of Annexin V and PI staining by FACS. As shown and BMP7-treated cells showed that the mean telomere length in in Fig. 3B, a single dose of BMP7 treatment resulted in a the BMP7-treated group could be f25% shorter than that in the significantly increase in HeLa cell apoptosis compared with controls (Fig. 2B). The fluorescence micrographs in Fig. 2C showed controls (7% versus 17%). typical images of telomere fluorescence (yellow dots) in different To investigate if BMP7-induced cell death is a consequence of sizes at the ends of chromosomes (blue) between control (left) hTERT gene repression and telomerase inhibition, we analyzed the and BMP7-treated (right) HeLa cells. effect of hTERT gene expression on BMP7-induced cell death in BMP7 induces cancer cell death by a mechanism largely HeLa cell cultures. Underexpression of hTERT with hTERT shRNA dependent on hTERT gene repression. To determine the for 48 hours mimicked the cell killing effect of BMP7 (Fig. 3B), functional consequence of BMP7-induced telomerase inhibition whereas overexpression of hTERT for 24 hours before BMP7 and telomere shortening, we examined the changes in cell number treatment inhibited subsequent BMP7-induced cell death (Fig. 3B), and apoptotic death in the cell cultures treated with or without suggesting that the proapoptotic effect of BMP7 is mediated in a different cytokines over a time course of several days. As shown in significant part by hTERT repression. To further attest the role of Fig. 3A, whereas the cells underwent population doubling in the hTERT in the BMP7-induced cell death, we transfected HeLa cells control group, a single-dose treatment of HeLa cells with BMP7 with GFP-hTERT, GFP-hTERT shRNA, or GFP-only gene expression (10ng/mL) resulted in a marked reduction of total cell number. plasmids, then isolated the GFP-positive transformants by FACS The reduction of cell number in the cell cultures treated with BMP7 and treated the cells with BMP7 (10ng/mL) for 24 hours. In these

Figure 2. BMP7 induces telomere shortening in cervical cancer cells. A, effect of BMP7 on telomere length in HeLa cells. Cells were treated with or without BMP7 (30 ng/mL) for 15 h followed by replacement with fresh medium on every second day for 2 wk. Telomeres were determined by Q-FISH, and the distributions of telomere length are presented by fluorescence intensity versus frequencies. B, histograms of mean telomere fluorescence intensities (FSE) in control and BMP7-treated cells. C, illustrative micrographs of fluorescence- labeled telomeres in metaphase spreads.

Cancer Res 2008; 68: (22).November 15, 2008 9160 www.aacrjournals.org

Figure 3. BMP7-induced cancer cell apoptosis requires hTERT gene repression. A, inhibition of cancer cell proliferation by BMP7. HeLa cells were incubated with or without 30 ng/mL of BMP7, BMP6, BMP5, BMP4, or BMP2 for different periods of time as indicated. Cell numbers were counted in 20 AL at each time when cells were subcultured or terminated in Coulter counter (Beckman Coulter; Z Series). *, significant difference from the nontreated control (P < 0.05). B, effects of BMP7 on apoptosis. HeLa cells transfected with empty plasmid, hTERTshRNA, or hTERT wild-type gene expression plasmid for 24 h and then incubated with or without BMP7 (10 ng/mL) for further 24 h, as indicated. Cells were stained with Annexin V and PI, and analyzed by FACS. C, effects of gene expression of hTERT or hTERT shRNA on BMP7-induced HeLa cell death. Cells transfected with pEGFP, pEGFP-hTERT, or pEGFP-hTERT shRNA for 24 h were isolated by FACS sorting and incubated with BMP7 (10 ng/mL) or diluent as indicated for 24 h. Apoptosis was analyzed by PI staining in FACS. The data were from one of three similar experiments. D, effects of BMP7 in telomerase-negative cell lines. GM847, Saos2, and HeLa cells were incubated in duplicate with different concentrations of BMP7 for 48 h. homogenously transfected cell cultures, we found that the the control (Fig. 4A). The inhibition was concentration dependent, expression of GFP-hTERT significantly decreased the levels of cell with the IC50 and maximal inhibition concentrations being 4 and death induced by BMP7 (from 10-22% versus from 10-17%; Fig. 3C), 20ng/mL, respectively (Fig. 4 A, right). To attest the effect of BMP7 corroborating that hTERT repression is required to a significant on tumor growth in vivo, we exploited the xenograft tumor model degree in BMP7-induced cell death. Consistently, down-regulation in immune-deficient nude mice and carried out intratumor of hTERT with GFP-hTERT shRNA increased cell death above basal injection of different concentrations of BMP7 on every second levels (Fig. 3C). Consistent with a significant role of telomerase day for 2 weeks. BMP7 markedly inhibited tumor growth compared inhibition in mediating BMP7-induced apoptosis, BMP7 did not with the controls (Fig. 4B). Significant tumor growth inhibition was induce significant cell death in the telomerase-negative GM847 and observed after the first one or two injections of 10ng/mL of BMP7, Saos2 cell cultures (Fig. 3D). Verification of BMP7 receptors by with maximal inhibition observed after three injections (Fig. 4B). Western blotting with specific antibodies confirmed the presence Inhibition was concentration dependent, with the IC50 being of both BMPR1A and BMPR2 receptors in Saos2 and HeLa cell f10ng/mL and maximal inhibitory concentration of 30ng/mL of lines, although BMPR1A was not detectable in GM847 cells BMP7 (Fig. 4C). The effective concentrations were comparable with (data not shown). that inducing telomerase inhibition (Fig. 1A) and inhibiting BMP7 inhibits tumor growth with hTERT gene repression thymidine incorporation in human myeloma cells (12). In 2 weeks and telomere shortening in mouse xenograft tumors. To of the treatment, BMP7 induced a maximal tumor growth inhibition investigate if BMP7 inhibits tumor growth, we examined the effect by f75% of the controls (Fig. 4C), whereas no significant inhibitory of BMP7 on tumor cell anchorage–independent growth in soft agar effect was observed on the tumors receiving BMP2, BMP4, BMP5, or assays. As shown in Fig. 4A (left), numerous tumor colonies grew BMP6 under the same experimental conditions (Fig. 4B). from the HeLa cell suspension under basal conditions. In the To determine the intermediate role of hTERT gene repression in presence of BMP7, however, there was a dramatic inhibition of the BMP7-induced tumor growth inhibition, we carried out tumor colony formation and growth, resulting in <15% colonies of underexpression and overexpression of hTERT in the xenograft www.aacrjournals.org 9161 Cancer Res 2008; 68: (22).November 15, 2008

Figure 4. BMP7 inhibition of tumor growth. A, micrograph of BMP7 suppression of tumor growth in soft agars. HeLa cells (3 Â 106) in 0.35% soft agar suspension were incubated with or without BMP7 at 10 ng/mL (left) or different concentrations (right) for 7 d. Tumor colonies were observed after staining with MTT. Columns, mean from four different determinations; bars, SD. B, effects of different BMPs on xenograft tumor growth in mice. HeLa cells (3 Â 106) were inoculated in nude mice s.c. Different BMPs (10 ng/mL in 50 AL PBS) were injected in each xenograft 24 h after inoculation and then on every second day for 2 wk. Tumor growth was measured before each injection. Columns, mean from multiple experiments (22 animals in PBS groups, 24 animals in BMP7-treated groups, and 4 to 8 animals in groups receiving other BMPs); bars, SD. *, significant difference from PBS control with P < 0.01. C, dose-dependent effect of BMP7 on xenograft tumor growth in mice. BMP7 of different doses was administered as indicated on every second day for 2 wk. Columns, mean from four animals in each group; bars, SD. D, expression of hTERT reverses BMP7-induced tumor growth inhibition. Empty plasmid, hTERT shRNA expression plasmid, or hTERT wild-type gene expression plasmid was intratumor injected 1 d after tumor inoculation and continued on every second day for 2 wk. One day after injections of empty plasmid and hTERT wild-type gene expression plasmid, and 2 d after HeLa cell inoculation, BMP7 (10 ng/mL) was administered, which was continued on every second day for 2 wk. Columns, mean from four animals in each treatment group; bars, SD. *, significant differences from that in PBS control.

Cancer Res 2008; 68: (22).November 15, 2008 9162 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2008 American Association for Cancer Research. BMP7 Induces Telomere Shortening in Cervical Tumors tumors and determined the effect of BMP7 in different back- 2 weeks was significantly shorter than that in the controls (Fig. 6A grounds of hTERT gene expression on tumor growth. Significant and B). The shortening of telomeres induced by the pulsatile inhibitory effect on tumor growth was observed in the animals treatment with BMP7 for 2 weeks was 25% to 30% of the controls. treated by intratumor injection of the hTERT shRNA expression On the contrary, overexpression of hTERT resulted in lengthened plasmid (Fig. 4D). Overexpression of wild-type hTERT has no telomeres to f130% of the control (Fig. 6A and B) in association significant effect on tumor growth in the cohort receiving with a significant reversal of BMP7-induced tumor growth arrest intratumor injection of hTERT expression plasmid. However, (Fig. 4D). Consistent with a functional consequence of telomere overexpression of hTERT significantly compromised the BMP7- shortening, we observed a significant decrease of cell proliferation induced inhibition of tumor growth, reversing tumor growth as indicated by Ki67 staining in BMP7-treated tumors. As shown in inhibition from 25% back to 70% to 80% of the control after Fig. 6C and D, BMP7 elicited an inhibition of cell proliferation by 2 weeks of treatments (Fig. 4D). Examinations of telomerase f60% as estimated in Ki67 staining on tumor tissue sections. activity in the tumors treated with or without BMP7 showed that These in vivo findings from the xenograft tumors mirrored the telomerase activity was almost completely inhibited in the tumor findings from cell culture studies, demonstrating that BMP7 tissues treated with BMP7 for 2 weeks, in contrast to the controls triggered an intracellular signaling pathway leading to repressed and other groups receiving treatments with BMP2, BMP4, BMP5, or gene expressions of c-myc and hTERT, increased gene expressions BMP6 (Fig. 5A and B). To determine altered gene expressions, we of cell cycle checkpoints, inhibition of telomerase activity, found that BMP7 treatment rendered a marked inhibition of the shortening of telomeres, and cancer cell apoptosis. endogenous hTERT gene expression after 2 weeks of treatment (Fig. 5C and D). In addition, BMP7 treatment also induced a significant inhibition of the oncogene c-myc gene expression Discussion (Fig. 5C and D), suggesting that BMP7 induces the repression of the The maintenance of telomeres is critical for cancer cell hTERT gene through a transcription-dependent mechanism proliferation particularly as telomeres are already very short in involving c-myc gene repression (30, 33). Furthermore, correspond- cancer cells (21, 24). Inhibition of telomere maintenance offers an ing with a BMP7-triggerd deregulation of telomeres and its asso- important mechanism to inhibit cancer cell proliferation. In this ciated DNA damage response, we observed significant increases in study, we show that the cytokine BMP7 triggers an inhibition of the levels of p53 and p16 tumor suppressors in the tumors treated telomere maintenance and cervical cancer cell growth arrest with BMP7 (Fig. 5C and D). The increases of p53 and p16 were in vitro and in tumor xenografts. A single application of significant compared with the control, reflecting an activation of recombinant BMP7 produces a significant inhibition of telomerase the cell cycle checkpoints and arrest of the cell cycle (Fig. 5D). activity critical in telomere maintenance in cultured cervical Examination of telomeres in the tumor tissues showed that the cancer HeLa cells. Continuous applications of BMP7 on every average telomere length in the tumors treated with BMP7 for second day for 2 weeks in HeLa cell cultures or xenograft tumors

Figure 5. BMP7 induces telomerase inhibition, hTERT gene repression, and p53 and p16 gene activation in mouse xenograft tumors. A, effects of BMPs on telomerase activity in tumors. Tumors were treated with or without different BMPs (10 ng/mL) on every second day for 2 wk. Telomerase activity in the tumors was measured by TRAP. B, telomerase activity quantified by densitometry. Columns, mean (n = 3); bars, SD. Telomere DNA and internal control are indicated. C, altered gene expressions of hTERT, c-myc, and p53 and p16 in the xenograft tumors treated with BMP7. Quantitative data of gene expressions of c-myc, hTERT, p53, p16, and actin in tumors treated with and without BMP7 as indicated. Data were obtained by densitometry; columns, mean from three determinations; bars, SD. D, semiquantitative RT-PCR determination of hTERT and c-myc, and Western blotting for p53 and p16. The results are representative of multiple experiments.

www.aacrjournals.org 9163 Cancer Res 2008; 68: (22).November 15, 2008

Figure 6. BMP7 induces telomere shortening and tumor cell growth arrest in mouse xenograft tumors. A, BMP7 induces telomere shortening in tumors. Xenograft tumors in mice were treated on every second day for 2 wk. The tumors were sectioned and examined for telomeres by Q-FISH (Materials and Methods). Red dots, illustrative micrographs of Q-FISH with merged images showing telomeres; blue dots, DNA. B, quantitative histogram of the mean telomere fluorescence intensity. Columns, mean from at least 50 cells chosen randomly from each of the four conditions indicated; bars, SE. P values indicate the levels of significant differences between the means of treated groups and control. C, decreased Ki67 as cell proliferation marker in xenograft tumors treated with BMP7. Sections from control and BMP7-treated tumors were fixed and incubated with Ki67 antibodies (red) and were stained for nuclei with Hoechst 33258 (blue). D, quantitative analysis of cell proliferation by counting Ki67 positively cells from 15 views of 3 sections in each group. Columns, mean of percentage in total cells; bars, SE.

result in a sustained depression of telomerase activity and the findings that c-myc is down-regulated in association with shortening of telomeres. Consistently, BMP7 induces a gene hTERT down-regulation suggest that repression of c-myc contrib- expression pattern of decreased hTERT and c-myc, and increased utes to the repression of the hTERT gene induced by BMP7. p16 and p53. Furthermore, BMP7 induces inhibition of cancer cell Previous studies show that BMP7 induces apoptosis of myeloma anchorage–independent growth in soft agar and xenograft tumor cells (12) and prostate cancer cells (13, 18). Consistently, our studies growth. These findings suggest that telomere homeostasis show that BMP7 also induces cervical cancer cell apoptosis. In is subject to regulation by extracellular cytokines in which addition, we show for the first time that BMP7 inhibits cervical BMP7 plays a significant role in eliciting a negative regulation of tumor growth in mice. Our in vivo findings in cervical cancer cells telomere maintenance and cell proliferative capacity in cancer. are consistent with recent studies showing BMP7 antitumor effect Furthermore, our data also suggest that BMP7 has a potential in breast cancer (19) and prostate cancer (19). Furthermore, we have utility in anticancer effect by targeting cancer cell telomere found that BMP7 inhibits telomerase activity and induces telomere maintenance. shortening as a fundamental mechanism in BMP7-induced cell The specificity of BMP7-induced inhibition of telomerase death. In line with the recent findings that BMP7 maintains activity, telomere maintenance and cancer cell proliferation was epithelial cell phenotypes, and inhibits epithelial-mesenchymal shown by testing different BMPs. The dose-dependent effect of transition and cancer cell metastasis (16, 17, 19), our data suggest BMP7, with an effective concentration within a physiologically that BMP7 serves a function to inhibit cancer cell proliferation by relevant range (11, 38, 39), supports a specific effect of BMP7 inducing intracellular signaling to counteract mitogenic stimulation mediated by BMP7 receptors on tumor cell surface. Time course of telomerase activity by factors such as c-myc. Endogenous BMP7 studies showing that BMP7 inhibition of telomerase activity occurs produced in cancer cells may thus be a protective mechanism in 24 hours of treatment further suggest that BMP7 inhibits against oncogenic development, and reinforcement with exogenous telomerase activity by modifying specific programs of gene recombinant BMP7 produces an imbalance in favor of BMP7 expression in the tumor cells. We identify that the effect of predominance by counterbalancing the actions of oncogenic BMP7 on cancer cell apoptosis is mediated by a mechanism growth factors and cytokines in extracellular microenvironment involving repression of the hTERT gene and inhibition of of cancer. BMP7-induced inhibition of telomerase activity and telomerase activity. Underexpression of hTERT mimics the cell shortening of telomeres may therefore represent a powerful killing effect of BMP7, whereas overexpression of hTERT mechanism to intercept the process of immortalization of significantly inhibits the BMP7-induced cell death. Although future neoplastic cells that use telomerase to maintain telomeres. To the studies are required to investigate the mechanisms underlying cancer cells that adopt alternative mechanisms to maintain BMP7-induced hTERT gene repression and telomere shortening, telomeres, BMP7 seems ineffective as we show that BMP7 does

Cancer Res 2008; 68: (22).November 15, 2008 9164 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2008 American Association for Cancer Research. BMP7 Induces Telomere Shortening in Cervical Tumors not induce significant cell death of Saos2 cells that express BMP7 cancer cell death may be prompted by a series of interchanges receptors and are telomerase negative. Because BMP7 inhibition of including telomerase inhibition, telomere shortening, telomere- prostate cancer cell proliferation is also dependent on the subtypes associated DNA damage responses, and subsequent cell senescence of derived cell lines (18), it would be interesting to determine if cells and apoptosis (48–50). In summary, our data show a novel that respond differently to BMP7 without undergoing apoptosis mechanism of BMP7-negative regulation of telomere maintenance assume a different mechanism to stabilize telomeres during and cell proliferation in cancer by a mechanism of hTERT gene continuous proliferation. However, our findings that constant repression and telomerase inhibition in vitro and in vivo. Our data presence of high levels of BMP7 results in sustained telomerase suggest that telomeres undergo continuous remodeling, which can inhibition and telomere shortening illustrate a novel model in be reprogramming by defined combinations of extracellular factors, cytokine therapeutic intervention of cancer development through with BMP7 serving as a major negative regulator. Further studies targeting telomerase maintenance of telomeres. should unveil a complete picture of extracellular regulation of It is currently thought that deregulation of telomeres instigates telomere remodeling by various factors in cancer and other highly rapid cell death via mechanisms including telomere decapping and proliferative cells such as stem cells. DNA damage response. Our observation that BMP7 induces telomerase inhibition and telomere shortening in association with imminent cancer cell death are consistent with the recent findings Disclosure of Potential Conflicts of Interest that inhibition of telomerase induces a rapid cancer cell death (40, No potential conflicts of interest were disclosed. 41). Our findings that BMP7 induces hTERT repression–dependent inhibition on tumor growth are also consistent with recent reports Acknowledgments that inhibition of telomerase induces telomere shortening and tumor growth arrest (42, 43). In support of telomeric DNA damage Received 4/9/2008; revised 8/15/2008; accepted 8/27/2008. Grant support: National Health and Medical Research Council of Australia, response, we show that BMP7 induces increased p16 and p53 cell Australia Research Council, and Cancer Council of Victoria, Australia. cycle checkpoint activities. Critical in tumor cell aging and death The costs of publication of this article were defrayed in part by the payment of page (44, 45), increased p16 and p53 activities have previously been charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. shown to be involved in coupling telomere deregulation with cell We thank Peter Lansdorp for the TFL-Telo 2.2 software program for telomere senescent and apoptotic response (46, 47). Thus, BMP7-induced analysis.

References apoptosis in human myeloma cells. Oncogene 2004;23: LINE-1 retrotransposition at mammalian telomeres. 3024–32. Nature 2007;446:208–12. 1. Patel SR, Dressler GR. BMP7 signaling in renal 13. Miyazaki H, Watabe T, Kitamura T, Miyazono K. BMP 24. Shay JW, Wright WE. Telomerase therapeutics for development and disease. Trends Mol Med 2005;11: signals inhibit proliferation and in vivo tumor growth of cancer: challenges and new directions. Nat Rev Drug 512–8. androgen-insensitive prostate carcinoma cells. Onco- Discov 2006;5:577–84. 2. Milan M. Sculpting a fly leg: BMP boundaries and cell gene 2004;23:9326–35. 25. Hahn WC. Telomere and telomerase dynamics in death. Nat Cell Biol 2007;9:17–8. 14. Beck SE, Jung BH, Del Rosario E, Gomez J, Carethers human cells. Curr Mol Med 2005;5:227–31. 3. Attisano L, Wrana JL. Signal transduction by the JM. BMP-induced growth suppression in colon cancer 26. Holt SE, Wright WE, Shay JW. Regulation of TGF-h superfamily. Science 2002;296:1646–7. cells is mediated by p21(WAF1) stabilization and telomerase activity in immortal cell lines. Mol Cell Biol 4. Massague J, Seoane J, Wotton D. Smad transcription modulated by RAS/ERK. Cell Signal 2007;19:1465–72. 1996;16:2932–9. factors. Genes Dev 2005;19:2783–810. 15. Jena N, Martin-Seisdedos C, McCue P, Croce CM. 27. Li H, Pinto AR, Duan W, Li J, Toh BH, Liu JP. 5. Foletta VC, Lim MA, Soosairajah J, et al. Direct BMP7 null mutation in mice: developmental defects Telomerase down-regulation does not mediate PC12 signaling by the BMP type II receptor via the in skeleton, kidney, and eye. Exp Cell Res 1997;230: pheochromocytoma cell differentiation induced by NGF, cytoskeletal regulator LIMK1. J Cell Biol 2003;162: 28–37. but requires MAP kinase signalling. J Neurochem 2005; 1089–98. 16. Buijs JT, Rentsch CA, van der Horst G, et al. BMP7, 95:891–901. 6. Wen Z, Han L, Bamburg JR, Shim S, Ming GL, Zheng a putative regulator of epithelial homeostasis in the 28. Kyo S, Inoue M. Complex regulatory mechanisms of JQ. BMP gradients steer nerve growth cones by a human prostate, is a potent inhibitor of prostate telomerase activity in normal and cancer cells: How can balancing act of LIM kinase and Slingshot phosphatase cancer bone metastasis in vivo. Am J Pathol 2007;171: we apply them for cancer therapy? Oncogene 2002;21: on ADF/cofilin. J Cell Biol 2007;178:107–19. 1047–57. 688–97. 7. Lu M, Lin SC, Huang Y, et al. XIAP induces NF-nB 17. Zeisberg EM, Tarnavski O, Zeisberg M, et al. 29. Takakura M, Kyo S, Inoue M, Wright WE, Shay JW. activation via the BIR1/TAB1 interaction and BIR1 Endothelial-to-mesenchymal transition contributes to Function of AP-1 in transcription of the telomerase dimerization. Mol Cell 2007;26:689–702. cardiac fibrosis. Nat Med 2007;13:952–61. reverse transcriptase gene (TERT) in human and mouse 8. Qiu T, Grizzle WE, Oelschlager DK, Shen X, Cao X. 18. Yang S, Zhong C, Frenkel B, Reddi AH, Roy-Burman cells. Mol Cell Biol 2005;25:8037–43. Control of prostate cell growth: BMP antagonizes P. Diverse Biological Effect and Smad Signaling of Bone 30. Li H, Xu D, Li J, Berndt MC, Liu JP. Transforming androgen mitogenic activity with incorporation of Morphogenetic Protein 7 in Prostate Tumor Cells. growth factor h suppresses human telomerase reverse MAPK signals in Smad1. EMBO J 2007;26:346–57. Cancer Res 2005;65:5769–77. transcriptase (hTERT) by Smad3 interactions with 9. Alarmo EL, Rauta J, Kauraniemi P, Karhu R, Kuukas- 19. Buijs JT, Henriquez NV, van Overveld PG, et al. Bone c-Myc and the hTERT gene. J Biol Chem 2006;281: jarvi T, Kallioniemi A. Bone morphogenetic protein 7 is morphogenetic protein 7 in the development and 25588–600. widely overexpressed in primary breast cancer. Genes treatment of bone metastases from breast cancer. 31. Li H, Liu JP. Mechanisms of action of TGF-h in Chromosomes Cancer 2006;45:411–9. Cancer Res 2007;67:8742–51. cancer: evidence for Smad3 as a repressor of the hTERT 10. Rothhammer T, Wild PJ, Meyer S, et al. Bone 20. Blackburn EH. Switching and signaling at the gene. Ann N Y Acad Sci 2007;1114:56–68. morphogenetic protein 7 (BMP7) expression is a telomere. Cell 2001;106:661–73. 32. Li H, Xu D, Toh BH, Liu JP. TGF-h and cancer: is potential novel prognostic marker for recurrence in 21. Xu L, Blackburn EH. Human cancer cells harbor Smad3 a repressor of hTERT gene? Cell Res 2006;16: patients with primary melanoma. Cancer Biomark 2007; T-stumps, a distinct class of extremely short telomeres. 169–73. 3:111–7. Mol Cell 2007;28:315–27. 33. Hu B, Tack DC, Liu T, Wu Z, Ullenbruch MR, Phan 11. Grijelmo C, Rodrigue C, Svrcek M, et al. Proinvasive 22. Negrini S, Ribaud V, Bianchi A, Shore D. DNA breaks SH. Role of Smad3 in the regulation of rat telomerase activity of BMP-7 through SMAD4/src-independent and are masked by multiple Rap1 binding in yeast: reverse transcriptase by TGFh. Oncogene 2006;25: ERK/Rac/JNK-dependent signaling pathways in colon implications for telomere capping and telomerase 1030–41. cancer cells. Cell Signal 2007;19:1722–32. regulation. Genes Dev 2007;21:292–302. 34. Fujiki T, Miura T, Maura M, et al. TAK1 represses 12. Ro TB, Holt RU, Brenne AT, et al. Bone morphoge- 23. Morrish TA, Garcia-Perez JL, Stamato TD, Taccioli transcription of the human telomerase reverse tran- netic protein-5, -6 and -7 inhibit growth and induce GE, Sekiguchi J, Moran JV. Endonuclease-independent scriptase gene. Oncogene 2007;26:5258–66. www.aacrjournals.org 9165 Cancer Res 2008; 68: (22).November 15, 2008

35. Lacerte A, Korah J, Roy M, Yang XJ, Lemay S, Lebrun formation and suppressing bone resorption. J Biol Chem tumour clearance is triggered by p53 restoration in JJ. Transforming growth factor-h inhibits telomerase 2006;281:25509–21. murine liver carcinomas. Nature 2007;445:656–60. through SMAD3 and E2F transcription factors. Cell 40. Li S, Crothers J, Haqq CM, Blackburn EH. Cellular 45. Krimpenfort P, Ijpenberg A, Song JY, et al. p15Ink4b is Signal 2008;20:50–9. and gene expression responses involved in the rapid a critical tumour suppressor in the absence of p16Ink4a. 36. Li H, Zhao LL, Funder JW, Liu JP. Protein growth inhibition of human cancer cells by RNA Nature 2007;448:943–6. phosphatase 2A inhibits nuclear telomerase activity interference-mediated depletion of telomerase RNA. 46. Smogorzewska A, de Lange T. Different telomere in human breast cancer cells. J Biol Chem 1997;272: J Biol Chem 2005;280:23709–17. damage signaling pathways in human and mouse cells. 16729–32. 41. Cao Y, Li H, Deb S, Liu JP. TERT regulates cell EMBO J 2002;21:4338–48. 37. Rufer N, Dragowska W, Thornbury G, Roosnek E, survival independent telomerase activity. Oncogene 47. Jacobs JJ, de Lange T. p16INK4a as a second effector Lansdorp PM. Telomere length dynamics in human 2002;21:3130–8. of the telomere damage pathway. Cell Cycle 2005;4: lymphocyte subpopulations measured by flow cytom- 42. Djojosubroto MW, Chin AC, Go N, et al. Telomerase 1364–8. etry. Nat Biotechnol 1998;16:743–7. antagonists GRN163 and GRN163L inhibit tumor growth 48. de Lange T. Protection of mammalian telomeres. 38. Ma T, Gutnick J, Salazar B, et al. Modulation of and increase chemosensitivity of human hepatoma. Oncogene 2002;21:532–40. allograft incorporation by continuous infusion of growth Hepatology 2005;42:1127–36. 49. ArtandiSE,ChangS,LeeSL,etal.Telomere factors over a prolonged duration in vivo. Bone 2007;41: 43. Pallini R, Sorrentino A, Pierconti F, et al. Telomerase dysfunction promotes non-reciprocal translocations 386–92. inhibition by stable RNA interference impairs tumor and epithelial cancers in mice. Nature 2000;406:641–5. 39. Simic P, Culej JB, Orlic I, et al. Systemically growth and angiogenesis in glioblastoma xenografts. Int 50. Denchi EL, de Lange T. Protection of telomeres administered bone morphogenetic protein-6 restores J Cancer 2006;118:2158–67. through independent control of ATM and ATR by TRF2 bone in aged ovariectomized rats by increasing bone 44. Xue W, Zender L, Miething C, et al. Senescence and and POT1. Nature 2007;448:1068–71.

Cancer Res 2008; 68: (22).November 15, 2008 9166 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2008 American Association for Cancer Research. Bone Morphogenetic Protein-7 Inhibits Telomerase Activity, Telomere Maintenance, and Cervical Tumor Growth

Lucy Cassar, He Li, Alexander Ruvantha Pinto, et al.

Cancer Res 2008;68:9157-9166.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/68/22/9157

Cited articles This article cites 49 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/68/22/9157.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/68/22/9157.full#related-urls

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

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

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/68/22/9157. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.