Characterization of a Novel Radiation-Induced Sarcoma Cell Line

Background

Radiation-induced sarcoma (RIS) is a potential complication of cancer treatment. No widely available cell line models exist to facilitate studies of RIS.

Methods

We derived a spontaneously immortalized primary human cell line, UACC-SARC1, from a RIS.

Results

Short tandem repeat (STR) profiling of UACC-SARC1 was virtually identical to its parental tumor. Immunohistochemistry (IHC) analysis of the tumor and immunocytochemistry (ICC) analysis of UACC-SARC1 revealed shared expression of vimentin, osteonectin, CD68, Ki67 and PTEN but tumor-restricted expression of the histiocyte markers α1-antitrypsin and α1-antichymotrypsin. Karyotyping of the tumor demonstrated aneuploidy. Comparative genomic hybridization (CGH) provided direct genetic comparison between the tumor and UACC-SARC1. Sequencing of 740 mutation hotspots revealed no mutations in UACC-SARC1 nor in the tumor. NOD/SCID gamma mouse xenografts demonstrated tumor formation and metastasis. Clonogenicity assays demonstrated that 90% of single cells produced viable colonies. NOD/SCID gamma mice produced useful patient-derived xenografts for orthotopic or metastatic models.

Conclusion

Our novel RIS strain constitutes a useful tool for pre-clinical studies of this rare, aggressive disease. UACC-SARC1 is an aneuploid cell line with complex genomics lacking common oncogenes or tumor suppressor genes as drivers of its biology. The UACC-SARC1 cell line will enable further studies of the drivers of RIS.

Synopsis

We derived a spontaneously immortalized primary human cell line, UACC-SARC1, from a radiation-induced sarcoma (RIS). Our novel RIS cell line constitutes a useful tool for pre-clinical studies of this rare, aggressive disease.

Keywords: sarcoma, radiation-induced, malignant fibrous histiocytoma

Tissue Harvest

The source of the cell line was a 52-year-old woman treated for high grade ductal carcinoma in situ (estrogen receptor (ER) positive) with lumpectomy and 66 Gray of XRT who developed a bulky RIS within the previously irradiated field 5 years later. As the resulting RIS was not responsive to treatment with neoadjuvant doxorubicin and ifosfamide, the patient underwent resection of her RIS with an extended radical mastectomy with widely negative margins.

Preoperatively, we had obtained informed consent from the patient for the use of residual tumor tissue, per our institution's Human Subjects Protection Program approved tumor biorepository protocol (#06-0609-04). At the time of surgery, we harvested fresh RIS tissue and snap-froze a portion in liquid nitrogen. We obtained whole peripheral blood from the patient, which was processed for DNA isolation.

The fresh sarcoma tissue was placed in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 1% antibiotics/antimycotics (AA) (Invitrogen, Carlsbad, CA). Immediately upon receipt of the specimen, the tissue was minced and digested for roughly 16 hours at 37°C with 1X recombinant pure collagenase/hyaluronidase (Stem Cell Technologies, Vancouver, Canada). The tissue digest was then transferred to a 50ml conical tube, and spun down at 800rpm for 1 minute. The fatty supernatant was discarded. The cell pellet was resuspended and plated in fresh DMEM with 10% FBS and 1% AA.

Culture Conditions

We maintained all cell lines in an incubator with 10% humidity and 5% CO₂ at 37° C. The synovial sarcoma cell lines HS-SY-II[10] and SYO-1[11], and human foreskin fibroblasts (HFF) were obtained from the American Type Culture Collection (ATCC) by the Cranmer laboratory, authenticated by STR genotyping and grown in DMEM with FBS and AA. All experiments were performed when cells reached 75% to 85% confluency, and all cultures were passaged when the cell culture reached around 80% confluence.

DNA Preparation and Short Tandem Repeat (STR) Profiling

To isolate genomic DNA, UACC-SARC1 and MDA-MB-231 cells at passage 20 were trypsinized (0.05% trypsin-EDTA; Gibco/Life Technologies, Grand Island NY), washed with phosphate-buffered saline (PBS), and pelleted by centrifugation. DNA was extracted from the cell lines, whole blood and sarcoma tumor with the QIAamp kit (Qiagen, Valencia, CA). UACC-SARC1 and the parental sarcoma tumor were profiled for amelogenin and 15 STR microsatellite loci using the Identifiler polymerase chain reaction (PCR) kit (Applied Biosystems, Foster City, CA) at the University of Arizona Genetics Core per manufacturer's specifications. PCR products were separated by capillary electrophoresis on an ABI 3730 DNA analyzer. Electropherogram analysis and allelic values assignment were performed using Gene Marker software (Soft Genetics, State College, PA). Heterozygosity (i.e., detection of different alleles) expressed as a percentage was calculated based on any difference in the number of peaks within the 15 STR loci (excluding amelogenin). The fractional allele percentage of similarity was inferred by dividing the number of identical alleles by 32 (the total number of alleles studied) and multiplying the result by 100[12].

IHC and ICC

10⁶ cultured UACC-SARC1 cells at passage 19 and mouse xenografts were fixed in 10% neutral buffered formalin for 24 hours. Specimens were then embedded in paraffin via our institution's Tissue Acquisition and Cellular/Molecular Analysis Shared Service. Routine hematoxylin and eosin (H&E) staining was performed on 3 micron thick tissue sections. Cell cultures for ICC were trypsinized and resuspended in 15ml of PBS with 2% FBS, then spun at 1500 rpm for 5 minutes. Cell pellets were then fixed overnight with 6 mL of cold 10% neutral buffered formalin (VWR, Radnor PA). The cell blocks for staining were then processed and paraffin embedded per the manufacturer's instructions using the ThermoShandon CytoBlock kit (Thermo Scientific, Fremont, CA).

IHC and ICC analyses were carried out with prediluted monoclonal (M) or polyclonal (P) antibodies to: vimentin (M) (Ventana Medical Systems, Inc. [VMSI], Tucson, AZ); osteonectin (secreted protein acidic and rich in cysteine [SPARC]) (M); (Abnova, Taipei City, Taiwan); α₁-antitrypsin (P) (VMSI); α₁-antichymotrypsin (P) (VMSI); CD68 (M) (VMSI); smooth muscle actin (SMA) (M) (Cell Marque, Rocklin, CA); CD34 (M) (VMSI); epithelial membrane antigen (EMA) (M) (VMSI); S100 (P) (VMSI); desmin (M) (VMSI); pancytokeratin 8, 14, 15, 16, 18, and 19 (P) (gift of the Nagle laboratory, mixture of MAK6 and 10:11) ; ER (M) (VMSI); PR (M) (VMSI); CD117 (anti-c-KIT) (M) (VMSI); lysozyme (P) (VMSI); epidermal growth factor receptor (EGFR) (M) (Zymed, South San Francisco, CA); p53 (M) (VMSI); and CD99 (M) (VMSI); PTEN (M) (VMSI), RB1 (M) (Epitomics, Burlingame CA) and Ki67 clone MIB-1(M) (Dako, Carpinteria, CA).

All steps of tissue section processing and staining were performed using a Discovery XT Automated Immunostainer (VMSI) and VMSI-validated reagents. They included deparaffinization, cell conditioning (antigen retrieval with a borate- EDTA buffer), staining with primary antibodies; detection and amplification, using a biotinylated-streptavidin-horseradish peroxidase (HRP) and diaminobenzidine (DAB) system; and hematoxylin counterstaining.

Images were acquired with a Paxcam 3 camera with PAX-it Digital Image Analysis and standardized for light intensity.

Human-Specific Lamin A/C Immunohistochemistry

Tissue sections from RIS and xenografts were de-paraffinized and re-hydrated using standard protocols in the Tlsty laboratory. In order to reduce the non-specific endogenous peroxidase background staining, sections were treated with 3% H₂O₂ for 10 minutes at room temperature. Slides were subjected to antigen retrieval by microwaving for 10 minutes in citrate buffer (pH=6.0) then cooled down for 20 minutes at room temperature. The sections were incubated for 60 minutes at room temperature with a primary monoclonal rabbit antibody against lamin A/C (Epitomics) diluted 1:500 in Antibody Dilution Buffer (Genemed, South San Francisco, CA), then for 30 minutes at room temperature with HRP Polymer (Thermo Scientific). After 4 washes, the sections were developed for 5 minutes using DAB as substrate. The sections were counterstained in Mayer's hematoxylin.

Matrigel Invasion Assay

The metastatic potential of the UACC-SARC1, HFF, SYO-1, and HS-SY-II cell lines were evaluated using the BD BioCoat Matrigel Invasion Chamber in duplicate (BD Biosciences, San Jose, CA) with wells prepared per manufacturer specifications. 2.5×10⁴ cells each were added to the Matrigel Basement Membrane Matrix inserts and control inserts in duplicate and incubated for 24 hours at 37°C. Non-invading cells were removed and the membrane stained with the Hema 3 Staining System (Fisher HealthCare, Houston, TX). Counting of invading cells was facilitated by photographing the membrane though the microscope. Results were quantified as the percent invasion through the Matrigel relative to the migration through the control membrane. All images were counted by a single investigator (WZ). The data was then analyzed using Prism Graph Pad Version 6 (La Jolla, CA) to plot bar graphs and compare cell lines using a one-way ANOVA test for non-parametric data to compare the mean percent invasion for each cell line.

Clonogenicity analysis

The UACC-SARC1 was diluted to contain one cell in 150ul of the culture medium (RPMI + 10%FBS) (verified microscopically), and 144 wells of two 96-well adherent plates were filled with 150ul of the medium containing one cell. The cell growth was checked daily and the cell colony formations were recorded on day 12 under the microscope.

A sphere formation assay was performed by culturing UACC-SARC1 in 6-well ultra-low adherent tissue culture plates with MammoCult Medium (Stem Cell Technologies, Vancouver Canada). Tertiary tumorspheres were generated by serial passaging non-adherent cultures as per the Stem Cell Technologies technical bulletin for culture of breast cancer cell lines as tumorspheres.

Cytogenetic Analysis of UACC-SARC1

Karyotype analysis was conducted at the Cytogenics Laboratory in the Department of Pathology of the University of Arizona. A T-25 flask (Thermo Fisher Scientific, Waltham MA) of UACC-SARC1 cells in complete medium was trypsinized then re-plated onto 8 separate culture dishes in order to provide mitoses for cytogenetic analysis. 48 hours after trypsinization, processed cells were G-banded then analyzed at 1000X magnification under a light microscope.

CGH

An Agilent (Santa Clara, CA) human CGH array with 169,628 features was processed in an R environment for statistical computing and graphics rendering[13]. Raw hybridization signals were normalized using the Bioconductor[14] package snapCGH[15]. Genomic regions with amplifications or deletions were identified using an algorithm implemented in the DNAcopy package[16]. P values less than 0.01 were considered significant. Positive regions were defined as regions covered by at least 5 probes. One microgram of DNA from RIS tumor, UACC-SARC1, and peripheral blood were compared to reference normal female human DNA. Features were annotated using the Human May 2004 (NCBI Build 35/hg17) assembly.

Sequencing

DNA derived from the UACC-SARC1 cell line, the RIS tumor, peripheral blood, and reference human male DNA were analyzed at 740 known mutation hotspots using the Ion Ampliseq Cancer Panel consisting of 190 amplicons derived from 46 genes (Ion Torrent Systems, Guilford CT). Genomic DNA was amplified, bar-coded, pooled and sequenced on the Ion Torrent Personal Genome Machine using primers and reagents according to manufacturer's protocols. Briefly, 10 nanograms of the same genomic DNA used for aCGH described above were amplified in a multiplexed PCR reaction. The primer-derived DNA was digested from the amplicons and bar-coded using sequencing adaptors ligated to the amplicon ends. The amplicon pools from the four samples were diluted and mixed in equimolar amounts to produce a library. Sequencing templates were generated from the library by emulsion PCR and template positive beads were selected and loaded onto an Ion Torrent 318 chip for sequencing. Data were analyzed using Ion Torrent Server Suite version 2.0 and Ampliseq Variant Caller plugin v2.0.1.1 with the two normal samples being used to remove false positives.

Severe Combined Immunodeficient (SCID) mouse colony and procedures

This study was 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. A SCID mouse colony was developed at the University of Arizona using SCID mice obtained from Taconic (Germantown, New York). UACC-SARC1 cells were trypsinized at roughly 80% confluence, spun down at 800 rpm, and resuspended in 2% FBS saline for counting by hemocytometry and trypsin inactivation. The cells were then re-spun at 800 rpm, and the cells were resuspended in a 1:1 mixture of Matrigel (BD) and sterile saline.

Cells were resuspended in a 1:1 mixture of Matrigel (BD) and sterile saline. 5 × 10⁶ cells or 10⁷ cells were injected subcutaneously (SC) on the flank of 8-week-old SCID female mice. The tumors grew for 90 days, then the mice were sacrificed and subjected to gross necropsy. The tumors were harvested for FFPE block preparation. All procedures were completed in accordance with the University of Arizona Institutional Animal Care and Use Committee (IACUC) to the Experimental Mouse Shared Service (EMSS) of the Arizona Cancer Center.

Cell Cycle Analysis

UACC-SARC1 cells grown in a T75 flask (Thermo Fisher Scientific, Waltham MA) were detached with 0.05% trypsin-EDTA. Trypsin was deactivated with DMEM containing serum. The cell suspension was transferred into a 50 ml conical tube that was filled with cold PBS and spun down at 1,500 rpm for 10 minutes. The medium was aspirated and the cells were fixed by adding 1 ml ice-cold 70% ethanol. Fixed cells were spun down, resuspended in 1ml cold PBS, supplemented with 1/20 volume of 10mg/RNAse A and 1/40 volume of 1.6mg/ml propidium iodide dissolved in water and incubated at 37°C for 30 minutes. Flow cytometry cell cycle analysis was carried out using a FACScan (BD).

Growth curve

UACC-SARC1 was thawed from a fresh batch taken from liquid nitrogen storage. The youngest passage number was used (passage 21). After thawing, the cells were re-suspended in fresh DMEM+10% FBS+1% AA in a T75 flask and were allowed to grow for 5 days in the incubator at 37°C/5% CO2/92% humidity to allow adequate time for re-animation. After cells reached 80% confluency, the flask was trypsinized with 2 ml of 0.05% trypsin for 5 minutes. After adequate trypsinization, 6-7 ml of medium was added to stop the reaction. The cells were transferred to a 15ml conical tube and were centrifuged at 1100 RPM for 5 minutes. The pellet was re-suspended in 5ml of medium and cells were counted with a hemocytometer. Cell suspension was diluted with medium to achieve a seeding concentration of ~30,000cells/ml. 1.0ml of our calculated concentration was seeded into 5 T25, 5 T75 and 3 T175 flasks (Thermo Fisher Scientific, Waltham MA) labeled day 2-13 and placed in the incubator. Every 24 hours, 1 flask was counted with the hemocytometer twice. The average was taken to establish the concentration of cells in the flask and was multiplied by the re-suspension volume to give the total number of cells per flask. These numbers were then plotted to establish the growth curves using Prism GraphPad Version 6.

NOD/SCID gamma mouse model

All NOD/SCID gamma mouse studies were performed according to the protocols approved by the University of Southern California Institutional Animal Care and Use Committee. Experiments were conducted in strict compliance with IACUC regulations. Mice showing any signs of distress or pain were humanely sacrificed. Female NOD/SCID gamma mice were purchased from Jackson Laboratory (Bar Harbor, ME) and were allowed to acclimate for two weeks prior to being used for our studies.

Wild type and GFP/Luciferase tagged UACC-SARC1 cells were cultured in DMEM supplemented with 10% FBS in T75 flasks. Flow cytometry was used on confirm that all UACC-SARC1 cells used for this in vivo model were GFP/luciferase positive. When the cells reached 85% confluency, they were trypsinized with 1.5ml 0.05% trypsin, followed by determination of the concentration of cells by hemocytometry. Cells for subcutaneous injection were mixed with Matrigel to obtain a 1:1 cell-Matrigel mixture containing proposed cell number in 100ul of the mixture. Two cellular concentrations were prepared of both the wild type and GFP/ Luciferase tagged cells for mouse modeling, 5×10⁵ (low dose) and 5×10⁶ (high dose) cells, which were subcutaneously injected into ten 6-week old female NOD/SCID gamma mice. Six mice were injected with the GFP/ Luciferase tagged cells (3 mice with 5×10⁵ (low dose) and 3 mice with 5×10⁶ (high dose). Additionally, four mice were injected with wild-type untagged cells - 2 mice with 5×10⁵ (low dose) and 2 mice with 5×10⁶ (high dose) cells. Tumor growth was monitored twice a week with digital calipers for the mice receiving subcutaneous injection.

Ten additional 6-week old NOD/SCID gamma mice were used for an intra-cardiac injection metastasis model as adapted from Gupta et al[17]. Flow cytometry was used on confirm that all UACC-SARC1 cells used for this in vivo model were GFP/luciferase positive. Wild type and GFP/Luciferase tagged UACC-SARC1 cells were cultured in DMEM supplemented with 10% FBS in T75 flasks. When the cells reached 85% confluency, they were trypsinized with 1.5ml 0.05% Trypsin, followed by determination of the concentration of cells by hemocytometry. Cells were resuspended in PBS and two cellular concentrations of both the wild-type and GFP/Luciferase tagged cells 5×10⁴ and 1×10⁵ were prepared in a total volume of 50ul for direct injection into the cardiac left ventricular outflow tract of anesthetized mice under ultrasound guidance (Vevo 2100, FUJIFILM, VisualSonics Inc., Toronto, Canada). Bioluminescence imaging was used to monitor the development of metastatic lesions.

Six mice were injected with the GFP/Luciferase tagged cells - 3 mice with 5×10⁴ (low dose) and 3 mice with 1×10⁵ (high dose). Additionally, four mice were injected with wild-type untagged cells (2 mice with 5×10⁴ (low dose) and 2 mice with 1×10⁵ (high dose).

In vivo Optical Imaging

All bioluminescence imaging studies were performed at the University of Southern California Molecular Imaging Center using the IVIS SPECTRUM pre-clinical in vivo imaging system and data analyzed using the Living Image 4.2 software (PerkinElmer, Waltham, MA). Briefly, firefly luciferin (PerkinElmer, Waltham, MA) in sterile saline was injected at 50 kg/mg intravenously through the tail vein and the mice imaged 1.5 min post injection. Bioluminescence data for all studies were normalized and 1-minute exposures were used to generate images. Average radiance was determined for each mouse based on regions of interest.

Clinicopathologic Data

Pathologic analysis of the patient's tumor revealed a high-grade sarcoma with angioinvasion. The clinical characterization found no expression of ER, progesterone receptor, HER2, pancytokeratin, or CD117. The Ki67 index was 30% in the clinical pathology report; this finding was also true in subsequent research laboratory IHC and ICC staining of tumor and UACC-SARC1. Given the histomorphologic appearance and the IHC results, the tumor was clinically categorized as sarcoma, not otherwise specified. A residual 4 cm viable tumor remained in the extended radical mastectomy specimen. Margins were widely negative and the patient remains without evidence of disease five years after surgery.

STR Profiling

STR profiling of UACC-SARC1 cells demonstrated that this line is genotypically a match to its parental tumor, with only minor loss of heterozygosity (Figure 1). Comparing the sarcoma tissue to UACC-SARC1, we found heterozygosity of 46.7% and overall allele percent similarity of 78.1%. Using a relaxed definition of percent similarity (i.e. allowing for a difference of a single STR per locus), we found a percent similarity of 84.4%, which met the criteria for positive identification of a cell line (reference ≥80%) [12].

IHC and ICC

Figure 2 provides a growth curve (A) and photomicrograph (B) of UACC-SARC1. The doubling time of UACC-SARC1 based on hemocytometer counts was 28.3 hours and the cell line demonstrated logarithmic growth kinetics. Four technical replicates of the growth curves were highly consistent, as shown in Figure 2. The doubling time remained consistent after being cultured to 50 passages to date. By light microscopy, the sarcoma tissue and UACC-SARC1 were morphologically similar, showing an admixture of plump spindle-shaped cells as well as pleomorphic cells that appeared morphologically to be more histiocytoid. IHC analysis of the tumor and ICC analysis of UACC-SARC1 revealed shared expression of vimentin, osteonectin (SPARC), and CD68, but tumor-restricted expression of the histiocyte markers α₁-antitrypsin and α₁-antichymotrypsin (Figure 3). The pericytes of normal vessels within the tumor were positive for SMA, while the tumor cells were negative, however UACC-SARC1 cells were focally positive for SMA within the cytoplasm. α₁-antichymotrypsin was 3+ positive in the cytoplasm of the IHC of the tumor. The ICC cultured cells were negative. For α₁-antitrypsin, the tissue cells were weakly positive in the cytoplasm while the ICC of cultured UACC-SARC1 cells were negative. Both tissue and cell line were negative for CD34, EMA, S100, desmin, pancytokeratin 8, 14, 15, 16, 18, and 19, ER, lysozyme, EGFR, and CD99 (data not shown).

Transwell Invasion Assay

UACC-SARC1 cells displayed invasiveness through Matrigel and 8-micron membrane pores (Figure 4A and B). UACC-SARC1 was more invasive (mean 86%) than SYO-I (mean 45.1%) or HS-SY-II cells (mean 20%) when compared to control wells. As a negative control, normal human fibroblasts were unable to invade the Matrigel membrane. The difference among mean percent invasion for all 3-cell lines taken together was statistically significant (p=0.02).

Cytogenetic Analysis of UACC-SARC1 and Cell Cycle Analysis

Karyotype analysis of UACC-SARC1 cells revealed the presence of two abnormal aneuploid composite cell lines in this G-banded study (Figure 5). The first was hyperdiploid (>2N), with a modal range of 48 to 57 chromosomes; this was observed in 12/18 (67%) metaphase cells scored. The second was hypertetraploid (>4N), with a modal number of >95 chromosomes; this was observed in 6/18 (33%) metaphase cells scored. The hypertetraploid clone appeared to be a doubling of the hyperdiploid clone, and was, most likely, due to cultural artifact. The >2N spindle-cell sarcoma karyotype[18] is as follows:

48-57,XX,add(1)(p36.1)×2,del(1)(q25),del(2)(p21p23),add(3)(p13),del(3)(q21),−5, del(6)(q21),+7,del(7)(q22q36),+9,+9,add(9)(p24),add(11)(q21),add(13)(p12),add(13)(q34), del(13)(q12q22)x2,+14,add(15)(p10),add(15)(p12),add(15)(q26),+19,+19,hsr(19)(q13.4)×2, del(20)(q11.2),-21,+22,add(22)(q13),+1-4mar[cp7]

Extra copies of whole chromosomes 7, 9, 14, 19, and 22, and losses of chromosomes 5 and 21 were noted. Structural anomalies included losses in 1q, 2p, 3q, 6q, 7q, 13q, and 20q. Cell cycle analysis by flow cytometry found that 62.8% of the cells were in G1; 16.2% in G2; and 21% in S phase fraction (SPF).

CGH

Chromosomes of the sarcoma tumor, UACC-SARC1, peripheral blood, and control DNA (normal male versus normal female) were aligned and compared (Figure 6). Although UACC-SARC1 was genetically similar to its tumor of origin, more probes were affected by changes in copy number in the cell line – the majority of these being losses. Quantitative chromosomal losses and gains for the RIS, UACC-SARC1, and peripheral blood are summarized in a Venn diagram (Figure 7). Raw CGH data were deposited in Array Express (accession E-MEXP-3419).

Sequencing

A total of 8.5 million reads were obtained from the Ion Torrent 318 chip. 5.6 million reads remained after removal of polyclonal, primer dimer, and low quality reads by the Ion Torrent Server Suite software. Reads were evenly split between the four bar-coded samples (1.1-1.6 million reads). This provided enough reads for all 190 amplicons to achieve 100-fold coverage in all samples. False positives were identified using the patient's peripheral blood and the human reference DNA. After removal of false positives, and synonymous mutations, no mutations nor indels were found in either the tumor or UACC-SARC1. The sequenced genes, amplicons, and hotspots, along with the raw sequence data, are available as supplemental data (Supplemental Tables 1 and 2).

SCID Mouse Xenografts

Tumorigenic properties of UACC-SARC1 were assessed in vivo. Tumors formed in 3 of 3 SCID mice injected SC with 5 × 10⁶ UACC-SARC1 cells and in 3 of 3 SCID mice injected SC with 10⁷ UACC-SARC1 cells. However, none of those 6 mice showed evidence of metastasis, an observation that correlated with the patient's clinical course. While all xenografts successfully established tumors in SCID mice, these tumors peaked in size at days 5-7 (maximal volume of 85.5 mm³ for the 5 × 10⁶ UACC-SARC1 cells and maximal volume of 58.5 mm³ for xenografts inoculated with 10⁷ UACC-SARC1 cells). All xenografts grossly regressed to pinpoint tumors after a median of 29 days but all demonstrated viable tumor microscopically. The morphology of the xenografts by H&E staining (Figure 8) closely resembled that of the primary tumor's spindle component, but lacked histiocytic infiltration. The xenograft tumors were moderately well vascularized, but due to the amount of extracellular matrix, the vasculature structures were compressed and not abundantly obvious. The presence of cells of human origin in mouse xenografts was confirmed by IHC after staining with a human-specific lamin A/C antibody (Figure 7). Thus, the RIS tumor showed widespread nuclear membrane staining while the murine tissue showed nuclear membrane staining restricted to the SCID mouse xenograft; the adjacent mouse adipose tissue and stroma exhibiting low non-specific staining but no nuclear staining.

Validation IHC and ICC

To validate our next generation sequencing results with the Ion AmpliSeq Cancer panel and to provide further evidence of the origin of this cell line, additional IHC and ICC was performed and is presented in Figure 9. Ki67 was verified to be 30% positive for both cell line and tumor. Mutant p53 was 1% positive for each. 100% of cells expressed PTEN in both tumor and UACC-SARC1. RB1 was negative (0%) for each.

Clonogenicity Assay

90% (130/144) of the wells had cell growth and at least one cell colony present which originated from single cells. Figure 10 presents a representative image of an adherent colony that originated from a single UACC-SARC1 cell. Similarly, tumorspheres could be generated when UACC-SARC1 was cultured in ultra-low adherent plates for suspension culture and could be passaged to tertiary tumorspheres successfully.

NOD/SCID Gamma Mouse Xenografts and Imaging

Nine of nine (100%) orthotopic xenografts produced tumors (mean 829.3mm³). One additional mouse received an orthotopic injection but died for unknown reasons 1 week post-injection (low dose cohort). No significant difference was found for tumor size at the conclusion of the experiment (0.07) and all mice were sacrificed either when tumors grew to 1cm³, mice lost 10% body weight or more, or showed signs of systemic illness. A growth curve comparing low dose to high dose orthotopic injections is shown in Figure 11. No significant difference in mean growth kinetics (p=0.96) or final tumor volume (p=0.08) was seen between the high versus low dose cohorts, although there was a trend for larger tumors at 6 weeks in the high dose cohort. The mean survival time for both the low dose and high dose orthotopic cohorts were identical at 46 days. Two mice with high dose injections developed visceral metastasis in the orthotopic cohort – 1 with a liver metastasis and 1 with a pleural metastasis. No mice in the low dose cohort developed gross evidence of metastases. Figure 12 presents a representative bioluminescence scan for our orthotopic model. The mouse shown in 12H had developed a bulky pleural metastasis on necropsy. Mice injected with a higher dose of UACC-SARC1 showed an early dose response with higher levels of luminescence based on normalized comparisons of average radiance for imaging time points up to day 26 (example: A-C vs E-G) but were equivalent by day 32 post-injection (example: D vs H).

Of the mice in the intra-cardiac injection xenografts, 8/10 (80%) produced gross metastasis at necropsy, with the most prominent visceral sites of metastasis being liver (n=7), lung (n=1), and kidney (n=1). Based on bioluminescence, 8/10 (80%) mice had evidence of bony metastasis but this could not be verified based on gross tissue morphology or computed tomography angiography to verify gross bony destruction. Tissues were procured for future histologic studies and molecular pathologic analysis in accordance with our IACUC protocol. Mice injected with a higher dose of UACC-SARC1 showed more numerous and more rapid metastasis. The mean survival for the high dose intra-cardiac injection cohort was 33.8 days, versus 42 days for the low dose cohort. Figure 13 demonstrates the GFP/luciferase positive UACC-SARC1 at numerous sites of distant disease and the rapid progression of this metastasis model.

Supp TableS1

Supp TableS2

• 1.Strobbe LJ, Peterse HL, van Tinteren H, et al. Angiosarcoma of the breast after conservation therapy for invasive cancer, the incidence and outcome. An unforseen sequela. Breast Cancer Res Treat. 1998;47:101–109. doi: 10.1023/a:1005997017102. [DOI] [PubMed] [Google Scholar]
• 2.Gladdy RA, Qin LX, Moraco N, et al. Do radiation-associated soft tissue sarcomas have the same prognosis as sporadic soft tissue sarcomas? J Clin Oncol. 2010;28:2064–2069. doi: 10.1200/JCO.2009.25.1728. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Penel N, Grosjean J, Robin YM, et al. Frequency of certain established risk factors in soft tissue sarcomas in adults: a prospective descriptive study of 658 cases. Sarcoma. 2008;2008:459386. doi: 10.1155/2008/459386. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 4.Cha C, Antonescu CR, Quan ML, et al. Long-term results with resection of radiation-induced soft tissue sarcomas. Ann Surg. 2004;239:903–909. doi: 10.1097/01.sla.0000128686.51815.8b. discussion 909-910. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 5.Bjerkehagen B, Smeland S, Walberg L, et al. Radiation-induced sarcoma: 25-year experience from the Norwegian Radium Hospital. Acta Oncol. 2008;47:1475–1482. doi: 10.1080/02841860802047387. [DOI] [PubMed] [Google Scholar]
• 6.Stojadinovic A, Leung DH, Allen P, et al. Primary adult soft tissue sarcoma: time-dependent influence of prognostic variables. J Clin Oncol. 2002;20:4344–4352. doi: 10.1200/JCO.2002.07.154. [DOI] [PubMed] [Google Scholar]
• 7.Sheth GR, Cranmer LD, Smith BD, et al. Radiation-Induced Sarcoma of the Breast: A Systematic Review. The Oncologist. 2012 doi: 10.1634/theoncologist.2011-0282. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.Riad S, Biau D, Holt GE, et al. The clinical and functional outcome for patients with radiation-induced soft tissue sarcoma. Cancer. 2011 doi: 10.1002/cncr.26543. [DOI] [PubMed] [Google Scholar]
• 9.Mills J, Matos T, Charytonowicz E, et al. Characterization and comparison of the properties of sarcoma cell lines in vitro and in vivo. Hum Cell. 2009;22:85–93. doi: 10.1111/j.1749-0774.2009.00073.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Sonobe H, Manabe Y, Furihata M, et al. Establishment and characterization of a new human synovial sarcoma cell line, HS-SY-II. Laboratory investigation; a journal of technical methods and pathology. 1992;67:498–505. [PubMed] [Google Scholar]
• 11.Kawai A, Naito N, Yoshida A, et al. Establishment and characterization of a biphasic synovial sarcoma cell line, SYO-1. Cancer letters. 2004;204:105–113. doi: 10.1016/j.canlet.2003.09.031. [DOI] [PubMed] [Google Scholar]
• 12.Lorenzi PL, Reinhold WC, Varma S, et al. DNA fingerprinting of the NCI-60 cell line panel. Mol Cancer Ther. 2009;8:713–724. doi: 10.1158/1535-7163.MCT-08-0921. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 13.R_Development_Core_Team . R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; InVienna, Austria: 2010. [Google Scholar]
• 14.Gentleman RC, Carey VJ, Bates DM, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol. 2004;5:R80. doi: 10.1186/gb-2004-5-10-r80. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 15.Smith ML, Marioni JC, McKinney S, et al. snapCGH: Segmentation, normalisation and processing of aCGH data. 2009.
• 16.Olshen AB, Venkatraman ES, Lucito R, Wigler M. Circular binary segmentation for the analysis of array-based DNA copy number data. Biostatistics. 2004;5:557–572. doi: 10.1093/biostatistics/kxh008. [DOI] [PubMed] [Google Scholar]
• 17.Gupta P, Adkins C, Lockman P, Srivastava SK. Metastasis of Breast Tumor Cells to Brain Is Suppressed by Phenethyl Isothiocyanate in a Novel Metastasis Model. PloS one. 2013;8:e67278. doi: 10.1371/journal.pone.0067278. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Shaffer LG, Slovak ML, Campbell LJ. “ISCN (2009): An International System for Human Cytogenetic Nomenclature. Basel, Switzerland: Karger: 2009. [Google Scholar]
• 19.Cowan JM, Beckett MA, Tarbell N, Weichselbaum RR. Symmetrical chromosome rearrangements in cell lines established from human radiation-induced sarcomas. Cancer Genet Cytogenet. 1990;50:125–137. doi: 10.1016/0165-4608(90)90246-7. [DOI] [PubMed] [Google Scholar]
• 20.Kelland LR, Bingle L. Radiosensitivity and characterisation of a newly established cell line from an epithelioid sarcoma. British journal of cancer. 1988;58:322–325. doi: 10.1038/bjc.1988.211. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Li WW, Takahashi N, Jhanwar S, et al. Sensitivity of soft tissue sarcoma cell lines to chemotherapeutic agents: identification of ecteinascidin-743 as a potent cytotoxic agent. Clin Cancer Res. 2001;7:2908–2911. [PubMed] [Google Scholar]
• 22.Masters JR, Thomson JA, Daly-Burns B, et al. Short tandem repeat profiling provides an international reference standard for human cell lines. Proceedings of the National Academy of Sciences of the United States of America. 2001;98:8012–8017. doi: 10.1073/pnas.121616198. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 23.Nishio J, Iwasaki H, Nabeshima K, et al. Establishment of a new human pleomorphic malignant fibrous histiocytoma cell line, FU-MFH-2: molecular cytogenetic characterization by multicolor fluorescence in situ hybridization and comparative genomic hybridization. Journal of experimental & clinical cancer research : CR. 2010;29:153. doi: 10.1186/1756-9966-29-153. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 24.Perot G, Chibon F, Montero A, et al. Constant p53 pathway inactivation in a large series of soft tissue sarcomas with complex genetics. Am J Pathol. 2010;177:2080–2090. doi: 10.2353/ajpath.2010.100104. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Soini Y, Vahakangas K, Nuorva K, et al. p53 immunohistochemistry in malignant fibrous histiocytomas and other mesenchymal tumours. J Pathol. 1992;168:29–33. doi: 10.1002/path.1711680106. [DOI] [PubMed] [Google Scholar]
• 26.Gibault L, Perot G, Chibon F, et al. New insights in sarcoma oncogenesis: a comprehensive analysis of a large series of 160 soft tissue sarcomas with complex genomics. J Pathol. 2010 doi: 10.1002/path.2787. [DOI] [PubMed] [Google Scholar]
• 27.Solimini NL, Xu Q, Mermel CH, et al. Recurrent hemizygous deletions in cancers may optimize proliferative potential. Science. 2012;337:104–109. doi: 10.1126/science.1219580. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Gustafson P, Ferno M, Akerman M, et al. Flow cytometric S-phase fraction in soft-tissue sarcoma: prognostic importance analysed in 160 patients. British journal of cancer. 1997;75:94–100. doi: 10.1038/bjc.1997.15. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Huuhtanen RL, Blomqvist CP, Wiklund TA, et al. S-phase fraction of 155 soft tissue sarcomas: correlation with clinical outcome. Cancer. 1996;77:1815–1822. doi: 10.1002/(SICI)1097-0142(19960501)77:9<1815::AID-CNCR9>3.0.CO;2-A. [DOI] [PubMed] [Google Scholar]
• 30.Fletcher CDM, Unni K, Mertens F. IARC Press; Lyon: 2002. “World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Soft Tissue and Bone. [Google Scholar]
• 31.Hartel PH, Bratthauer G, Hartel JV, Fanburg-Smith JC. Primary malignant fibrous histiocytoma (myxofibrosarcoma/pleomorphic sarcoma not otherwise specified) of the breast: clinicopathologic study of 19 cases. Ann Diagn Pathol. 2011;15:407–413. doi: 10.1016/j.anndiagpath.2011.05.007. [DOI] [PubMed] [Google Scholar]
• 32.Dontu G, Abdallah WM, Foley JM, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & development. 2003;17:1253–1270. doi: 10.1101/gad.1061803. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.Liu H, Patel MR, Prescher JA, et al. Cancer stem cells from human breast tumors are involved in spontaneous metastases in orthotopic mouse models. Proceedings of the National Academy of Sciences of the United States of America. 2010;107:18115–18120. doi: 10.1073/pnas.1006732107. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Al-Hajj M, Wicha MS, Benito-Hernandez A, et al. Prospective identification of tumorigenic breast cancer cells. Proceedings of the National Academy of Sciences of the United States of America. 2003;100:3983–3988. doi: 10.1073/pnas.0530291100. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 35.Wang CY, Wei Q, Han I, et al. Hedgehog and Notch signaling regulate self-renewal of undifferentiated pleomorphic sarcomas. Cancer research. 2012;72:1013–1022. doi: 10.1158/0008-5472.CAN-11-2531. [DOI] [PubMed] [Google Scholar]

Supplementary Materials