Tissue samples and RNA isolation
Frozen OS tissue samples and normal bone samples (femur/ tibia) of similar age group individuals were obtained through the tissue procurement facility at the University of Minnesota (Supplementary Table 1
). Total RNA was isolated from 75-100 mg of frozen tissue using the miRvana total RNA isolation kit (Ambion Inc, Austin TX, USA) following the manufacturer's protocol. RNA was quantified using the Nanodrop 8000 (Nanodrop Technologies LLC, Wilmington, DE, USA). The quality of the RNA was tested on a 1.2% formaldehyde agarose gel with ethidium bromide staining, and RNA integrity was analyzed using a Nano Labchip (Agilent). Samples with RNA index number (RIN) values of 6 or more were included in this study.
Whole genome miRNA expression profiling
The miRNA expression patterns of OS samples were profiled using the human Illumina miRNA microarrays with 1135 miRNA assay probes (Illumina Inc, San Diego, CA) following the manufacturer's instructions [22
]. The array matrix was imaged using an Illumina BeadArray Reader, which measured the fluorescence intensity at each addressed bead location. Intensity files were analyzed using BeadStudio version 3, and expression levels were converted to an average Beta value. Data analysis was carried out based on the criteria mentioned in Sarver 2010 [23
cDNA was quantified with the miRscript SYBR green detection kit (Qiagen) using an miRNA-specific forward primer and a universal primer, following the manufacturer's instructions, in an ABI 7500 optical cycler (Applied Biosystems). The oligonucleotide primer sequence used for analysis is provided in Supplementary Table 2
. U6 snRNA and
GAPDH were used as controls for miRNA and mRNA qRT-PCR analyses, respectively. Threshold cycle (Ct) values calculated by the SDS v1.2.1 software (Applied Biosystem, USA) were exported and subjected to statistical analysis. Cycle threshold values obtained from duplicate reactions were subjected to statistical analysis, and expression was calculated following the comparative Ct method [24
]. qRT-PCR reactions were carried out in triplicates and the average values were plotted as mean +_SD values.
Saos2 and U2os cells were grown in McCoy's5A medium (Lonza) with 15% fetal bovine serum. Hos cells were grown in Iscove's Modified Dulbecco's Medium (Lonza) with 15% fetal bovine serum, and HEK 293 cells were grown in DMEM (Thermofisher) with 15% fetal bovine serum. All cells were grown at 37°C with 5% CO2.
Luciferase reporter assays
A PSGG luciferase reporter construct containing the firefly luciferase sequence and the 3′UTR of cMYC cloned into Nhe I and Xho I sites were purchased from Switchgear genomics (Menlo Park, CA). HEK 293 cells (0.8 × 106 cells) were transfected with 500 ng of PSGG plasmid and 75 ng of miRNA over-expression plasmid, pRL-TK (Promega, Madison, WI), with or without double stranded RNA oligonucleotides (5.0 μl of 10 μM solution) in duplicate using Lipofectamine 2000 and following the manufacturer's protocol (Invitrogen). Cells were collected after 48 hrs and assayed using the dual luciferase reporter system (Promega) following the manufacturer's protocol. Firefly luciferase values were normalized using Renilla luciferase expression, and average values were calculated from 2 sets of experiments. Cells transfected with pSGG plasmid containing only firefly luciferase were used as controls.
Nucleofection of miRNA mimics
miRNA-specific miRNA mimics were purchased from a commercial manufacturer (Dharmacon). Cells grown in T-75 flasks were trypsinized, and 0.1 × 106 cells were nucleofected with double-stranded RNA oligonucleotides (10 nM final concentration) in 99.3 μl of nucleofection solution V using Nucleofector II equipment (Lonza), in duplicate. Nucleofected cells were transferred to 1.5 ml microcentrifuge tubes, incubated in a 37°C incubator for 20 min, transferred to 12-well plates after changing the media 6 hrs after treatment the cells were incubated up to 72 hrs at 37°C with 5% CO2. Programs Q-01, D-024 and X-001 were used to nucleofect HEK293, Saos2 and U2os cells, respectively. Cells were subjected to qualitative analysis at 48 hrs after nucleofection, with sham treated and nonnucleofected cells as controls. The efficiency of nucleofection was calculated by nucleofecting cells with 2.0 μg of pmaxGFP plasmid (Lonza) and monitoring the expression 20 hrs after nucleofection.
Apoptosis assay and ZVAD-fmk treatment
The rate of apoptosis was determined by flow cytometry following the quick method of Riccardi and Nicoletti, 2006 [25
]. Briefly, cells were stained with subsaturation quantity of propidium iodide(PI) as recommended and fluorescence data was collected in logarithmic scale. PI-stained cells (n=10,000) were subjected to flow cytometry analysis 48 hrs after treatment in a FACS Calibur (BD Biosciences, San Jose, CA) using the FL2 channel. As the intake of PI is proportional to the fraction of cells with damaged membranes resulted by apoptotic events, PI positive cells with the fluorescence value of more than 2 logs relative to sham and control cells were scored positive. For the apoptosis inhibition experiments, cells were continuously incubated with 20 μM of ZVAD-fmk (#550377, BD Biosciences) immediately after nucleofection until analysis.
RNAi for cMYC
Saos2 cells (0.1 × 106) were nucleofected with 500 ng of plasmid DNA expressing shRNA against cMYC (origene #TGS311323, construct GI345288 and GI345287) with scrambled controls (#TR30013) using the programs D-024, in duplicates.
Western blot for cMYC
Treated cells were harvested and lysed in lysis buffer (25 mM Hepes pH 7.6, 300 mM NaCl, 1.5 mM Mgcl2, 0.2 mM EDTA and 0.1 % Triton x-100) containing a protease inhibitor cocktail (Sigma 8340). Protein was estimated by Bradford dye binding with a colorimetric assay (BioRad 500-0006). Total cellular protein (15 μg) was resolved on a precast 10% Tris-Hcl Criterion 18-well gel (Biorad) at 100 V (300 mAmp) for the first 15 min and later at 200 V (300 mAmp) for 1 hr. The gel was wet-transferred to a PVDF membrane for 1 hr, blocked with PBST containing 5% instant dry non-fat milk powder for 30 minutes at room temperature and incubated with anti-cMYC (9E10) mouse monoclonal antibody (Sc40, Santacruz) at 1:200. All primary and secondary antibody incubations were carried out with 5% milk powder at 4°C for overnight and 1.0 hr respectively, after which the membrane was washed with PBST for 5 minutes (×5 times). The washed blot was incubated with goat anti-mouse IgG-AP (Santa Cruz sc-2008) diluted to 1:5000 and developed with ECF reagent (GE Healthcare 9RPN-5785). The membrane was stripped using western blot stripping buffer (Pierce 21059), incubated with rabbit Anti-GAPDH (Sigma G9545) at 1:10000, washed and incubated with goat anti-rabbit IgG-AP (Santa Cruz sc-2007) at 1:5000 and developed as described above.
Osteoblast cell culture and differentiation assay
Human (hFOB1 19, ATCC CRL-11372) was purchased from ATCC and grown as recommended [26
]. For osteoblast differentiation assays, hFOB and MC3T3 cells were seeded separately at 80% confluence (0.8×106
cells/ well) in 6-well plates with or without 50 μg/mL ascorbic acid and 10 mM β-glycerol-phosphate for 3 weeks. Media was replaced every alternate day and cells were assayed for differentiation by staining with alizarin red dye following the protocol of the manufacturer (osteogenesis assay kit, #ECM815, Millipore). Briefly the cells were fixed in 4% formalin for 10 minutes, washed 4 times with PBS and stained with alizarin red for 20 minutes, washed with PBS and imaged at 100× magnification.
Cell proliferation assay
CellTiter 96 Aqueous One Solution Cell Proliferation Assay (Promega, Madison, WI) was used according to manufacturer's instruction. Briefly, 50 μl of CellTiter 96 Aqueous One Solution Reagent was added into each well of the 24-well assay plates containing cells in 1ml of culture medium. After 1 hr of incubation, 100 μl of culture medium from each well were transferred to a white opaque 96-well plate (Falcon, BD) for measuring absorbance at 490 nm using Synergy2 microplate reader (BioTek).
Cell migration assay
24-well cell migration assay (Cultrex, Trevigen, Gaithersburg, MD) were used to investigate cell migration following the manufacturer's instructions with minor modifications. Briefly, 1 × 105 cells were seeded into the top inserts after being starved in DMEM containing 0.5% FBS for 24 hrs. DMEM containing 10% FBS was added into the receiver wells to trigger cell migration. After 24 hours incubation, both top inserts and bottom receiving wells were washed with PBS, and Calcein AM/ Cell dissociation solution was added into the bottom chamber of each well. After 1 hr of incubation, top inserts were discarded and the Calcein AM /cell dissociation solution with cells detached from the bottom chamber were transferred to a black 96 well plate (Fluotrac 200, Greiner Bio-One, Monroe, NC 28110), which was then read by Synergy 2 (BioTek) at 485nm excitation, 520nm emission. Standard curves were generated to convert relative fluorescence units into cell number.
Cell invasion assay
Transwell Inserts coated with Cultrex basement membrane extract (Corning, Cat No. 3458) were used for cell invasion assay according to the instructions of manufacturer. Briefly, human osteoblasts nucleofected as mentioned elsewhere were starved in DMEM with 0.5% FBS for 24hrs before being seeded into the inserts (105 cells/insert). DMEM with 10%FBS was added into the receiver wells to promote cell invasion. After 24 hours incubation, both top inserts and bottom receiving wells were washed with PBS, and Calcein AM/ Cell dissociation solution was added into the bottom chamber of each well. After 1.0 hr of incubation, top inserts were discarded and the Calcein AM /cell dissociation solution with cells from the bottom chamber were transferred to a black plate (Fluotrac 200, Greiner Bio-One, Monroe, NC 28110), which was then read by Synergy 2 (BioTek) at 485nm excitation, 520nm emission. Standard curves were generated to convert relative fluorescence units into cell number.
Genomic DNA was isolated from patient samples using Gentrapure DNA isolation kits (Qiagen, Valencia, CA) following the manufacturer's protocol.
Array Comparative Genome Hybridization
Hybridizations were performed using a Human 1Mb BAC array (HuCGH v1.1.1) (Sanger Institute, UK) (Fiegler et al., 2003), which consisted of over 3,000 clones spaced at approximately 1 Mb intervals across the entire genome, and the Agilent Human CGH 8×60K microarrays, which contain unique 60-mer oligonucleotide probes (~50 kb resolution) and 60,000 features (Agilent Technologies). Reference DNA consisted of genomic DNA from unaffected female or male blood sample mixes (Promega). The Bioprime Labeling Kit (Invitrogen) was used to label 1 μg of tumor and reference DNA for the 1 Mb BAC array platforms with Cyanine-3-dCTP and Cyanine-5-dCTP, respectively. The probe mixture containing sex-mismatched labeled Cy3 and Cy5 probes and 135 μl (1 mg/ml) of human Cot1 DNA (Invitrogen) was precipitated at -80°C for 45 min and resuspended in 44 μl of hybridization buffer (50% deionized formamide, 2.5% dextran sulfate, 2X SSC, 0.1% Tween 20, and 0.01 mM Tris HCL, pH 8.0). The probe mixture was then denatured at 72°C for 10 min and incubated at 37°C for 1 hr. Next, the mixture was applied to the microarray for 48 hr at 37°C. Following posthybridization stringency washes, the arrays were scanned at 10 μm resolutions (using a Perkin Elmer ScanArray Express), and the fluorescence intensity at each genomic locus was quantified and used to calculate the log2 test: reference ratio for each locus on the array.
For the Agilent 8×60K human aCGH platform, 500 ng of tumor and reference DNA were labeled using the Bioprime Labeling Kit (Invitrogen) and then combined in the presence of 2 μl (1 mg/ml) of human Cot1 DNA (Invitrogen). The sex-mismatched probe mixture was then precipitated at -80°C for 45 min and resuspended in 22 μl of HPLC water, 5.5 μl of Agilent 10X blocking agent, and 27.5 μl of Agilent HI-RPM hybridization buffer. This mixture was denatured at 95°C for 5 min and incubated at 37°C for 30 min before application to the array. Hybridization was carried out for 24 hr at 65°C using an Agilent microarray hybridization chamber in a rotating oven. Arrays were washed according to the manufacturer's recommendations, air dried, and scanned at 3 μm resolution using an Agilent DNA Microarray Scanner with SureScan High resolution technology (Agilent Technologies).
DNA copy number analysis
To detect DNA copy number aberrations, using the 1 Mb aCGH platform, fluorescence intensities were quantified, threshold limits of log2 ratio values of tumor DNA to reference DNA were set as 1.5:1 (copy number gain) and 0.667:1 (copy number loss), and data were analyzed by Bluefuse for microarray software (Cambridge, UK). Bluefuse analysis was performed using the aCGH Smoothing Algorithm, which contains a clustering-based approach to identify homogenous groups of probes that define regions of gain or loss. The Agilent 8×60K arrays were assessed for data quality by the ‘Quality Metrics’ report produced in Agilent's Feature extraction software (v10.5) (Agilent Technologies). The data report from Feature Extraction was then imported into Agilent Genome Workbench Software (v5.0) (Agilent Technologies). To produce aberration calls and also assess overall array quality, an estimate of the noise was obtained by calculating the spread of the ratio differences between consecutive probes (DLRspread) along all chromosomes. The ADM-2 algorithm was then used to scan for chromosomal intervals of any length, for which the mean interval ratio was significantly different from zero and had ratios outside the threshold standard deviations. The threshold value for the ADM-2 algorithm was set 6 to 6 to reduce inherent sample noise without losing true consistent intervals along each chromosome. Also, to detect the extent of the copy number aberrations, we set the mean aberration log2 ratio thresholds at or above 0.5.