The authors have declared that no competing interests exist.
Conceived and designed the experiments: SYS. Performed the experiments: CML YCH SJL TFH SJ SYS. Analyzed the data: CLH SYS MCH. Contributed reagents/materials/analysis tools: MCH JAL SYS LWKC. Wrote the paper: SYS.
Cancer cells respond to stress by activating a variety of survival signaling pathways. A disintegrin and metalloproteinase (ADAM) 9 is upregulated during cancer progression and hormone therapy, functioning in part through an increase in reactive oxygen species. Here, we present
Occurring in more than 80% of advanced-stage prostate cancer cases, skeletal metastases correlates with a high level of morbidity; a 5 year survival rate of 25% and median survival of approximately 40 months
Despite recent advances in therapeutic strategies, many malignant cancers still develop resistance to radiation and targeted therapies
In the present study, we assess the feasibility of lentiviral vector-delivered small hairpin RNA (shRNA) against ADAM9 for the treatment of androgen-independent and bone metastatic human prostate cancer in an experimental animal model. The molecular mechanism underlying the therapeutic action of ADAM9 targeted gene therapy was also elucidated.
Retroviral vectors containing shRNA that targets ADAM9 and control shRNA were obtained from Open Biosystems (Lafayette, CO). Lentiviral vector ADAM9 shRNA and controls were obtained from the National RNAi Core Facility at the Institute of Molecular Biology/Genomic Research Center, Academia Sinica, Taiwan. The anti-ADAM9 antibodies were obtained from R&D Systems (Minneapolis, MN). Anti-human EF1-α obtained from Millipore (Billerica, MA) was used as a control antibody.
The androgen-independent metastatic prostate cancer cell line, PC3 and androgen-dependent prostate cancer cell line, LNCaP, were used in previous studies
The retroviral plasmid pSM2C and shRNA specific for ADAM9, as well as the control plasmid pGIPZ and shRNA specifically targeting REG4 were obtained from Open Biosystems. The lentiviral pLKO, control shGFP, and shRNA targeting the mRNA of the ADAM9 coding sequence were obtained from the National RNAi Core Facility at the Institute of Molecular Biology/Genomic Research Center, Academia Sinica, supported by the National Research Program for Genomic Medicine Grants of NSC (NSC 97-3112-B-001-016). Target information is listed in
Protein extracts from the cell lines were analyzed on SDS-polyacrylamide gels (15 µg per lane) and transferred to Hybone ECL nitrocellulose membranes (GE Healthcare Life Science, Piscataway, NJ). Blots were probed with monoclonal mouse anti-human antibodies against ADAM9 (R&D systems), p21, and p27 (kind gift from Dr. Yun-Lung Yu at China Medical University & Hospital) according to the manufacturer’s instructions. BT474 cell lysate (cells were treated with 500 nM doxorubicin overnight) was obtained from Dr. Wei-Chien Huang at China Medical University & Hospital. For loading control, blots were probed with an anti-EF1-α monoclonal antibody (1∶10,000; Millipore). After incubation with an HRP-conjugated secondary antibody (1∶5000; GE Healthcare Life Science), chemiluminescent signals were detected using an ECL Plus kit and the blots were exposed to Hyperfilm ECL (GE Healthcare Life Science). Protein band quantification was carried out using ImageJ software (
Cells were seeded on dishes at 70–80% confluence and starved overnight. For quantification analysis, cells were trypsinized, centrifuged, and resuspended in HBSS with or without 5% FBS (depending on whether they were exposed to radiation or starvation). Hydrogen peroxide was detected using dichlorofluorescindiacetate (DCF, 2 µM) (Invitrogen). Superoxide was detected using dihydroethidium (DHE, 10 µM) (Invitrogen). Superoxide imaging was detected using MitoSox Red mitochondrial superoxide indicator (Invitrogen). Samples were incubated for 40 min at room temperature in the dark on a rotator. Analysis of DCF and DHE fluorescence was performed using a FACSCalibur (BD Biosciences, San Jose, CA) according to manufacturer’s instructions. For imaging mitochondrial superoxide generation, cells were starved for 24 hours followed by loading with MitoSOX Red, Alexa-488 WGA and DAPI (Invitrogen) for 30 min and viewing under a fluorescence microscope.
The effect of ADAM9 shRNA on cell proliferation was measured by directly counting the number of cells. Briefly, cells were plated at a density of 1×105 on 60-mm dish. At designated times, the cells were removed by trypsinization, and the number of viable cells was counted in a hemocytometer with the use of trypan blue (0.4%) staining.
For the clonogenic cell proliferation assay, cell suspensions were prepared by treatment with 0.25% trypsin and 0.05% EDTA. After calculating cell concentration using a hemocytometer, 100 cells were seeded in 6-well plates with T-medium and 5% FBS for 2 weeks. The number of colonies in each well was determined by counting the number of cells in each colony. Only colonies with ≥50 cells were defined as successful. To identify colonies, medium was withdrawn and 3.7% formaldehyde was added for 15 minutes, followed by incubation in 1 mL crystal violet for 15 minutes at room temperature. Colony images were taken and cell numbers calculated using ImgaeJ.
The invasiveness of cancer cells was assessed using 24-well Transwell (Corning, Lowell, MA) plates. In brief, 2×105 cells in media containing 0.5% FBS were added to the upper chamber containing 8 µm pore polycarbonate coated with 1 mg/mL matrigel; the lower chamber was filled with media containing 5% FBS. After 16 h incubation, the upper surface of the membrane was scrubbed with a cotton-tipped swab. The invading cells on the lower surface of the membrane were fixed and stained with 0.5% crystal violet. Random fields (5 per membrane) were photographed at 40x magnification for calculating cell number. In addition, cells were quantified by measuring the absorbance of dye extracts at 570 nm in 100 µl of Sorenson’s solution (9 mg tirsodium citrate, 305 mL distilled water, 195 mL 0.1 N HCl, 500 mL 90% ethanol).
Cells resuspended in culture medium were seeded into 24-well plates. A single wound was created in the center of the cell monolayer by gentle removal of the attached cells with a sterile plastic pipet tip after cell cultures reached ≥90% confluence. The debris was removed by washing with serum-free media. Cells that migrated into the wounded area or those protruding from the border of the wound were visualized and photographed under a Zeiss Axioplan microscope (Carl Zeiss MicroImaging, Thornwood, NY) with a 10× objective at five preselected time points (0, 2, 4, 6, and 8 h). Each experiment was independently performed at least three times.
Five-micron thick paraffin-embedded tissue sections were deparaffinized and rehydrated. The tissue sections were incubated for 2 hours with mouse monoclonal anti-human ADAM9 antibody (1∶50 dilution, R & D Systems). After washing to remove unbound primary antibody, sections were treated with a dextran polymer backbone conjugated to secondary antibodies and labeled with horseradish peroxidase according to manufacturer’s instructions (DAKO Envision system for mouse and rabbit primary antibodies, DAKO Corporation, Carpinteria, CA) for 30 minutes. Tissue sections were incubated in the chromogenic peroxidase substrate, diaminobenzidine, for 5 minutes; sections were then counterstained with automation hematoxylin (DAKO) for 15 minutes. The specificity of labeling by this procedure was verified by negative control reactions using buffer to replace the primary antibody and isotype-specific IgG. Masson’s trichrome staining was performed using a Masson’s trichrome stain kit (Sigma-Aldrich) following the standard protocol provided by the manufacturer.
Immunostaining with TRAcP was carried out according to the manufacturer’s protocol (Takara Bio Inc, Shiga, Japan). Briefly, paraffin-embedded mouse tibia bone sections were deparaffinized and rehydrated for the TRAc Posteolytic reaction. The bone sections were incubated with the substrate solution (0.1 volume of sodium tartrate in Naphthol-AS-BI-phosphate/Fast Red Violet LB substrate solution, pH 5.2) at 37°C for 45 minutes. After washing the slide with distilled water and adding glycerol to prevent dehydration, the samples were examined under the microscope.
Cancer cells were plated at 1×106 cells per 100 mm dish and starved for either 24 or 48 hours. Cells were tryspinized and subjected to cell cycle analysis using propidium iodide as reported previously
Five-week-old male athymic nude (nu/nu) mice obtained from National Laboratory Animal Center at Taiwan were used for subcutaneous (s.c.), intracardiac, and intratibial tumor implantation. Cells were cultured to 100% confluence, trypsinized and enumerated. Mice were sedated with 1.7% isoflurane mixed with air for s.c. tumor implantation. Xenograft tumors were established by s.c. injection of 106 PC3 cells expressing pSM2c or shADAM9 cells in 100 µL of PBS into both sides of each mouse. Animals were sacrificed after 8 weeks. There were no significant body weight differences between the animals with and without s.c. tumors at the time of sacrifice. For intratibial tumor injection, a 27-gauge needle was used to bore a hole in the marrow cavity through the tibial plateau into both the left and right tibias. A 28-gauge Hamilton syringe was used to inject 5×105 cells in a 50 µL volume into the marrow cavity. Tumor growth was monitored by both bioluminescence and x-rays once per week. Animals were sacrificed after 8 weeks. Tibias were removed, washed in PBS and fixed in 10% formaldehyde at room temperature for 24 h. Bone specimens were washed with PBS followed by decalcification with 0.25 M EDTA in PBS (pH 7.4) at room temperature for 6 weeks. The EDTA decalcification solutions were changed weekly. Tumor growth was monitored weekly using a bioluminescence imaging system (Xenogen IVIS 200 Series, Caliper, Hopkinton, MA).
Total RNA was isolated from cultured cells (Roche Applied Science, Mannheim, Germany) following the manufacturer’s protocol. Purified RNA samples were submitted to Phalanx Biotech for microarray analysis using the Human Whole Genome OneArray® Microarray v5 (Phalanx Biotech Group, Inc., Hsinchu, Taiwan). Each microarray contains 29,187 human genome probes and 1088 experimental control probes. Detailed description of microarray procedures and whole genome gene & probe lists are available from
The full length REG4 coding region was amplified by PCR using the primers listed in Materials and Methods S1, and subcloned into p3XFLAG-CMV-7.1 (Stratagene, Santa Clara, CA) or pcDNA3.1 (Invitrogen) vectors. The REG4 constructs were transiently transfected into PC3shADAM9 using lipofectamine 2000 (Invitrogen) following the standard protocol. Overexpressed REG4 was confirmed by immunoblotting (R&D System). Secreted REG4 in the conditioned medium was collected and concentrated using centricon YM-3 centrifuge filters (Millipore).
The animal studies were carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Instituional guidelines and an Animal Research Committee of China Medical University was followed for the mouse studies (IACUC# 99-68-N). All surgery was performed under Zoletil (tiletamine-zolazepam) at a dose of 20 mg/kg anesthesia by intraperitoneal injection, and bioluminescence imaging was taken under control by Isoflurance anesthesia. All efforts were made to minimize suffering.
All data from the
To validate the role of ADAM9 in prostate cancer progression, we stably downregulated ADAM9 expression in androgen-independent, bone metastatic PC3 prostate cancer cells and androgen-dependent LNCaP prostate cancer cells with either retroviral or lentiviral vectors carrying ADAM9-specific shRNA. The resultant ADAM9 knockdown cell lines PC3shADAM9 and LNCaPshADAM9, which showed an approximately 88% and 76% reduction respectively in ADAM9 protein expression when compared with either empty vector (pSM2C retroviral vector) or control shRNA (lentiviral-shGFP) (
Cell proliferation decreased starting 2 days after seeding (1×104 cells/well) of ADAM9-knockdown PC3 (a) and LNCaP (b) (* Student’s t-test,
To assess the growth difference between ADAM9-proficient and ADAM9-deficient cancer cells in bone, the primary metastatic site for prostate cancer, luciferase-tagged PC3shGFP and PC3shADAM9 were intratibially injected into the same nude mouse but at different legs (
(a) PC3shGFP and PC3shADAM9 were injected into the left and right tibia of the same mouse (n = 10). (b) Bioluminescence imaging at 40 days post-inoculation. Black arrows show the only signal detected in a PC3shADAM9 bone lesion. (c) Computerized radiographic scanning of lesion area. Upper image: Bone destruction is extensive in the left leg compared to the right leg; Lower image: mouse with a destructive bone lesion in the left leg and no lesion in the right leg. (d) Histological imaging showing that both trabecular and cortical bone was replaced by the PC3shGFP tumor. In contrast, trabecular and cortical bone was still intact in the PC3shADAM9 tumor (H & E). Masson’s trichrome staining indicated only traces of bone left in PC3shGFP compared to PC3shADAM9 tumors (Trichrome). (e) TRAcP analysis of osteoclast activity. Arrow indicates osteoclasts observed in a PC3shGFP tumor. Red highlight indicates loci of 40× enlarge.
Since, PC3 cells are known to induce osteolytic lesions
To examine whether shRNA-mediated inhibition of ADAM9 expression represents a promising strategy for prostate cancer gene therapy, PC3 cells were subcutaneously inoculated into both flanks of nude mice. The tumors were allowed to grow to a size of approximately 200 mm3. Ten of fifteen mice that had PC3 tumors on both flanks were selected to receive the intra-tumoral injection of 1×105 cfu lentiviral vectors (LV) carrying shGFP (left site flank) or shADAM9 (right site flank) weekly for a total of 6 weeks (
(a) Schematic of ADAM9 knockdown therapy experiment. Mice were injected subcutaneously with PC3 cells in both flanks, and tumor size was measured twice a week until the size reached approximately 200 mm3. Mice then received injections of either PBS into both tumor sites (n = 5) or shGFP lentivirus into the left side tumor and shADAM9 into the right side (n = 10). Viruses were injected and tumors measured weekly for 6 consecutive weeks. (b) After shADAM9 therapy, tumor volumes did not increase compared to shGFP therapy or PBS controls (**
To determine the cellular responses associated with LV-shADAM9 therapy, tumor proliferation (Ki-67 index,
To further delineate the mechanism of LV-shADAM9 therapy-induced tumor growth inhibition, the cell cycle distribution between ADAM9-proficient PC3shGFP and ADAM9-deficient PC3shADAM9 cells cultured under serum-starvation or serum-supplemented (5% FBS) conditions were analyzed. While PC3shGFP and PC3shADAM9, under normal conditions, did not show a change in the cell cycle profiles, the S-phase population was significantly decreased with time (from 24 to 48 hr) in serum-starved PC3shADAM9 cells as compared with that in PC3shGFP. This reduction was accompanied by the increased percentage of cells in the G1 phase (
Starvation reduced the S-phase cell population in ADAM9 knockdown cancer cells compared to that of controls in PC3 (a), LNCaP (b). (c) Expression levels of p21Cip1/WAF1, p27Kip1, and cyclin D1 increased in shADAM9 cells under starvation stress conditions in PC3 cells. (d) Expression levels of p21Cip1/WAF1 and p27Kip1 increased in LNCaPshADAM9 under both normal and starvation conditions.
Owing to our previous finding that ADAM9 is a stress-responsive protein that may support prostate cancer cell survival and progression under intense oxidative stress conditions
(a) PC3shGFP and PC3shADAM9 cells were cultured in either 5% FBS or under serum starvation for 24 hours, followed by treatment with 0.5 µM MitoSOX fluorescence. Equal intensity of fluorescence was used in each image capture for quantification of superoxide levels. (b) Elevation of endogenous superoxide concentrations level can be detected during serum starvation in both PC3shGFP and PC3shADAM9. (c) Endogenous hydrogen peroxide levels did not differ between cells cultured in 5% FBS and under serum starvation.
To further determine the underlying mechanism for shADAM9-induced G1 phase growth arrest under stress conditions, cDNA microarray analysis of differential gene expression was conducted in PC3shADAM9 cells. To exclude any off-target effects of shRNA on ADAM9 driven genes, two sets of PC3shADAM9 cell line, PC3shADAM9-17130 and PC3shADAM9-46980, which were established with two distinct shRNA constructs by a retroviral system and a lentiviral system (
(a) Quantitative PCR confirmed the decrease of REG4 mRNA and increase of CD33 mRNA. (b) Overexpression of REG4 decreased p21Cip1/WAF1 expression under FBS starvation conditions in PC3shADAM9. (c) The starvation-induced increase in superoxide levels was reversed by overexpression of REG4 in PC3shADAM9. (d) The strongest inhibition of REG4 expression by shREG4 lentiviral vectorwas observed in PC3shREG4-225663 by RT-PCR. (e) PC3shREG4-225663 reduced the S-phase and increased the G1-phase populations under starvation conditions.
Androgen-deprivation therapy is generally the initial treatment for men with advanced prostate cancer. Although patients have high response rates to the initial hormone therapy, nearly all of them eventually develop progressive and metastatic castrate-resistant disease. Docetaxel in combination with prednisone is now regarded as the standard first-line chemotherapy in patients with symptomatic castration-resistant prostate cancer (CRPC)
In the present study, we demonstrated the ability of ADAM9 shRNA impaired androgen-independent prostate cancer PC3 tumor formation and cancer-induced osteolysis in a skeletal metastasis xenograft model. We also presented data indicating the feasibility of using
The regenerating islet-derived (REG) family is a group of small secretory proteins that belong to the C-type lectin superfamily. The physiological function of REG family is mainly involved in proliferation, differentiation and inflammation of cells in the digestive system
In conclusion, the findings presented here thus establish that ADAM9 is essential for cancer progression during the transition to malignancy. Loss of ADAM9 in prostate cancer cells results in the impairment of proliferation and the osteolytic reaction, as well as causing G1/G0 cell cycle arrest. We identified the G1-to-S phase negative regulators p21Cip1/WAF1and p27Kip1 as downstream targets of the ADAM9-REG4 signaling pathway. In a tumor microenvironment, this pathway couples the stress-induced production of ROS to proliferation, G1 phase arrest, and cellular senescence induced by combination therapies.
(PDF)
(a) Immunoblotting validated ADAM9 expression and revealed the knockdown of ADAM9 by Retroviral shADAM9-17130. Vector, pSM2C, were also transfected into PC3 and used as negative control. To select knockdown clone, several clones were also examined. Clone 7 was selected and used in most of our studies. (b) Immunoblotting confirmed knockdown effects of ADAM9 gene in PC3 using lentivral knockdown systems. 46978 and 46980 were selected and shGFP were used as control studies. (c) Immunoblotting confirmed the knockdown of ADAM9 in LNCaP.
(PDF)
(PDF)
(PDF)
(PDF)
(PDF)
(DOCX)
We thank MedcomAsia for their editorial assistance, Miss Li-Chin Wu for TRAcP staining and Drs. Yun-Lung Yu and Wei-Chien Huang for supporting materials and data discussion.