Downregulated MTAP expression in myxofibrosarcoma: A characterization of inactivating mechanisms, tumor suppressive function, and therapeutic relevance

Myxofibrosarcomas are genetically complex and involve recurrently deleted chromosome 9p, for which we characterized the pathogenically relevant target(s) using genomic profiling. In 12 of the 15 samples, we detected complete or partial losses of 9p. The only aggressiveness-associated, differentially lost region was 9p21.3, spanning the potential inactivated methylthioadenosine phosphorylase (MTAP) that exhibited homozygous (4/15) or hemizygous (3/15) deletions. In independent samples, MTAP gene status was assessed using quantitative- and methylation-specific PCR assays, and immunoexpression was evaluated. We applied MTAP reexpression or knockdown to elucidate the functional roles of MTAP and the therapeutic potential of L-alanosine in MTAP-preserved and MTAP-deficient myxofibrosarcoma cell lines and xenografts. MTAP protein deficiency (37%) was associated with MTAP gene inactivation (P < 0.001) by homozygous deletion or promoter methylation, and independently portended unfavorable metastasis-free survival (P = 0.0318) and disease-specific survival (P = 0.014). Among the MTAP-deficient cases, the homozygous deletion of MTAP predicted adverse outcome. In MTAP-deficient cells, MTAP reexpression inhibited cell migration and invasion, proliferation, and anchorage-independent colony formation and downregulated cyclin D1. This approach also attenuated the tube-forming abilities of human umbilical venous endothelial cells, attributable to the transcriptional repression of MMP-9, and abrogated the susceptibility to L-alanosine. The inhibiting effects of MTAP expression on tumor growth, angiogenesis, and the induction of apoptosis by L-alanosine were validated using MTAP-reexpressing xenografts and reverted using RNA interference in MTAP-preserved cells. In conclusion, homozygous deletion primarily accounts for the adverse prognostic impact of MTAP deficiency and confers the biological aggressiveness and susceptibility to L-alanosine in myxofibrosarcomas.


Method S1: Conditions of real-time quantitative PCR used to measure MTAP gene dosage
The DNA extracted from myxofibrosarcoma cell lines and LCM-isolated pure myxofibrosarcoma cells was quantified using an ND-1000 spectrophotometer (Nanodrop). The relative fold change in the MTAP gene copy number between normal and tumor tissues or various cell lines was measured using a quantitative PCR assay, incorporating the comparative threshold cycle (C t ) method. The TaqMan assays targeting PIK3R1 (Hs06028467_cn) in 5q13.1 and exon 8 of the MTAP gene (Hs02079487_cn) in 9p21.3 were ordered from Applied Biosystems and separately amplified using the ABI StepOnePlus ™ Real-Time PCR System (Applied Biosystems). The PCR was performed in a total volume of 20 μL in each well, which contained 10 μL of TaqMan Genotyping MasterMix, 20 ng of genomic DNA, and 1 μL of 20x TaqMan assay probe. The PCR conditions involved an initial denaturation step of 95°C for 10 min, followed by 40 cycles at 95°C for 15 s, and 60°C for 1 min. All reactions were conducted in duplicate, and a negative control with no template was run in parallel with each PCR reaction.
The MTAP gene copy numbers of the tumor samples were derived by calculating the gene ratios of MTAP to PIK3R1 by using the 2 ΔΔCt method and normalizing them to the gene copy number of PIK3R1, defined as 2. A reference standard curve was established using input templates of 100, 20, 8, and 0.4 ng human genomic DNA (Clontech) as the calibrator. This gene ratio (R) was determined using the following formula: R = 2 ΔΔCt , where ΔΔCt = ΔCp (calibrator) -ΔCp(sample) and ΔCp = Cp(MTAP) -Cp(PIK3R1). The average ΔCp of the standard curve points was set as ΔCp(calibrator) at a gene ratio of 1. Accordingly, a normalized 2-gene copy test sample was expected to yield a gene ratio of 1.
Regarding the formalin-fixed samples, the MTAP/PIK3R1 gene ratio < 0.2 was considered as representing homozygous deletion based on the assumption that the normal tissue contamination and intratumoral heterogeneity should collectively account for <20% of experimental deviation. The samples exhibiting measured gene ratios between  0.2 and ≤ 0.8, between > 0.8 and ≤1.2, and > 1.2 were classified as the hemizygous deletion, normal copy number, and gain of the MTAP gene, respectively. Regarding the 3 myxofibrosarcoma cell lines, homozygous deletion was determined when the gene ratio approximated 0 and no MTAP gene copy was detectable in the presence of amplifiable PIK3R1.
Method S2: Real-time RT-PCR used to quantify MTAP transcript and to identify its candidate angiogenesis-regulating mediators by using an RT-PCR array The RNeasy Mini kit (Qiagen, Valencia, CA, USA) was used to extract total RNA from the NMFH-1 myxofibrosarcoma cell lines transduced with stable shMTAP-mediated knockdown of endogenous MTAP expression and shLacZ controls. The RNAs were further reverse-transcribed using the SuperScript ™ III First-Strand Synthesis System (Invitrogen, Carlsbad, CA, USA) according to the manufacturer instructions. Real-time PCR array was performed to quantify the expression level of MTAP transcript by using predesigned TaqMan assay reagents (MTAP Hs00559618_mL, MMP-9 Hs00234579_mL, and POLR2A [i.e., RNA polymerase polypeptide A] Hs01108291_mL, from Applied Biosystems, Foster City, CA, USA) and the ABI StepOnePlus™ Real-Time PCR System. The obtained data were normalized based on the expression of the POLR2A housekeeping transcript. The relative expression fold of the MTAP or MMP-9 transcript was then determined based on 2 -∆∆Cp , where ∆∆C T = ∆C T () -∆C T (sarcoma cells) and ∆C T (CCD966SK fibroblasts) represented the C T of MTAP subtracted from the C T of POLR2A. Only samples yielding a C T value < 32 for POLR2A were considered to exhibit acceptable RNA quality and were included in the analyses.
To explore potential mediators of MTAP that could implicate its antiangiogenic function, we applied an RT-PCR array (PAHS-128, SABioscience) to analyze the differential expression profiles of MTAPnull OH931 and NMFH-2 cells stably engineered to reexpress MTAP and of MTAP-expressing NMFH-1 cells stably silenced by shMTAP. The PCR reactions for 84 angiogenesis-associated genes were run on the ABI StepOnePlus ™ System (Applied Biosystems). Expression was normalized to housekeeping genes and presented as fold expression relative to the empty vector or shLacZ controls.

Method S4: Enzyme-linked immunosorbent assay used to quantify the protein concentration in conditioned media
MTAP-reexpressing myxofibrosarcoma cells and vector-alone controls were plated in a 24-well culture dish (5 × 10 5 /mL) and maintained for 96 h in serum-free media. After being filtered, the supernatant of media from various transfectants was percolated through a 0.22 um filter to measure the concentration of MMP-9 by using ELISA (R&D). The absorbance of the samples was determined using an ELISA reader (Promega) at 450 nm, using the absorbance at 560 nm as a reference.

Method-S5: Cell viability assay
The investigated cells were plated on gelatinized 96-well plates at a density of 5 × 10 3 cells per well and analyzed on an ELISA microplate reader at 492 nm by using the XTT assay kit (Roche).

Method-S6: Bromodeoxyuridine (BrdU) assay and electric cell-substrate impedance (ECIS) sensing assays
DNA synthesis was assessed using ELISA-based and colorimetric BrdU assays (Roche Diagnostics). Various plasmid vector-transfected, shRNA-transduced, and L-alanosine-or PBS-treated cells were plated onto a 96-well plate at a density of 2000, 3000, and 3000 cells per well for OH931, NMFH-1, and NMFH-2, respectively. In addition, DNA syntheses were evaluated at 24, 48, and 72 h. After incubation with BrdU for 3 h at 37°C in 5% CO 2 , the labeling medium was removed, and the samples were fixated, then underwent a final incubation with an anti-BrdU-POD solution. The absorbance of the samples was measured using an ELISA reader (Promega) at 450 nm, and the absorbance at 690 nm was used as reference.
To measure cell proliferation accurately, an 8-electrode culture dish (Applied Biophysics) involving one electrode per well was precoated with 400 uL of media and experimental additives and then incubated for 2 h. A total of 4 × 10 4 MTAP-reexpressing or empty control NMFH-2 cells were then seeded into the well. The ECIS collected the resistance values every 10 min for a total run time of 72 h.

Method-S7: Flow cytometric analysis of cell cycle
Stable pools of NMFH-2 myxofibrosarcoma cells transfected with MTAP-expressing plasmids or empty vectors were pelleted and fixed overnight in 75% cold ethanol at −20°C. The cells were washed twice in cold PBS containing 10 mg/mL of DNase-free RNase. Subsequently, the cells were labeled using propidium iodide (PI) at a concentration of 0.05 mg/mL and analyzed using the FACScan flow cytometer (BD Biosciences) on WinMDI2.9 software to determine the percentage of cells in each phase of the cell cycle. In all experiments, more than 10 4 cells were sorted after gating the fixation artifacts and cell debris.

Method-S8: Soft agar assay
To assess the level of anchorage-independent growth, 2 × 10 4 NMFH-2 myxofibrosarcoma cells transfected with pCMV6-ASS1-DDK-Myc or empty vectors were trypsinized, mixed in 1 mL of 0.3% agar in a complete medium, and then seeded onto a 6-well plate containing 1.5 mL of 0.6% agar in a complete medium. After being further cultured for 14 d, the colonies were visualized using 0.5% P-iodonitrotetrazolium violet stain overnight and those >100 μm were counted under the microscope in 10 random fields in 3 independent triplicate assays.

Method-S9: Wound healing and invasion assays
An artificial "wound" was created using a 200 μL pipette tip on confluent cell monolayers on 6-well plates. Photographs were captured at 0, 12, 24, 48, and 72 h. Quantitative analysis of the wound closure was calculated by measuring the percentage of healed area relative to the initial wound at Day 0.
The level of cell invasion was determined using a 24-well Collagen-based Cell Invasion Assay (Millipore). Briefly, each insert was added to a 0.3-mL serumfree medium to rehydrate the collagens, replaced with 0.3 mL of a serum-free suspension of 10 6 cells in the upper chamber, and incubated for 24 h to enable the cells to migrate toward the lower chamber containing 10% fetal bovine serum. After noninvading cells in the upper chamber were removed, cells invading through the inserts were stained using provided dye, dissolved in an extraction buffer, and transferred to 96-well plates for colorimetric reading at 560 nm.

Method S10: HUVEC tube formation assay
To analyze the effect of MTAP on tumor cell-induced tube formation in HUVECs, the conditioned media of myxofibrosarcoma cells exhibiting exogenous MTAP reexpression, shMTAP, or their corresponding controls, were harvested using serum-free starvation for 96 h. The 96-well plates were precoated with matrigel after polymerization of the matrix at 37°C. The HUVECs were then plated at a density of 2 × 10 4 cells/well and incubated using various conditioned media. After 6 h of treatment, the capillary-like tube formation was assessed by measuring the length of the capillary mesh under an inverted photomicroscope.

Method-S11: Cell apoptosis analysis
For the cell apoptosis analysis, 10 5 NMFH-2 or OH931 cells/Petri dish (60 mm) were plated for 24 h, harvested after incubation with L-alanosine or vehicle controls for the indicated doses and time, and then stained using the Annexin V-FITC kit (Bender MedSystems) containing propodium iodine for 15 min. The cell percentages at the stages of early apoptosis and late apoptosis were calculated based on 3 independent experiments. Supplementary Table S1. The regions on chromosome 9 recurrently showing copy number alterations in at least 20% of tumor and cell line samples analyzed by aCGH. In the short arm of chromosome 9, short stretches of DNA gains are interspersed among 5 major discontinuous deletion cores. Of these, 9p22.2 -p21.1 harbors the 9p21.3 region which are differentially lost in cases exhibiting clinical aggressiveness. In contrast, the copy number alterations in the long arm of chromosome 9 are predominantly DNA gains, except for the telomeric part where the 9q34 region shows DNA losses.