Supplemental Experimental Methods

Supplemental Experimental Methods

Microarray

Primary OS samples were homogenized in Trizol reagent (Invitrogen) using an electric tissue homogenizer. Cells (both OS cell lines and primary in vitro differentiated osteoblasts) were lysed in Trizol reagent. All RNA was purified using RNeasy kit (Qiagen) and ~5-30ug of total RNA was submitted for expression profiling. Gene expression was measured using Affymetrix A430 2.0 oligonucleotide arrays. cRNA was non-competitively hybridized to an array and after hybridization scanned to acquire the data files. CEL files were processed and normalized using dChip software (Li et al. 2001) (http://biosun1.harvard.edu/complab/dchip/). All microarray samples were processed by the Dana-Farber Cancer Institute Microarray Core Facility in accordance with the manufacturer’s instructions (http://chip.dfci.harvard.edu/).

GSEA lists

Common human OS cytogenetic rearrangements were identified from the literature (Bridge et al. 1993; Batanian et al. 2002). The consensus genes in these regions were identified from the UCSC Genome browser and Ensembl human genome annotation. Lists were generated containing the HUGO gene symbols of genes from annotations of listed regions. Redundancy was collapsed. Gene lists were formatted and GSEA was performed (http://www.broad.mit.edu/gsea/). Primary OS samples and OS cell lines were compared to in vitro differentiated primary osteoblasts. Synovial sarcoma samples were compared to WT muscle (Haldar et al. 2007). A complete list of human genes in all regions analyzed can be found in Supplemental Table 5. GSEA plots for the comparison of primary OS tumor vs osteoblast; OS cell line vs osteoblast; synovial sarcoma vs muscle are in Supplemental Figures 7, 8 and 9 respectively.

Isolation of single cell suspensions

Primary tumor samples were digested to obtain single cell suspensions prior to flow cytometry, cell line isolation. Primary samples were coarsely minced with a sterile scalpel blade followed by digestion for 45-60 minutes at 37°C with agitation in digestion media (DMEM:F12 media supplemented with 2mg/mL collagenase A, 0.1mg/mL hyaluronidase, 60U/mL nystatin, 0.1mg/mL gentamicin). Following digestion the cell suspension was filtered and either prepared for flow cytometry staining or plated in aMEM media supplemented with 10% FCS (Hyclone, non-heat inactivated). Cell lines were maintained in aMEM media supplemented with 10% FCS (Hyclone, non-heat inactivated).

Calvarial osteoblast populations were isolated using a modified version of the protocol described by Gu et al., (Gu et al. 2005a; Gu et al. 2005b) using sequential digestion of primary calvaria in 1mg/mL collagenase 1 and collagenase 2 mix (1:3 ratio of collagenase 1 to 2) in aMEM media supplemented with 0.1% BSA and 25mM Hepes at 37C. Each digestion step was performed for 15 minutes with shaking and then the calvari was transferred to fresh digestion media and then digestion was continued. Fractions 3 to 6 were pooled and utilized as primary calvarial osteoblasts.

Kaplan-Meier Survival Analysis

K-M survival plots and statistical analysis (Log Rank Test) were prepared using Prism software.

Flow Cytometry

Single cell suspensions from femur derived compact bone and primary tumors were prepared as described above or OS cell lines from culture and stained with antibodies: CD45 (clone A20 and 104, eBioscience), Ter119 (eBioscience), CD11b (M1/70, eBioscience), CD31 (390, eBioscience), CD51 (RMV-7, Pharmingen), Sca-1 (D7, eBioscience). Flow cytometry was performed on a FACSCalibur interfaced with CellQuest software. FACS analysis was performed on 13 tumors composed of 10 primary tumors and 3 secondary tumors. 6 cell lines by FACS were also analyzed.

In vitro differentiation of primary osteoblasts

Primary osteoblasts were isolated from crushed, digested (as described above) femur/tibia/iliac crest bones. Hematopoietic cells were filtered out (40mm cell strainer) and the remaining bone was digested. Following digestion the samples were filtered (40mm cell strainer) and plated in aMEM media supplemented with 20% FCS. Once cells reached approximately 80% confluence, any contaminating macrophages/hematopoietic cells were immunomagnetically depleted using affinity purified rat anti-mouse anti-Cd11b and anti-CD45 (eBioscience) and sheep anti-rat Dynabeads (Invitrogen). Hematopoietic depleted cells were plated in osteoblast differentiation media (aMEM media supplemented with 20% FCS, 10-8M dexamethasone, 100mM L-ascorbic acid and 4mM inorganic phosphate) with media changed every 3 days for 3 weeks.

Doxycycline treatment

Where indicated animals were treated with doxycyline in the drinking water at a dose of 2mg/mL (protected from light). Doxycycline was administered on establishment of the breeding pair and pups weaned and maintained on doxycyline until 4 weeks of age.

Quantitative PCR analysis

RNA was prepared from primary tissues and cell lines as described in Microarray section. All RNA was DNase treated prior to reverse transcription. cDNA was prepared using Superscript III reverse transcriptase as directed by the manufacturer (Invitrogen). Quantitative real-time PCR (qPCR) was performed using iQ SYBR Green mix (Bio-Rad) on a Bio-Rad iCycler PCR machine.

Genomic DNA was prepared from primary tumor samples and control tissue from the corresponding animal using a Puregene DNA Purification Kit (Gentra Systems) using the manufacturers protocol. Genomic DNA was isolated from OS cell lines using a DNeasy Tissue Kit (Qiagen) as described by the manufacturer.

Primers for qPCR

Primers for qPCR were either obtained from published protocols or from Primer Bank (Wang et al. 2003)

Gene / Forward / Reverse / Source
pRb / gaagcaactctggggcatct / gctcagtaaaagtgaatggcatc
p53 / aagatccgcgggcgtaa / catcctttaactctaaggcctcattc
Runx2 / tttagggcgcattcctcatc / tgtccttgtggattaaaaggacttg / (Kieslinger et al. 2005)
Osterix / actcatccctatggctcgtg / ggtagggagctgggttaagg / (Kieslinger et al. 2005)
Alkaline Phosphatase / cacgcgatgcaacaccactcagg / gcatgtccccgggctcaaaga / (Kieslinger et al. 2005)
Collagen 1a1 / caccctcaagagcctgagtc / gttcgggctgatgtaccagt / (Kieslinger et al. 2005)
Ebf2 / aagtgaaggctggaccacag / attgtgtggtgtgccgtaga / (Kieslinger et al. 2005)
BMPR1a / TGGCTCGTCGTTGTATTACAGG / GCTCCAACTTACTTCATCGCT / (Wang et al. 2003)
BMP4 / TTCCTGGTAACCGAATGCTGA / CCTGAATCTCGGCGACTTTTT / (Wang et al. 2003)
MMP13 / CTTCTTCTTGTTGAGCTGGACTC / CTGTGGAGGTCACTGTAGACT / (Wang et al. 2003)
PTHR1 / TGGCTCGTCGTTGTATTACAGG / GCTCCAACTTACTTCATCGCT / (Wang et al. 2003)
Osteoprotegerin / gtgctgcagttcgtgtgttt / cactgcacagtgaggaggaa / (Kieslinger et al. 2005)
Bone Sialoprotein / taccggccacgctactttctttat / gaccgccagctcgttttcatcc / (Kieslinger et al. 2005)
Osteocalcin / accctggctgcgctctgtctct / gatgcgtttgtaggcggtcttca / (Kieslinger et al. 2005)
FGFR2 / GTGCTTATTGGGGAGTATCTCCA / GATCCAAGTTTCACTGTCTACCG / (Wang et al. 2003)
MAPK7 / GTGGGGGACGAGTACGAGAT / TGGTCACCACATCAAAAGCAT / (Wang et al. 2003)
c-Fos / CCCATCCTTACGGACTCCC / GAGATAGCTGCTCTACTTTGCC / (Wang et al. 2003)
Bub3 / GATGGCATCTCCTCGGTTAAG / AATTGGCGGGCACATCGTAG / (Wang et al. 2003)
Topo3a / ATCTTCCCGGTCACCCTCTTA / CAGAGAACTTTTCGCACTCCTC / (Wang et al. 2003)
Mdm2 / TGTCTGTGTCTACCGAGGGTG / TCCAACGGACTTTAACAACTTCA / (Wang et al. 2003)
p15 / TCAGAGACCAGGCTGTAGCAATC / CCCCGGTCTGTGGCAGAA / (Passegue et al. 2005)
p16 / CCCAACGCCCCGAACT / GCAGAAGAGCTGCTACGTGAA / (Krishnamurthy et al. 2004)
p19 / GCAGAAGAGCTGCTACGTGAA / TGAGGCTAGAGAGGATCTTGAGA / (Krishnamurthy et al. 2004)
Cyclin E1 / GCAGCGAGCAGGAGACAGA / GCTGCTTCCACACCACTGTCTT / (Passegue et al. 2005)
c-Myc / GGAATTTTTGTCTATTTGGGGACAG / TAGTCGAGGTCATAGTTCCTGTTGG / (Passegue et al. 2005)

Supplemental References:

Batanian, J.R., Cavalli, L.R., Aldosari, N.M., Ma, E., Sotelo-Avila, C., Ramos, M.B., Rone, J.D., Thorpe, C.M., and Haddad, B.R. 2002. Evaluation of paediatric osteosarcomas by classic cytogenetic and CGH analyses. Mol Pathol 55(6): 389-393.

Bridge, J.A., Bhatia, P.S., Anderson, J.R., and Neff, J.R. 1993. Biologic and clinical significance of cytogenetic and molecular cytogenetic abnormalities in benign and malignant cartilaginous lesions. Cancer Genet Cytogenet 69(2): 79-90.

Gu, G., Hentunen, T.A., Nars, M., Harkonen, P.L., and Vaananen, H.K. 2005a. Estrogen protects primary osteocytes against glucocorticoid-induced apoptosis. Apoptosis 10(3): 583-595.

Gu, G., Mulari, M., Peng, Z., Hentunen, T.A., and Vaananen, H.K. 2005b. Death of osteocytes turns off the inhibition of osteoclasts and triggers local bone resorption. Biochem Biophys Res Commun 335(4): 1095-1101.

Haldar, M., Hancock, J.D., Coffin, C.M., Lessnick, S.L., and Capecchi, M.R. 2007. A conditional mouse model of synovial sarcoma: insights into a myogenic origin. Cancer Cell 11(4): 375-388.

Kieslinger, M., Folberth, S., Dobreva, G., Dorn, T., Croci, L., Erben, R., Consalez, G.G., and Grosschedl, R. 2005. EBF2 regulates osteoblast-dependent differentiation of osteoclasts. Dev Cell 9(6): 757-767.

Krishnamurthy, J., Torrice, C., Ramsey, M.R., Kovalev, G.I., Al-Regaiey, K., Su, L., and Sharpless, N.E. 2004. Ink4a/Arf expression is a biomarker of aging. J Clin Invest 114(9): 1299-1307.

Li, C. and Wong, W.H. 2001. Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc Natl Acad Sci U S A 98(1): 31-36.

Passegue, E., Wagers, A.J., Giuriato, S., Anderson, W.C., and Weissman, I.L. 2005. Global analysis of proliferation and cell cycle gene expression in the regulation of hematopoietic stem and progenitor cell fates. J Exp Med 202(11): 1599-1611.

Wang, X. and Seed, B. 2003. A PCR primer bank for quantitative gene expression analysis. Nucleic Acids Res 31(24): e154.

Supplemental Movie and Figure Legends

Supplemental Movie 1. Multifocal disease as visualized using F18 sodium fluoride.

Supplemental Movie 2. Jaw tumor present at 8 weeks of age in an Osx-Cre+p53fl/flpRbfl/fl.

Supplemental Movie 3. Osx-Cre- control at 8 weeks of age.

Supplemental Movie 4. Tibial tumor present at 12 weeks of age in an Osx-Cre+p53fl/flpRbfl/+.

Supplemental Figure 1. Metastatic Disease.

Example of lung and liver metastasis as visualized by microPET.

Supplemental Figure 2. Delayed gene inactivvation until 4 weeks of age leads to the development of a completely metastatic model.

a) Kaplan-Meier survival analysis of Osx-Cre+p53fl/flpRbfl/fl animals that were not treated with doxycycline (black line, Cre active during development from time of Osx expression; n=24) and a cohort of Osx-Cre+p53fl/flpRbfl/fl animals treated with doxyxycline until 4 weeks of age (repressing Cre expression until 4 weeks of age; n=6). b) Summary of disease observed in the doxycycline treated animals. c) Photo of lung metastasis present in mouse #493.

Supplemental Figure 3. Multifocal Disease.

Two representative examples of multifocal disease revealed by microPET. Location of lesion is indicated.

Supplemental Figure 4. Spectral Karyotyping reveals aneuploidy of mouse OS.

a) Representative metaphases of tumors 188, 330 and 331 respectively. Arrows highlight translocations present in tumor 330 affecting chromosome 8 and 19 with higher magnification images as indicated by color. b) Summary of results for indicated tumors.

Supplemental Figure 5. Metagene analysis with entire human sarcoma data set (exluding tumor defined as sarcoma in set).

Metagene projection of entire human sarcoma data set and mouse primary OS. mOS associates with human OS.

Supplemental Figure 6. mOS cell lines show a similar transcriptional profile to human OS.

Metagene projection of human sarcoma data and mouse primary OS, mOS cell lines, mOS secondary tumors and mOsteoblasts. All mOS samples show a close association with human OS.

Supplemental Figure 7. GSEA enrichment profiles for the comparison of mOS and osteoblasts using cytogenetic region enrichment analysis.

Supplemental Figure 8. GSEA enrichment profiles for the comparison of mOS cell lines and osteoblasts using cytogenetic region enrichment analysis.

Supplemental Figure 9. GSEA enrichment profiles for the comparison of mSynovial Sarcoma and primary muscle using cytogenetic region enrichment analysis.

Supplemental Figure 10. Evidence of loss of heterozygosity of p53 in tumors.

Genomic DNA was extracted from donor matched ear tissue and tumor tissue of 3 independent tumors. PCR genotyping was performed and loss of heterozygosity can be observed for p53. We have not observed this for pRb.

Supplemental Figure 11. Comparison of transcriptional profiles of tumors derived from different initial genotypes.

The mean log intensity of gene expression of all 45,000 probes on the Affymetrix A430 2.0 microarray were compared to determine the correlation in gene expression profiles amongst tumors derived from different initial genotypes. The correlation coefficient between each data set was derived from the trend line. The correlation of gene expression between samples of the same initial genotype arrayed on different days was 0.944 and 0.926 respectively. The correlation of gene expression between samples of different genotypes from the pooled array data (derived from 2 separate experiments run independently and arrays at different times) was >0.96 in all cases demonstrating a remarkable conservation of gene expression profiles amongst the tumors independent of the initial genotype.