Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Radiotherapy resistance in chondrosarcoma cells; a possible correlation with alterations in cell cycle related genes

Radiotherapy resistance in chondrosarcoma cells; a possible correlation with alterations in cell... Background: Conventional chondrosarcomas are malignant cartilage tumors considered radioresistant. Neverthe‑ less, retrospective series show a small but significant survival benefit for patients with locally advanced disease treated with radiotherapy. And, in daily practice when considered inoperable their irradiation is an accepted indication for proton beam radiotherapy. Therefore, we investigated the sensitivity of chondrosarcoma cell lines and ‑tissue samples towards radiotherapy and screened for biomarkers to identify predictors of radiosensitivity. Methods: Proliferation and clonogenic assays were performed in chondrosarcoma cell lines after γ‑radiation in combination with mutant IDH1 inhibitor AGI‑5198. In addition, glutathione levels were measured using mass spec‑ trometry. Chondrosarcoma tumor explants were irradiated after which γ‑H2AX foci were counted. Mutation analysis was performed using the Ion AmpliSeq Cancer Hotspot Panel and immunohistochemical staining’s were performed for P‑S6, LC‑3B, P53, Bcl‑2, Bcl‑xl and Survivin. Results were correlated with the number of γ ‑H2AX foci. Results: Chondrosarcoma cell lines were variably γ‑radiation resistant. No difference in radiosensitivity, nor glu‑ tathione levels was observed after treatment with AGI‑5198. Irradiated chondrosarcoma patient tissue presented a variable increase in γ‑H2AX foci compared to non‑radiated tissue. Samples were divided into two groups, high and low radioresistant, based on the amount of γ‑H2AX foci. All four highly resistant tumors exhibited mutations in the pRb pathway, while none of the less radioresistant tumors showed mutations in these genes. Conclusions: Chondrosarcoma cell lines as well as primary tumors are variably radioresistant, particularly in case of a defective Rb pathway. Whether selection for radiotherapy can be based upon an intact Rb pathway should be further investigated. Keywords: Chondrosarcoma, Radiotherapy resistance, Cell cycle alterations (2%), and periosteal chondrosarcoma (1%). Conventional Background chondrosarcoma can be subdivided into three histologi- Chondrosarcomas are a heterogeneous group of cartilage cal grades, representing the most important prognostic producing tumors of which the most prevalent subtype is factor. Atypical cartilaginous tumors (previously referred conventional chondrosarcoma (85%). More rare subtypes to as grade I) have a relatively good prognosis, exhibit- include dedifferentiated chondrosarcoma (10%), mesen - ing a 10  years survival of 83%. Probability of surviving chymal chondrosarcoma (2%), clear cell chondrosarcoma 10 years is about 64% for grade II chondrosarcomas and about 29% for grade III chondrosarcomas [1–3]. Based *Correspondence: j.v.m.g.bovee@lumc.nl on their location conventional chondrosarcoma can be Department of Pathology, Leiden University Medical Center, further subdivided into central (85%) and peripheral Albinusdreef 2, 2333 ZA Leiden, The Netherlands Full list of author information is available at the end of the article © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 2 of 11 (15%) subtypes. Rare chondrosarcoma subtypes dedif- Methods ferentiated, mesenchymal and clear cell chondrosarcoma Cell culture occur also centrally in the bone [4–6], while periosteal Conventional chondrosarcoma cell lines JJ012 (Grade II, chondrosarcoma is located and originates from the peri- IDH1 mutant) [16], SW1353 (Grade II, IDH2 mutant) osteal surface of the bone [3]. In this study, we focused (ATCC) and CH2879 (Grade III, IDH wild type) [17] were on central chondrosarcomas, as these are of highest cultured in RPMI-1640 (Gibco, Invitrogen Life-Tech- prevalence. Central chondrosarcomas carry mutations nologies, Scotland, UK) supplemented with 10% Fetal in the genes encoding the enzymes isocitrate dehydroge- Calf Serum, at 37 °C in a humidified incubator (5% CO ). nase 1 or -2 (IDH1 or IDH2) [7–9] in approximately 50% Short tandem repeat analysis was performed before and of cases, resulting in production of the oncometabolite after completion of experiments to confirm identity of d-2-hydroxyglutarate (D-2HG). the cell lines by using the Cell ID Gene Print 10 system The only curative treatment for patients with chon - (Promega Benelux BV, Leiden, The Netherlands). Myco - drosarcoma is radical surgery, since they are considered plasma tests were performed on a regular basis. resistant towards conventional chemo- and radiotherapy. For this reason, linear accelerator based (conventional) Compounds radiotherapy is only given to patients with metastatic The specific mutant IDH1 inhibitor AGI-5198 (14624, disease, incomplete resection or inoperable tumors in Cayman Chemicals, Michigan, USA) was dissolved in difficult sites like the base of skull or the sacrum [2, 10]. DMSO according to the manufacturer’s instructions and Several causes have been suggested for chondrosarcoma stored in − 20 °C. AGI-5198 was used at a concentration radioresistance, such as the relative hypoxic chondro- of 10  µM since our group previously showed that this sarcoma microenvironment impairing the induction leads to a complete inhibition of D-2HG production [18]. of reactive oxygen species (ROS) and DNA damage, or (2R)-Octyl-α-hydroxyglutarate (16366, Cayman Chemi- their relatively slow dividing rate  (though not applicable cals), a cell-permeable derivative of D2-HG, was freshly to high grade tumors). Additionally, knock down of anti- dissolved in PBS before use and used at a final concentra - apoptotic Bcl-2 family members Bcl-2, Bcl-xl and XIAP tion of 250 µM. has been shown to increase cell death after radiation in chondrosarcoma cell lines [11]. Furthermore, restora- Clonogenic assays tion of p16 activity sensitized chondrosarcoma cell lines Chondrosarcoma cell lines SW1353 and JJ012 were to radiotherapy [12]. More recently, a role for IDH muta- plated in optimal cell densities to obtain sufficient colo - tions has been found in predicting radiotherapy response nies after treatment. Cells were allowed to adhere over- in glioma due to altered redox responses in IDH mutant- night before treatment with a wide range (0, 1, 2, 4 or compared to wild type cells thereby enhancing radiosen- 6 Gy) of γ-radiation using a Cs source (YXLON, Comet sitivity [13]. technologies USA). Colony formation was assessed after A recent retrospective analysis suggested that a sub- 14  days of treatment by fixing and staining with 0.5% group of chondrosarcoma patients with locally advanced, crystal violet/6% glutaraldehyde. Colonies were counted unresectable disease showed a favorable overall survival manually and the surviving fraction (SF) was calcu- after conventional radiotherapy [10]. As chondrosarcoma lated by normalizing towards the plating efficiency of patients are not commonly treated with radiotherapy, untreated controls. The α/β ratios were calculated based prognostic biomarkers for radiosensitivity were inves- on the linear quadratic model. α/β ratios describe the tigated using in  vitro and ex  vivo methods. Therefore, slope of the cell-survival curve; acute responding tissues the aim of our study was to examine whether sensitivity show a higher ratio compared to late responding tissues. to γ-radiation can be observed in central conventional chondrosarcoma cell lines by determining clonogenic Viability assay survival and γ-H2AX  foci induction after radiation. In Chondrosarcoma cells were counted using a Burker Turk addition, radiosensitivity of chondrosarcoma patient counting chamber and seeded in optimized cell densities samples was determined by counting γ-H2AX foci after in 96 well plates. After attachment overnight cells were ex vivo radiation [14, 15]. Mutation and expression analy- irradiated with increasing doses. Seventy-two hours after ses were performed to investigate prognostic biomarkers radiation cell viability was measured using PrestoBlue for radiosensitivity which could then be used to select viability reagent (Invitrogen, Life-Technologies, Scot- patients for radiotherapy. land, UK) according to the manufacturer’s instructions. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 3 of 11 Fluorescence was measured at 590  nM using a Wallac with γ-H2AX results. Cells were harvested by scrap- plate reader (Victor3V, 1420 multilabel counter, Perkin ing after 1  h. For each individual replicate 1x10 cells Elmer, the Netherlands). Experiments were performed were fixed overnight using ice-cold acetone. The super - three times in triplicate. natant was then removed and dH O added (10 µL), fol- lowed by a freeze–thaw cycle (10 s dry ice-cooled MeOH, Cell proliferation measurement 2  min 37  °C heating block). After 2 freeze thaw cycles To measure cell proliferation in real time the RTCA glutathione was extracted with EtOH (50 µL) and the xCELLigence system (Roche Applied Sciences, Almere, supernatant was mixed 1:1 (v:v) with matrix solution the Netherlands) was used. JJ012, SW1353 and CH2879 (N-(1-naphthyl) ethylenediamine dihydrochloride in 70% cells were pre-treated for 72  h with either 10  µM AGI- MeOH) for analysis. Samples were measured using a 9.4T 5198 or 0.1% DMSO in T25 flasks. Thereafter, cells were SolariX XR Fourier transform ion cyclotron resonance plated in the presence of AGI-5198 or DMSO and left (FT-ICR) mass spectrometer (Bruker Daltonics, Bremen, for 20 h to attach to the wells before low-dose (2 Gy) or Germany) equipped with a CombiSource and a dynam- high dose (4  Gy) γ-radiation. After 125  h the measure- ically harmonized ParaCell . The mass spectrometric ment was stopped, and the results were analyzed using analysis was performed in negative ion mode in the m/z- RTCA software. Normalized cell index was calculated range 101.7–1000  Da, with a 1  M data point transient by normalizing against the amount of cells in each well (0.3670  s duration) and an estimated resolving power after 20 h. Experiments were performed in duplicate and of 66,000 at m/z 400. Spectra were acquired of 10 aver- repeated at least two times. aged scans, acquired in a random pattern covering the entire spot, where each scan consisted of 500 laser shots γ‑H2AX staining of chondrosarcoma cell lines (laser power 70%, 1000 Hz frequency). Spectra were ana- Optimal cell amounts for JJ012 (550,000 cells), SW1353 lyzed using DataAnalysis version 4.2 (Bruker Daltonics, (500,000 cells) and CH2879 (1,000,000 cells) cell lines Bremen, Germany). Using a custom algorithm, intensi- were cultured on alcohol sterilized APES coated slides ties of the compounds of interest were extracted and in a culture dish. IDH mutant chondrosarcoma cell lines exported to Excel 2016 (Microsoft) for further analysis. SW1353 and JJ012 were pre-treated with 10  µM AGI- 5198 or 0.1% DMSO for 72 h, in order to inhibit mutant Chondrosarcoma tissue samples activity, and IDH wild type chondrosarcoma cell line Chondrosarcoma patient samples (n = 9, see Tables 1 and CH2879 was pre-treated with 250  µM D-2HG for 24  h, 2) were obtained at surgery and collected in RPMI- in order to mimic mutant activity. After overnight attach- 1640 medium supplemented with 20% FBS and 1% P/S. ment cells were radiated to 5  Gy, in order to achieve All samples were coded according to the Dutch code of highest response without significant damage to sample, proper secondary use of human material as accorded by and fixed after 2 or 24 h with buffered 4% formaldehyde the Dutch society of pathology (Federa), and as approved (VWR Chemicals) at 37  °C. The procedure was per - by the LUMC ethical board (B17.019). Samples were pro- formed as described previously [19]. Slides were stained cessed by manual dissection into approximately 2  mm with γ-H2AX antibody (JBW301, Millipore) for 60  min, fragments and incubated on a rotation shaker at 37 °C in washed, then a mixture of Alexa fluor 657 labelled sec - a humidified incubator (5% CO ). The tissue was treated ondary antibody and 0.5 µM Hoechst 33342 was applied to 5 Gy, consistent with the cell line treatment, and incu- for 60  min. Slides were covered with ProLongGold anti- bated for 2 or 24  h on a rotation shaker at 37  °C in a fade reagent and were examined using a confocal micro- humidified incubator as described previously [21]. After scope by taking tile scans of areas of interest using ZEN overnight fixation with formalin, tissue was processed, light software. Pictures were exported in TIF format and embedded in paraffin and sections were cut and stained loaded into FoCo, a previously published format using for γ-H2AXfoci. Image J and Matlab software [20]. γ‑H2AX staining and quantification of paraffin embedded Mass spectrometry measurements of glutathione chondrosarcoma tissue Chondrosarcoma cell lines JJ012, SW1353 and CH2879 Sections were deparaffinized and antigen retrieval was were seeded in triplicate in T25 culture flasks and performed using DAKO antigen retrieval solution pH allowed to adhere overnight. IDH mutant chondrosar- 9.0 by boiling the slides in a microwave for 12 min. After coma cell lines JJ012 and SW1353 were pre-treated with cooling, slides were washed with TBS and blocking was 10  µM AGI-5198 for 72  h, while IDH wildtype cell line performed using TBS/5% normal goat serum/1% BSA CH2879 was pre-treated with 250  µM D-2HG for 24  h. and 0.2% Triton-x-100. Anti γ-H2AX antibody (Millipore Samples were treated with 0 or 5 Gy, in order to correlate 2310355) was applied in a 1:1000 dilution and incubated de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 4 of 11 Table 1 Patient samples treated with 5 gy of radiation Sample Grade IDH mutation status Other alterations Mean + SEM basic foci level Mean + SEM Difference P value after 5gy between means radiation L5213 1 WT 3.053 ± 0.6277 14.28 ± 1.737 11.23 ± 1.733 < 0.0001 n = 38 n = 32 L5678 3 IDH1 R132C 0.2295 ± 0.07165 8.239 ± 1.477 8.01 ± 1.284 < 0.0001 n = 61 n = 46 L5676 2 IDH2 R172S 3.625 ± 0.9297 9 ± 1.493 5.375 ± 1.784 0.003712 n = 32 n = 34 L5541 Dediff IDH1 R132S 2.378 ± 0.5747 6.702 ± 1.225 4.324 ± 1.212 0.0005 n = 90 n = 57 L5546 2 WT CDKN2A deletion 3.982 ± 0.9036 6.846 ± 1.172 2.864 ± 1.468 0.0537 n = 56 n = 52 L5302 3 WT CDKN2A deletion 2.246 ± 0.4531 5.082 ± 0.7412 2.836 ± 0.9108 0.0023 n = 61 n = 73 Foci were counted 2 h after treatment Table 2 Patient samples treated with 5 gy of radiation Sample Grade IDH mutation status Other alterations Mean + SEM basic Mean + SEM Difference P value foci level after 5gy radiation between means L5847 2 WT 2.644 ± 0.7399 7.467 ± 1.526 4.823 ± 1.501 0.0018 n = 59 n = 30 L5718 3 IDH1 R132L CDKN2A deletion 2.437 ± 0.584 4.088 ± 0.6256 1.651 ± 0.8659 0.0580 n = 87 n = 102 L5850 2 WT RB1 deletion 0.6327 ± 0.1409 1.55 ± 0.1809 0.9173 ± 0.2463 0.0002 n = 98 n = 140 Foci were counted 24 h after treatment overnight at 4 °C. Slides were washed with TBS and incu- standard laboratory methods as previously described bated with secondary labelled GAM Alexa fluor 647 [22]. All slides were scored by two independent observ- antibody. Mounting was performed with ProLongGold ers (JVMGB, YDJ) using a scoring system assessing stain- containing DAPI (Thermofisher Scientific, Waltham, ing intensity (0 = no, 1 = weak , 2 = moderate, 3 = strong) Massachusetts, USA). Positive control slides consisted as well as percentage of staining (0 = no, 1 = 1–24%, of JJ012 cells radiated at 4  Gy as these displayed a con- 2 = 25–49%, 3 = 50–74%, 4 = 75–100%) [23]. This scoring sistent, high γ-H2AX signal. The cells were embedded in system was used for all the different proteins. ™ ™ paraffin with the use of Shandon Cytoblock Cell Block Preparation system (Thermofisher Scientific). Slides Mutation analysis were analyzed using a confocal microscope by taking tile Frozen tissue of chondrosarcoma patients was col- scans of areas of interest using ZEN light software. Pic- lected at surgery and DNA was isolated using the wiz- tures were exported in TIF format and loaded into FoCo, ard genomic DNA purification kit (Promega, Madison, a previously published format using Image J and Matlab WI, USA) according to the manufacturer’s instructions. software [20]. The settings to quantify the foci were first Samples were subjected to next generation sequenc- optimized using the positive control slides, after which ing using the in-house developed Ion AmpliSeq Can- the complete series was analysed using the same settings. cer Hotspot Panel v2  (Life Technologies, Thermo Fisher Scientific, USA, catalog number 4475346) as described previously [24, 25]. The following genes and regions were Immunohistochemical staining analysed on an Ion Torrent PGM/Proton for all sam- Protein expression of Bcl-2, Bcl-xl, Survivin, P-S6, ples (exons between brackets): ARAF (7,10); CTNNB1 P53 and LC-3B were evaluated in irradiated con- (1,2,4,7,8,12,15); KRAS (2–4); NRAS (2–4); HRAS (2–3); trol patient samples (see Additional file  1: Table  S1). BRAF (11,15); EGFR (3,7,15,18–21); GNAQ (5); GNAS Immunohistochemistry was performed according to (8–9); H3F3A (2); H3F3B (2); IDH1 (4); IDH2 (4); KIT de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 5 of 11 (2,9-18); MYD88 (3b,5), MUTYH (7,13); PDGFRA in repairing double strand breaks and more sensitive to (12,14,15,18,23); PIK3CA (2,5,6–10,14,18,21); POLE radiation. (9,11,13,14); RET (10–12,15,16); TP53 (1–11). In addi- tion hotspot regions in the following genes were also γ‑Radiosensitivity of chondrosarcoma cell lines included: ABL1; AKT1; ALK; APC; ATM; CARD11; is not correlated with IDH mutation status CD79A; CD79B; CDH1; CDKN2A; CSF1R; CTNNB1; Inhibition of D-2HG production by JJ012 cells using ERBB2; ERBB4; EZH2; FBXW7; FGFR1; FGFR2; FGFR3; the mutant IDH1 inhibitor AGI-5198 did not show any FLT3; GNA11; HNF1A; JAK2; JAK3; KDR; MET; MLH1; difference in radiosensitivity (Additional file  1: Figure MPL; NOTCH1; NPM1; PTEN; PTPN11; RB1; SMAD4; S1A). In addition no differences were observed in prolif - SMARCB1; SMO; SRC; STK11; VHL. Libraries were pre- eration capacity and foci formation between cells treated pared with 10 ng of genomic DNA, and each sample was with AGI-5198 and radiation or cells treated with radia- uniquely barcoded. Ion Proton chips were prepared using tion only (Additional file  1: Figure S1B, C). Since previ- the Ion Chef System. The unaligned bam files generated ous reports [13] suggest that IDH mutant cells have a by the Proton sequencer were mapped against the human reduced capability of producing GSH, GSH levels were reference genome (GRCh37/hg19) using the TMAP 5.0.7 measured 1  h after radiation in combination with AGI- software with default parameters (https ://githu b.com/ 5198, for the mutant cell lines or D-2HG (oncometabo- ionto rrent /TS). The Torrent Variant Caller (TVC)-5.0.2 lite resulting from an IDH mutation) for the wild type was used for variant calling and variant interpretation cell line. The most radioresistant SW1353 cells displayed was done using Geneticist Assistant (http://softg eneti the highest baseline levels of GSH, followed by CH2879 cs.com/Genet icist Assis tant_2.php) as described. Chro- and JJ012, however no differences were observed mosomal gains and deletions were assessed by calculat- between different treatment conditions, indicating that ing the median base coverage per amplicon, which was GSH levels are not influenced by D-2HG inhibition normalized using the median value of all amplicons in (Additional file  1: Figure S1D). These results suggest that that sample. IDH and D-2HG do not play a role in chondrosarcoma radiosensitivity. Results Chondrosarcoma cell lines are variably resistant Chondrosarcoma samples that are more resistant to γ‑radiation towards radiotherapy have an increased incidence Clonogenic assay  SF2 values (surviving fraction of cells of mutations in cell cycle regulators after 2  Gy radiation) suggested that JJ012 cells (SF2 Chondrosarcoma patient samples showed an increase in 0.55) were more radiosensitive compared to SW1353 γ-H2AX foci after 5 Gy radiation treatment. In Tables  1, (SF2 0.88) (Fig.  1a). JJ012 cells showed a high α/β ratio 2 and Additional file  1: Figure S2 quantified results are of 38.47, in contrast to SW1353 with a very low α/β shown for nine chondrosarcoma explant tissue samples ratio of − 0.75, suggesting that SW1353  cells may ben- analyzed 2 or 24 h after treatment with 5 gy of radiation. efit from hypofractionation. CH2879 cells were unable Radiation response was heterogeneous across the sam- to form colonies and were therefore not included in this ples; the largest difference between control and radiated assay. In addition, dose response curves were made to samples was 11.2 foci (sample L5213, Fig.  2a, b), while assess viability after 72  h. JJ012 cells were most sensi- the smallest difference was 0.9 foci. Samples analyzed tive followed by the CH2879 and SW1353 cells (Fig. 1b). 24 h after radiation treatment in general showed a lower An X-CELLigence assay was performed to determine amount of foci compared to samples analyzed 2  h after the effect on proliferation in real time after 2 or 4  Gy radiation treatment. Samples that showed a difference of of radiation. All cell lines showed a reduction in prolif- 4 or more foci were subjected to the less radioresistant eration after radiation. However, CH2879 cells showed group, while samples that showed < 4 foci difference were a similar response when irradiated to 2 or 4  Gy, while designated as radioresistant for further analysis. This JJ012 and SW1353 showed a dose response relation- cutoff value was taken for both the 2  h and 24  h groups ship (Fig.  1c). The amount of double strand breaks was and both time points were analyzed together since sam- determined by quantifying γ-H2AX  foci 2 and 24  h ple size impeded separate analyses. The division between after radiation treatment. JJ012, SW1353 and CH2879 more or less radioresistant was made prior to mutation cells exhibited an increase in foci after 2  h, which was analysis. No correlation was observed between histo- decreased again after 24 h in SW1353 and CH2879 cells logical grade or IDH mutation status and the amount of (Fig.  1d–e). Conversely JJ012 cells still showed a sub- γ-H2AX  foci, consistent with the results obtained in the stantial significant amount of foci compared to non- chondrosarcoma cell lines (Table  1). Mutation analysis radiated cells, indicating this cell line is less proficient was performed on 50 known cancer related genes and Dose (Gy) Dose (Gy) CH2879 SW1353 CH2879 JJ012 Time-Interval (Hour) JJ012 Time-Interval (Hour) SW1353 Time-Interval (Hour) de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 6 of 11 ab 0.1 SW1353 SW1353 SF2: 0.88, α/β ratio: -0.75 JJ012 JJ012 SF2: 0.55, α/β ratio: 38.47 CH2879 01 2 56 0 10 c d 10 250 250 * 250 ns 0 Gy **** **** **** **** 2 Gy 200 200 200 4 Gy 150 ns 150 150 **** **** 100 100 100 50 50 50 0 0 0 0255075 100 125 0 Gy 5 Gy 5 Gy 0 Gy 5 Gy 5 Gy 0 Gy 5 Gy 5 Gy 0 h 2 h 24 h 0 h 2 h 24 h 0 h 2 h 24 h e CH2879 JJ012 SW1353 0 Gy 2 Gy 4 Gy 0 Gy 0 h 0255075 100 125 5 Gy 2 h 0 Gy 2 Gy 4 Gy 5 Gy 24 h 0255075 100 125 Fig. 1 Chondrosarcoma cell lines exhibit variable γ‑radioresistance. a Colony forming assay of SW1353 and JJ012 chondrosarcoma cells. b Viability of JJ012, SW1353 and CH2879 cells, 72 h after treatment with increasing doses of γ‑radiation measured using presto blue viability reagent. c Normalized cell index of CH2879, JJ012 and SW1353 cells irradiated to 0 (black), 2 (blue) or 4 Gy (green) radiation. d, e γH2AX foci/cell in CH2879, JJ012 and SW1353 cells 2 and 24 h after 5 Gy radiation.****P values < 0.0001, *P values < 0.05 Normalized Cell index Normalized Cell index Normalized Cell index Surviving fraction Foci/cell Viability (%) de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 7 of 11 expression of Bcl-2, Bcl-xl, Survivin, P-S6, LC3B and P53, mutant IDH1 using AGI-5198 did not lead to any changes previously identified to play a role in chondrosarcoma in radiosensitivity, nor in altered GSH levels. Also, no dif- [26–30], was determined using immunohistochemistry. ference was observed in γ-H2AX foci formation between No significant difference was observed in protein expres - IDH mutant and IDH wild type chondrosarcoma explants. sion of selected markers (Fig. 2c, Additional file  1: Figure This indicates that, unlike the observations in gliomas, S3), however three CDKN2A deletions (3/4) and one RB1 IDH1 or IDH2 mutations in chondrosarcoma do not cor- deletion (1/4) were found in the highly radioresistant relate with radiosensitivity. This is in line with our previ - group and none in the less radioresistant group indicat- ous results in which we also did not find any correlation ing that a defective Rb pathway may be able to impair between IDH1 or -2  mutation status in sensitivity for the response to γ-radiation in chondrosarcoma (Fig.  2d, inhibitors of glutaminolysis [34], NAD synthesis [35] or Tables 1 and 2). Interestingly a mutation in CDKN2A was Bcl-2 family members in chondrosarcoma cells, while in also identified in the SW1353 cell line (Additional file  1: other tumor types there was a clear difference in sensitiv - Table S2), which is the most radioresistant cell line. ity [36–38]. This difference indicates that IDH mutations may have a tissue specific effect rather than a more gen - Discussion eral effect in different tumor types. Chondrosarcomas are relatively radioresistant tumors Our results suggest that deletions in cell cycle regu- and therefore, after multidisciplinary discussions, very lators CDKN2A and RB1  are associated with increased few of these patients are offered radiotherapy [10]. In this radioresistance in chondrosarcoma explant tissue. In study we investigated the sensitivity of chondrosarcoma line with this, the most resistant SW1353 cells harbored cells and tumor explants to γ-radiation. In addition, we a mutation in the splice site region of CDKN2A  in screened for biomarkers that could select patients that addition to mutations in TP53 and a kRAS mutation might benefit from γ-radiation. Chondrosarcoma cell [39]. CDKN2A is a gene encoding p16(INK4A) and lines showed a heterogeneous response to radiotherapy p14(ARF), which are two tumor suppressor proteins with relatively high SF2 values [31]. JJ012 cells presented controlling the cell cycle. P16 inhibits CDK4 and CDK6, a SF2 value of 0.55, while SW1353 were more resistant two inhibitors of Rb1 phosphorylation, while p14- showing a SF2 value of 0.88. Compared to cell lines of ARF protects p53 from being broken down by inhibit- other tumor types, these values are relatively high, con- ing MDM2. Alterations in the pRB pathway have been firming chondrosarcoma  radio resistance. A previous described in the majority of high grade conventional published study by Hamdi et  al. showed a SF2 value of chondrosarcomas [40–43]. Previous studies in chon- 0.64 for the SW1353 cell line [32]. This difference might drosarcoma cell lines (CS-7, CS-8, CS-9) showed that be attributed to the fact that different methods were restoring P16 expression, and thereby increasing Rb1 used to perform the clonogenic assay. In our study the phosphorylation, resulted in an increased radiosensi- cells were seeded prior to radiation, while in the study tivity [12], in line with our results. Although we see a of Hamdi et  al. the cells were subconfluently seeded in clear difference in the amount of foci in tumors with culture flasks and subsequently radiated and seeded. In and without deletions in CDKN2A or RB1, this study addition the differences in culture conditions (normoxic is based on a small heterogenous group of chondrosar- vs hypoxic) can very well influence how cells respond comas and measurements are taken after 2 or 24  h. In to radiation treatment. Quantification of γ-H2AX foci addition the threshold of 4 foci/cell is taken arbitrar- showed that SW1353 and CH2879 cells were able to ily. This complicates making firm conclusions based on repair double strand breaks within 24  h after radiation this data and further studies should focus on; includ- while JJ012 cells were less capable of doing so, in line with ing more patients and analyze amount of foci after 24 h the lower SF2 values observed in JJ012 cells. to determine the damage remaining after DNA repair. In contrast to published studies in glioma [13, 33], we In addition when radiotherapy is included in the treat- did not observe a correlation between radiosensitivity and ment plan, follow the response towards radiation and IDH mutation status in chondrosarcoma. Inhibition of correlate this towards mutation status. (See figure on next page.) Fig. 2 γ‑Radioresistance in chondrosarcoma tissues correlates with mutations in cell cycle related genes. a L5213 chondrosarcoma tissue sample showing a large induction of γH2AX foci after5 Gy(right) compared to controls (right). b Amount of foci/cell in sample L5213 after 5 Gy radiation compared to controls. c Protein expression of Bcl‑2, Bcl‑xl, Survivin, P ‑S6, LC3B and P53 in highly radioresistant (< 4 foci difference) compared to less radioresistant (> 4 foci difference) chondrosarcomas. d Difference between mean amount of γH2AX foci before and after radiation in chondrosarcomas with and without alterations in RB1 or CDKN2A. Each dot represents one sample. Samples in green indicate 2 h, while samples in blue indicate samples 24 h after radiation de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 8 of 11 L5213 P = 1.217e-008 0 Gy 5 Gy Bcl-xl Survivin Bcl-2 8 10 P=0.07331 P=0.5496 P=0.4825 4 4 2 2 0 0 P-S6 LC-3B P53 8 5 4 P=0.7128 P=0.6195 6 3 4 2 2 P=0.8383 1 0 0 -1 Mean amount of foci P=0.01768 >4 foci difference <4 foci difference >4 foci difference >4 foci difference <4 foci difference <4 foci difference >4 foci difference >4 foci difference <4 foci difference <4 foci difference no cell cycle alterations deletion of CDKN2A or RB >4 foci difference <4 foci difference difference between means IHC score IHC score IHC score foci/cell IHC score de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 9 of 11 Abbreviations In contrast to our findings in chondrosarcoma, deletion D‑2HG: d ‑2‑hydroxyglutarate; IDH1 or IDH2: isocitrate dehydrogenase 1 or ‑2; of RB1 has been described to enhance radiosensitivity in ROS: reactive oxygen species; SF: surviving fraction. breast, prostate and bladder cancer [44–47]. This observa - Acknowledgements tion might therefore be tissue and context specific; only a We would like to thank Matty Meijers and MaaikeVreeswijk for technical assis‑ limited number of cancer types have been investigated. In tance with the tumor culture model and Ben Floot ( The Netherlands Cancer addition, pRb has multiple functions, not only in the cell Institute, Amsterdam) for his help with the colony formation assay. In addition we would like to thank Liam McDonnell for his advice and discussions on cycle but also in chromatin organization, transcription the optimization of the glutathione measurements. Also we would like to patterns, metabolic pathways and the proteome [48]. Sev- thank Francois Chevalier and Yannick Saintigny for fruitful discussions. We are eral studies found that loss of pRB expression resulted in grateful to Dr. JA Block (Rush University Medical Centre, Chicago, IL, USA), who provided us with the JJ012 cell line and Professor A Llombart Bosch (University an increase in glutamine consumption and an increased of Valencia, Spain) for the CH2879 cell line. glutamine incorporation into GSH. Upon RB knock down cells increased the expression of glutamine transport- Authors’ contributions YDJ, RH and JB were involved in the design of the experiments and writing the ers and upregulated glutaminase activity, indicating that manuscript. YDJ, MI and IP performed the mass spectrometry experiments. the pRB pathway can regulate glutamine metabolism MI, IP and BH analysed the mass spectrometry experiments. IB, AK, SV and YDJ and that cells with inactivated pRB are potentially more performed all in vitro cell culture experiments. IB and AMCJ performed and analysed the sequencing experiments. All authors read and approved the final sensitive for targeted anti-glutamine treatment [49–51]. manuscript. Interestingly, we recently reported that interfering with glutamine metabolism can be a therapeutic target for Funding This work was financially supported by the Dutch Cancer Society (UL2010‑ high grade chondrosarcoma [34]. The SW1353 chondro - 4873 and UL2013‑6103) and EuroSarc under Grant Agreement 278742. sarcoma cell line was particularly sensitive for inhibition of glutaminase, which is in line with the high basal GSH Availability of data and materials All data generated or analysed during this study are included in this published levels measured in this cell line. We can hypothesize that article (and its additional files). the mutation in CDKN2A observed in this cell line might contribute to the glutamine dependence, however more Ethics approval and consent to participate All samples were coded according to the Dutch code of proper secondary use research is needed to further investigate this. of human material as accorded by the Dutch society of pathology (Federa), In addition to conventional radiotherapy, research has and as approved by the LUMC ethical board (B17.019). been focusing increasingly on proton [ H] and carbon Consent for publication [ C] therapy, which have several advantages compared Not applicable. to γ-radiation. Both these beam qualities have dose dis- tribution advantages, causing less damage to surrounding Competing interests The authors declare that they have no competing interests. healthy tissues, making it possible to deliver higher doses to the tumor. Chondrosarcomas of the skull base and Author details spine are increasingly treated with proton beam radiation, Department of Pathology, Leiden University Medical Center, Albinusdreef 2, 2333 ZA Leiden, The Netherlands. Center for Proteomics and Metabolomics, showing promising results [52]. Whether chondrosar- Leiden University Medical Center, Leiden, The Netherlands. Department coma radiosensitivity differs between photon (γ-) beams of Radiation Oncology, Leiden University Medical Center, Leiden, The Nether‑ and proton beams is subject for further investigation. lands. Department of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands. Received: 27 February 2019 Accepted: 20 May 2019 Conclusion In conclusion, this study confirms a heterogeneous radio - sensitivity of chondrosarcoma cell lines and fresh explant tissue. In addition, we identified alterations in CDKN2A/ References 1. Evans HL, Ayala AG, Romsdahl MM. Prognostic factors in chondrosarcoma RB1 as predictors of decreased double strand break for- of bone: a clinicopathologic analysis with emphasis on histologic grad‑ mation after radiotherapy in chondrosarcoma tissue, and ing. Cancer. 1977;40(2):818–31. although further research is necessary in a larger group, 2. Gelderblom H, Hogendoorn PCW, Dijkstra SD, van Rijswijk CS, Krol AD, Taminiau AH, et al. The clinical approach towards chondrosarcoma. this suggests that these patients might benefit less from Oncologist. 2008;13(3):320–9. radiation therapy. 3. Hogendoorn PCW, Bovee JVMG, Nielsen GP. Chondrosarcoma (grades I‑III), including primary and secondary variants and periosteal chon‑ drosarcoma. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO classification of tumours of soft tissue and bone. Lyon: IARC; Additional file 2013. p. 264–8. 4. Inwards C, Hogendoorn PCW. Dedifferentiated chondrosarcoma. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO clas‑ Additional file 1. Supplementary material. sification of tumours of soft tissue and bone. Lyon: IARC; 2013. p. 269–70. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 10 of 11 5. McCarthy EF, Hogendoorn PCW. Clear cell chondrosarcoma. In: Mertens F, 25. van Riet J, Krol NMG, Atmodimedjo PN, Brosens E, Jansen M, et al. SNPitty: editor. WHO classification of tumours of soft tissue and bone. Lyon: IARC; an intuitive web application for interactive B‑allele frequency and copy 2013. p. 273–4. number visualization of next‑ generation sequencing data. J Mol Diagn. 6. Nakashima Y, de Pinieux G, Ladanyi M. Mesenchymal chondrosarcoma. In: 2018;20(2):166–76. Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO clas‑ 26. de Jong Y, van Oosterwijk JG, Kruisselbrink AB, Briaire‑ de Bruijn IH, Agrogi‑ sification of tumours of soft tissue and bone. Lyon: IARC; 2013. p. 271–2. annis G, Baranski Z, et al. Targeting survivin as a potential new treatment 7. Amary MF, Bacsi K, Maggiani F, Damato S, Halai D, Berisha F, et al. IDH1 for chondrosarcoma of bone. Oncogenesis. 2016;5:e222. and IDH2 mutations are frequent events in central chondrosarcoma 27. Perez J, Decouvelaere AV, Pointecouteau T, Pissaloux D, Michot JP, Besse and central and periosteal chondromas but not in other mesenchymal A, et al. Inhibition of chondrosarcoma growth by mTOR inhibitor in an tumours. J Pathol. 2011;224(3):334–43. in vivo syngeneic rat model. PLoS ONE. 2012;7(6):e32458. 8. Damato S, Alorjani M, Bonar F, McCarthy SW, Cannon SR, O’Donnell P, 28. Terek RM, Healey JH, Garin‑ Chesa P, Mak S, Huvos A, Albino AP. p53 muta‑ et al. IDH1 mutations are not found in cartilaginous tumours other than tions in chondrosarcoma. Diagn Mol Pathol. 1998;7(1):51–6. central and periosteal chondrosarcomas and enchondromas. Histopa‑ 29. van Oosterwijk JG, Herpers B, Meijer D, Briaire‑ de Bruijn IH, Cleton‑ Jansen thology. 2012;60(2):363–5. AM, Gelderblom H, et al. Restoration of chemosensitivity for doxorubicin 9. Pansuriya TC, d’Adamo P, van Ruler MA, Kuijjer ML, Oosting J, et al. and cisplatin in chondrosarcoma in vitro: BCL‑2 family members cause Somatic mosaic IDH1 and IDH2 mutations are associated with enchon‑ chemoresistance. Ann Oncol. 2012;23(6):1617–26. droma and spindle cell hemangioma in Ollier disease and Maffucci 30. Zhang YX, van Oosterwijk JG, Sicinska E, Moss S, Remillard SP, et al. Func‑ syndrome. Nat Genet. 2011;43(12):1256–61. tional profiling of receptor tyrosine kinases and downstream signaling 10. van Maldegem AM, Gelderblom H, Palmerini E, Dijkstra SD, Gambarotti in human chondrosarcomas identifies pathways for rational targeted M, Ruggieri P, et al. Outcome of advanced, unresectable conventional therapy. Clin Cancer Res. 2013;19:3796. central chondrosarcoma. Cancer. 2014;120(20):3159–64. 31. Fertil B, Malaise EP. Intrinsic radiosensitivity of human cell lines is cor‑ 11. Kim DW, Seo SW, Cho SK, Chang SS, Lee HW, Lee SE, et al. Targeting of cell related with radioresponsiveness of human tumors: analysis of 101 pub‑ survival genes using small interfering RNAs (siRNAs) enhances radiosensi‑ lished survival curves. Int J Radiat Oncol Biol Phys. 1985;11(9):1699–707. tivity of Grade II chondrosarcoma cells. J Orthop Res. 2007;25(6):820–8. 32. Hamdi DH, Barbieri S, Chevalier F, Groetz JE, Legendre F, Demoor M, et al. 12. Moussavi‑Harami F, Mollano A, Martin JA, Ayoob A, Domann FE, Gitelis In vitro engineering of human 3D chondrosarcoma: a preclinical model S, et al. Intrinsic radiation resistance in human chondrosarcoma cells. relevant for investigations of radiation quality impact. BMC Cancer. Biochem Biophys Res Commun. 2006;346(2):379–85. 2015;15:579. 13. Molenaar RJ, Botman D, Smits MA, Hira VV, van Lith SA, Stap J, et al. Radio‑ 33. Li S, Chou AP, Chen W, Chen R, Deng Y, Phillips HS, et al. Overexpression protection of IDH1‑mutated cancer cells by the IDH1‑Mutant Inhibitor of isocitrate dehydrogenase mutant proteins renders glioma cells more AGI‑5198. Cancer Res. 2015;75(22):4790–802. sensitive to radiation. Neuro Oncol. 2013;15(1):57–68. 14. Menegakis A, De Colle C, Yaromina A, Hennenlotter J, Stenzl A, Scharpf 34. Peterse EFP, Niessen B, Addie RD, de Jong Y, Cleven AHG, Kruisselbrink M, et al. Residual gamma H2AX foci after ex vivo irradiation of patient AB, et al. Targeting glutaminolysis in chondrosarcoma in context of the samples with known tumour‑type specific differences in radio ‑respon‑ IDH1/2 mutation. Br J Cancer. 2018;118:1074. siveness. Radiother Oncol. 2015;116(3):480–5. 35. Peterse EF, van den Akker B, Niessen B, Oosting J, Suijker J, de Jong Y, et al. 15. Menegakis A, von Neubeck C, Yaromina A, Thames H, Hering S, Hennen‑ NAD synthesis pathway interference is a viable therapeutic strategy for lotter J, et al. gammaH2AX assay in ex vivo irradiated tumour specimens: chondrosarcoma. Mol Cancer Res. 2017;15:1714–21. a novel method to determine tumour radiation sensitivity in patient‑ 36. Chan SM, Thomas D, Corces‑Zimmerman MR, Xavy S, Rastogi S, Hong WJ, derived material. Radiother Oncol. 2015;116(3):473–9. et al. Isocitrate dehydrogenase 1 and 2 mutations induce BCL‑2 depend‑ 16. Scully SP, Berend KR, Toth A, Qi WN, Qi Z, Block JA. Marshall Urist Award. ence in acute myeloid leukemia. Nat Med. 2015;21(2):178–84. Interstitial collagenase gene expression correlates with in vitro invasion in 37. Grassian AR, Parker SJ, Davidson SM, Divakaruni AS, Green CR, Zhang human chondrosarcoma. Clin Orthop Relat Res. 2000;376:291–303. X, et al. IDH1 mutations alter citric acid cycle metabolism and increase 17. Gil‑Benso R, Lopez‑ Gines C, Lopez‑ Guerrero JA, Carda C, Callaghan RC, dependence on oxidative mitochondrial metabolism. Cancer Res. Navarro S, et al. Establishment and characterization of a continuous 2014;74(12):3317–31. human chondrosarcoma cell line, ch‑2879: comparative histologic and 38. Tateishi K, Wakimoto H, Iafrate AJ, Tanaka S, Loebel F, Lelic N, et al. genetic studies with its tumor of origin. Lab Invest. 2003;83(6):877–87. Extreme vulnerability of IDH1 mutant cancers to NAD+ depletion. Can‑ 18. Suijker J, Oosting J, Koornneef A, Struys EA, Salomons GS, Schaap FG, cer Cell. 2015;28(6):773–84. et al. Inhibition of mutant IDH1 decreases D‑2‑HG levels without affect ‑ 39. Bernhard EJ, Stanbridge EJ, Gupta S, Gupta AK, Soto D, Bakanauskas VJ, ing tumorigenic properties of chondrosarcoma cell lines. Oncotarget. et al. Direct evidence for the contribution of activated N‑ras and K ‑ras 2015;6(14):12505–19. oncogenes to increased intrinsic radiation resistance in human tumor cell 19. Corver WE, Demmers J, Oosting J, Sahraeian S, Boot A, Ruano D, et al. lines. Cancer Res. 2000;60(23):6597–600. ROS‑induced near ‑homozygous genomes in thyroid cancer. Endocr Relat 40. Schrage YM, Lam S, Jochemsen AG, Cleton‑ Jansen AM, Taminiau AH, Cancer. 2018;25(1):83–97. Hogendoorn PC, et al. Central chondrosarcoma progression is associated 20. Lapytsko A, Kollarovic G, Ivanova L, Studencka M, Schaber J. FoCo: a sim‑ with pRb pathway alterations: CDK4 down‑regulation and p16 overex ‑ ple and robust quantification algorithm of nuclear foci. BMC Bioinform. pression inhibit cell growth in vitro. J Cell Mol Med. 2009;13(9A):2843–52. 2015;16:392. 41. Asp J, Brantsing C, Benassi MS, Inerot S, Sangiorgi L, Picci P, et al. Changes 21. Naipal KA, Verkaik NS, Ameziane N, van Deurzen CH, Ter Brugge P, Meijers in p14(ARF) do not play a primary role in human chondrosarcoma tissues. M, et al. Functional ex vivo assay to select homologous recombination‑ Int J Cancer. 2001;93(5):703–5. deficient breast tumors for PARP inhibitor treatment. Clin Cancer Res. 42. Asp J, Sangiorgi L, Inerot SE, Lindahl A, Molendini L, Benassi MS, et al. 2014;20(18):4816–26. Changes of the p16 gene but not the p53 gene in human chondrosar‑ 22. Baranski Z, Booij TH, Cleton‑ Jansen AM, Price LS, van de Water B, Bovee coma tissues. Int J Cancer. 2000;85(6):782–6. JV, et al. Aven‑mediated checkpoint kinase control regulates proliferation 43. van Beerendonk HM, Rozeman LB, Taminiau AH, Sciot R, Bovee JV, and resistance to chemotherapy in conventional osteosarcoma. J Pathol. Cleton‑ Jansen AM, et al. Molecular analysis of the INK4A/INK4A‑ARF 2015;236(3):348–59. gene locus in conventional (central) chondrosarcomas and enchondro‑ 23. van Oosterwijk JG, Meijer D, van Ruler MA, van den Akker BE, Oost‑ mas: indication of an important gene for tumour progression. J Pathol. ing J, Krenacs T, et al. Screening for potential targets for therapy in 2004;202(3):359–66. mesenchymal, clear cell, and dedifferentiated chondrosarcoma reveals 44. Bosco EE, Wang Y, Xu H, Zilfou JT, Knudsen KE, Aronow BJ, et al. The Bcl‑2 family members and TGFbeta as potential targets. Am J Pathol. retinoblastoma tumor suppressor modifies the therapeutic response of 2013;182(4):1347–56. breast cancer. J Clin Invest. 2007;117(1):218–28. 24. Sibinga Mulder BG, Mieog JS, Handgraaf HJ, Farina Sarasqueta A, Vasen 45. Ertel A, Dean JL, Rui H, Liu C, Witkiewicz AK, Knudsen KE, et al. RB‑ HF, Potjer TP, et al. Targeted next‑ generation sequencing of FNA‑ derived pathway disruption in breast cancer: differential association with disease DNA in pancreatic cancer. J Clin Pathol. 2017;70(2):174–8. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 11 of 11 subtypes, disease‑specific prognosis and therapeutic response. Cell Cycle. 50. Nicolay BN, Gameiro PA, Tschop K, Korenjak M, Heilmann AM, Asara 2010;9(20):4153–63. JM, et al. Loss of RBF1 changes glutamine catabolism. Genes Dev. 46. Pollack A, Wu CS, Czerniak B, Zagars GK, Benedict WF, McDonnell TJ. 2013;27(2):182–96. Abnormal bcl‑2 and pRb expression are independent correlates of 51. Reynolds MR, Lane AN, Robertson B, Kemp S, Liu Y, Hill BG, et al. Control radiation response in muscle‑invasive bladder cancer. Clin Cancer Res. of glutamine metabolism by the tumor suppressor Rb. Oncogene. 1997;3(10):1823–9. 2014;33(5):556–66. 47. Sharma A, Comstock CE, Knudsen ES, Cao KH, Hess‑ Wilson JK, Morey 52. Frisch S, Timmermann B. The evolving role of proton beam therapy for LM, et al. Retinoblastoma tumor suppressor status is a critical deter‑ sarcomas. Clin Oncol (R Coll Radiol). 2017;29(8):500–6. minant of therapeutic response in prostate cancer cells. Cancer Res. 2007;67(13):6192–203. Publisher’s Note 48. Dyson NJ. RB1: a prototype tumor suppressor and an enigma. Genes Dev. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 2016;30(13):1492–502. lished maps and institutional affiliations. 49. Clem B. RB in glutamine metabolism. Oncoscience. 2014;1(5):304–5. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Clinical Sarcoma Research Springer Journals

Radiotherapy resistance in chondrosarcoma cells; a possible correlation with alterations in cell cycle related genes

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/radiotherapy-resistance-in-chondrosarcoma-cells-a-possible-correlation-st50aCQewY

References (59)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology
eISSN
2045-3329
DOI
10.1186/s13569-019-0119-0
Publisher site
See Article on Publisher Site

Abstract

Background: Conventional chondrosarcomas are malignant cartilage tumors considered radioresistant. Neverthe‑ less, retrospective series show a small but significant survival benefit for patients with locally advanced disease treated with radiotherapy. And, in daily practice when considered inoperable their irradiation is an accepted indication for proton beam radiotherapy. Therefore, we investigated the sensitivity of chondrosarcoma cell lines and ‑tissue samples towards radiotherapy and screened for biomarkers to identify predictors of radiosensitivity. Methods: Proliferation and clonogenic assays were performed in chondrosarcoma cell lines after γ‑radiation in combination with mutant IDH1 inhibitor AGI‑5198. In addition, glutathione levels were measured using mass spec‑ trometry. Chondrosarcoma tumor explants were irradiated after which γ‑H2AX foci were counted. Mutation analysis was performed using the Ion AmpliSeq Cancer Hotspot Panel and immunohistochemical staining’s were performed for P‑S6, LC‑3B, P53, Bcl‑2, Bcl‑xl and Survivin. Results were correlated with the number of γ ‑H2AX foci. Results: Chondrosarcoma cell lines were variably γ‑radiation resistant. No difference in radiosensitivity, nor glu‑ tathione levels was observed after treatment with AGI‑5198. Irradiated chondrosarcoma patient tissue presented a variable increase in γ‑H2AX foci compared to non‑radiated tissue. Samples were divided into two groups, high and low radioresistant, based on the amount of γ‑H2AX foci. All four highly resistant tumors exhibited mutations in the pRb pathway, while none of the less radioresistant tumors showed mutations in these genes. Conclusions: Chondrosarcoma cell lines as well as primary tumors are variably radioresistant, particularly in case of a defective Rb pathway. Whether selection for radiotherapy can be based upon an intact Rb pathway should be further investigated. Keywords: Chondrosarcoma, Radiotherapy resistance, Cell cycle alterations (2%), and periosteal chondrosarcoma (1%). Conventional Background chondrosarcoma can be subdivided into three histologi- Chondrosarcomas are a heterogeneous group of cartilage cal grades, representing the most important prognostic producing tumors of which the most prevalent subtype is factor. Atypical cartilaginous tumors (previously referred conventional chondrosarcoma (85%). More rare subtypes to as grade I) have a relatively good prognosis, exhibit- include dedifferentiated chondrosarcoma (10%), mesen - ing a 10  years survival of 83%. Probability of surviving chymal chondrosarcoma (2%), clear cell chondrosarcoma 10 years is about 64% for grade II chondrosarcomas and about 29% for grade III chondrosarcomas [1–3]. Based *Correspondence: j.v.m.g.bovee@lumc.nl on their location conventional chondrosarcoma can be Department of Pathology, Leiden University Medical Center, further subdivided into central (85%) and peripheral Albinusdreef 2, 2333 ZA Leiden, The Netherlands Full list of author information is available at the end of the article © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 2 of 11 (15%) subtypes. Rare chondrosarcoma subtypes dedif- Methods ferentiated, mesenchymal and clear cell chondrosarcoma Cell culture occur also centrally in the bone [4–6], while periosteal Conventional chondrosarcoma cell lines JJ012 (Grade II, chondrosarcoma is located and originates from the peri- IDH1 mutant) [16], SW1353 (Grade II, IDH2 mutant) osteal surface of the bone [3]. In this study, we focused (ATCC) and CH2879 (Grade III, IDH wild type) [17] were on central chondrosarcomas, as these are of highest cultured in RPMI-1640 (Gibco, Invitrogen Life-Tech- prevalence. Central chondrosarcomas carry mutations nologies, Scotland, UK) supplemented with 10% Fetal in the genes encoding the enzymes isocitrate dehydroge- Calf Serum, at 37 °C in a humidified incubator (5% CO ). nase 1 or -2 (IDH1 or IDH2) [7–9] in approximately 50% Short tandem repeat analysis was performed before and of cases, resulting in production of the oncometabolite after completion of experiments to confirm identity of d-2-hydroxyglutarate (D-2HG). the cell lines by using the Cell ID Gene Print 10 system The only curative treatment for patients with chon - (Promega Benelux BV, Leiden, The Netherlands). Myco - drosarcoma is radical surgery, since they are considered plasma tests were performed on a regular basis. resistant towards conventional chemo- and radiotherapy. For this reason, linear accelerator based (conventional) Compounds radiotherapy is only given to patients with metastatic The specific mutant IDH1 inhibitor AGI-5198 (14624, disease, incomplete resection or inoperable tumors in Cayman Chemicals, Michigan, USA) was dissolved in difficult sites like the base of skull or the sacrum [2, 10]. DMSO according to the manufacturer’s instructions and Several causes have been suggested for chondrosarcoma stored in − 20 °C. AGI-5198 was used at a concentration radioresistance, such as the relative hypoxic chondro- of 10  µM since our group previously showed that this sarcoma microenvironment impairing the induction leads to a complete inhibition of D-2HG production [18]. of reactive oxygen species (ROS) and DNA damage, or (2R)-Octyl-α-hydroxyglutarate (16366, Cayman Chemi- their relatively slow dividing rate  (though not applicable cals), a cell-permeable derivative of D2-HG, was freshly to high grade tumors). Additionally, knock down of anti- dissolved in PBS before use and used at a final concentra - apoptotic Bcl-2 family members Bcl-2, Bcl-xl and XIAP tion of 250 µM. has been shown to increase cell death after radiation in chondrosarcoma cell lines [11]. Furthermore, restora- Clonogenic assays tion of p16 activity sensitized chondrosarcoma cell lines Chondrosarcoma cell lines SW1353 and JJ012 were to radiotherapy [12]. More recently, a role for IDH muta- plated in optimal cell densities to obtain sufficient colo - tions has been found in predicting radiotherapy response nies after treatment. Cells were allowed to adhere over- in glioma due to altered redox responses in IDH mutant- night before treatment with a wide range (0, 1, 2, 4 or compared to wild type cells thereby enhancing radiosen- 6 Gy) of γ-radiation using a Cs source (YXLON, Comet sitivity [13]. technologies USA). Colony formation was assessed after A recent retrospective analysis suggested that a sub- 14  days of treatment by fixing and staining with 0.5% group of chondrosarcoma patients with locally advanced, crystal violet/6% glutaraldehyde. Colonies were counted unresectable disease showed a favorable overall survival manually and the surviving fraction (SF) was calcu- after conventional radiotherapy [10]. As chondrosarcoma lated by normalizing towards the plating efficiency of patients are not commonly treated with radiotherapy, untreated controls. The α/β ratios were calculated based prognostic biomarkers for radiosensitivity were inves- on the linear quadratic model. α/β ratios describe the tigated using in  vitro and ex  vivo methods. Therefore, slope of the cell-survival curve; acute responding tissues the aim of our study was to examine whether sensitivity show a higher ratio compared to late responding tissues. to γ-radiation can be observed in central conventional chondrosarcoma cell lines by determining clonogenic Viability assay survival and γ-H2AX  foci induction after radiation. In Chondrosarcoma cells were counted using a Burker Turk addition, radiosensitivity of chondrosarcoma patient counting chamber and seeded in optimized cell densities samples was determined by counting γ-H2AX foci after in 96 well plates. After attachment overnight cells were ex vivo radiation [14, 15]. Mutation and expression analy- irradiated with increasing doses. Seventy-two hours after ses were performed to investigate prognostic biomarkers radiation cell viability was measured using PrestoBlue for radiosensitivity which could then be used to select viability reagent (Invitrogen, Life-Technologies, Scot- patients for radiotherapy. land, UK) according to the manufacturer’s instructions. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 3 of 11 Fluorescence was measured at 590  nM using a Wallac with γ-H2AX results. Cells were harvested by scrap- plate reader (Victor3V, 1420 multilabel counter, Perkin ing after 1  h. For each individual replicate 1x10 cells Elmer, the Netherlands). Experiments were performed were fixed overnight using ice-cold acetone. The super - three times in triplicate. natant was then removed and dH O added (10 µL), fol- lowed by a freeze–thaw cycle (10 s dry ice-cooled MeOH, Cell proliferation measurement 2  min 37  °C heating block). After 2 freeze thaw cycles To measure cell proliferation in real time the RTCA glutathione was extracted with EtOH (50 µL) and the xCELLigence system (Roche Applied Sciences, Almere, supernatant was mixed 1:1 (v:v) with matrix solution the Netherlands) was used. JJ012, SW1353 and CH2879 (N-(1-naphthyl) ethylenediamine dihydrochloride in 70% cells were pre-treated for 72  h with either 10  µM AGI- MeOH) for analysis. Samples were measured using a 9.4T 5198 or 0.1% DMSO in T25 flasks. Thereafter, cells were SolariX XR Fourier transform ion cyclotron resonance plated in the presence of AGI-5198 or DMSO and left (FT-ICR) mass spectrometer (Bruker Daltonics, Bremen, for 20 h to attach to the wells before low-dose (2 Gy) or Germany) equipped with a CombiSource and a dynam- high dose (4  Gy) γ-radiation. After 125  h the measure- ically harmonized ParaCell . The mass spectrometric ment was stopped, and the results were analyzed using analysis was performed in negative ion mode in the m/z- RTCA software. Normalized cell index was calculated range 101.7–1000  Da, with a 1  M data point transient by normalizing against the amount of cells in each well (0.3670  s duration) and an estimated resolving power after 20 h. Experiments were performed in duplicate and of 66,000 at m/z 400. Spectra were acquired of 10 aver- repeated at least two times. aged scans, acquired in a random pattern covering the entire spot, where each scan consisted of 500 laser shots γ‑H2AX staining of chondrosarcoma cell lines (laser power 70%, 1000 Hz frequency). Spectra were ana- Optimal cell amounts for JJ012 (550,000 cells), SW1353 lyzed using DataAnalysis version 4.2 (Bruker Daltonics, (500,000 cells) and CH2879 (1,000,000 cells) cell lines Bremen, Germany). Using a custom algorithm, intensi- were cultured on alcohol sterilized APES coated slides ties of the compounds of interest were extracted and in a culture dish. IDH mutant chondrosarcoma cell lines exported to Excel 2016 (Microsoft) for further analysis. SW1353 and JJ012 were pre-treated with 10  µM AGI- 5198 or 0.1% DMSO for 72 h, in order to inhibit mutant Chondrosarcoma tissue samples activity, and IDH wild type chondrosarcoma cell line Chondrosarcoma patient samples (n = 9, see Tables 1 and CH2879 was pre-treated with 250  µM D-2HG for 24  h, 2) were obtained at surgery and collected in RPMI- in order to mimic mutant activity. After overnight attach- 1640 medium supplemented with 20% FBS and 1% P/S. ment cells were radiated to 5  Gy, in order to achieve All samples were coded according to the Dutch code of highest response without significant damage to sample, proper secondary use of human material as accorded by and fixed after 2 or 24 h with buffered 4% formaldehyde the Dutch society of pathology (Federa), and as approved (VWR Chemicals) at 37  °C. The procedure was per - by the LUMC ethical board (B17.019). Samples were pro- formed as described previously [19]. Slides were stained cessed by manual dissection into approximately 2  mm with γ-H2AX antibody (JBW301, Millipore) for 60  min, fragments and incubated on a rotation shaker at 37 °C in washed, then a mixture of Alexa fluor 657 labelled sec - a humidified incubator (5% CO ). The tissue was treated ondary antibody and 0.5 µM Hoechst 33342 was applied to 5 Gy, consistent with the cell line treatment, and incu- for 60  min. Slides were covered with ProLongGold anti- bated for 2 or 24  h on a rotation shaker at 37  °C in a fade reagent and were examined using a confocal micro- humidified incubator as described previously [21]. After scope by taking tile scans of areas of interest using ZEN overnight fixation with formalin, tissue was processed, light software. Pictures were exported in TIF format and embedded in paraffin and sections were cut and stained loaded into FoCo, a previously published format using for γ-H2AXfoci. Image J and Matlab software [20]. γ‑H2AX staining and quantification of paraffin embedded Mass spectrometry measurements of glutathione chondrosarcoma tissue Chondrosarcoma cell lines JJ012, SW1353 and CH2879 Sections were deparaffinized and antigen retrieval was were seeded in triplicate in T25 culture flasks and performed using DAKO antigen retrieval solution pH allowed to adhere overnight. IDH mutant chondrosar- 9.0 by boiling the slides in a microwave for 12 min. After coma cell lines JJ012 and SW1353 were pre-treated with cooling, slides were washed with TBS and blocking was 10  µM AGI-5198 for 72  h, while IDH wildtype cell line performed using TBS/5% normal goat serum/1% BSA CH2879 was pre-treated with 250  µM D-2HG for 24  h. and 0.2% Triton-x-100. Anti γ-H2AX antibody (Millipore Samples were treated with 0 or 5 Gy, in order to correlate 2310355) was applied in a 1:1000 dilution and incubated de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 4 of 11 Table 1 Patient samples treated with 5 gy of radiation Sample Grade IDH mutation status Other alterations Mean + SEM basic foci level Mean + SEM Difference P value after 5gy between means radiation L5213 1 WT 3.053 ± 0.6277 14.28 ± 1.737 11.23 ± 1.733 < 0.0001 n = 38 n = 32 L5678 3 IDH1 R132C 0.2295 ± 0.07165 8.239 ± 1.477 8.01 ± 1.284 < 0.0001 n = 61 n = 46 L5676 2 IDH2 R172S 3.625 ± 0.9297 9 ± 1.493 5.375 ± 1.784 0.003712 n = 32 n = 34 L5541 Dediff IDH1 R132S 2.378 ± 0.5747 6.702 ± 1.225 4.324 ± 1.212 0.0005 n = 90 n = 57 L5546 2 WT CDKN2A deletion 3.982 ± 0.9036 6.846 ± 1.172 2.864 ± 1.468 0.0537 n = 56 n = 52 L5302 3 WT CDKN2A deletion 2.246 ± 0.4531 5.082 ± 0.7412 2.836 ± 0.9108 0.0023 n = 61 n = 73 Foci were counted 2 h after treatment Table 2 Patient samples treated with 5 gy of radiation Sample Grade IDH mutation status Other alterations Mean + SEM basic Mean + SEM Difference P value foci level after 5gy radiation between means L5847 2 WT 2.644 ± 0.7399 7.467 ± 1.526 4.823 ± 1.501 0.0018 n = 59 n = 30 L5718 3 IDH1 R132L CDKN2A deletion 2.437 ± 0.584 4.088 ± 0.6256 1.651 ± 0.8659 0.0580 n = 87 n = 102 L5850 2 WT RB1 deletion 0.6327 ± 0.1409 1.55 ± 0.1809 0.9173 ± 0.2463 0.0002 n = 98 n = 140 Foci were counted 24 h after treatment overnight at 4 °C. Slides were washed with TBS and incu- standard laboratory methods as previously described bated with secondary labelled GAM Alexa fluor 647 [22]. All slides were scored by two independent observ- antibody. Mounting was performed with ProLongGold ers (JVMGB, YDJ) using a scoring system assessing stain- containing DAPI (Thermofisher Scientific, Waltham, ing intensity (0 = no, 1 = weak , 2 = moderate, 3 = strong) Massachusetts, USA). Positive control slides consisted as well as percentage of staining (0 = no, 1 = 1–24%, of JJ012 cells radiated at 4  Gy as these displayed a con- 2 = 25–49%, 3 = 50–74%, 4 = 75–100%) [23]. This scoring sistent, high γ-H2AX signal. The cells were embedded in system was used for all the different proteins. ™ ™ paraffin with the use of Shandon Cytoblock Cell Block Preparation system (Thermofisher Scientific). Slides Mutation analysis were analyzed using a confocal microscope by taking tile Frozen tissue of chondrosarcoma patients was col- scans of areas of interest using ZEN light software. Pic- lected at surgery and DNA was isolated using the wiz- tures were exported in TIF format and loaded into FoCo, ard genomic DNA purification kit (Promega, Madison, a previously published format using Image J and Matlab WI, USA) according to the manufacturer’s instructions. software [20]. The settings to quantify the foci were first Samples were subjected to next generation sequenc- optimized using the positive control slides, after which ing using the in-house developed Ion AmpliSeq Can- the complete series was analysed using the same settings. cer Hotspot Panel v2  (Life Technologies, Thermo Fisher Scientific, USA, catalog number 4475346) as described previously [24, 25]. The following genes and regions were Immunohistochemical staining analysed on an Ion Torrent PGM/Proton for all sam- Protein expression of Bcl-2, Bcl-xl, Survivin, P-S6, ples (exons between brackets): ARAF (7,10); CTNNB1 P53 and LC-3B were evaluated in irradiated con- (1,2,4,7,8,12,15); KRAS (2–4); NRAS (2–4); HRAS (2–3); trol patient samples (see Additional file  1: Table  S1). BRAF (11,15); EGFR (3,7,15,18–21); GNAQ (5); GNAS Immunohistochemistry was performed according to (8–9); H3F3A (2); H3F3B (2); IDH1 (4); IDH2 (4); KIT de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 5 of 11 (2,9-18); MYD88 (3b,5), MUTYH (7,13); PDGFRA in repairing double strand breaks and more sensitive to (12,14,15,18,23); PIK3CA (2,5,6–10,14,18,21); POLE radiation. (9,11,13,14); RET (10–12,15,16); TP53 (1–11). In addi- tion hotspot regions in the following genes were also γ‑Radiosensitivity of chondrosarcoma cell lines included: ABL1; AKT1; ALK; APC; ATM; CARD11; is not correlated with IDH mutation status CD79A; CD79B; CDH1; CDKN2A; CSF1R; CTNNB1; Inhibition of D-2HG production by JJ012 cells using ERBB2; ERBB4; EZH2; FBXW7; FGFR1; FGFR2; FGFR3; the mutant IDH1 inhibitor AGI-5198 did not show any FLT3; GNA11; HNF1A; JAK2; JAK3; KDR; MET; MLH1; difference in radiosensitivity (Additional file  1: Figure MPL; NOTCH1; NPM1; PTEN; PTPN11; RB1; SMAD4; S1A). In addition no differences were observed in prolif - SMARCB1; SMO; SRC; STK11; VHL. Libraries were pre- eration capacity and foci formation between cells treated pared with 10 ng of genomic DNA, and each sample was with AGI-5198 and radiation or cells treated with radia- uniquely barcoded. Ion Proton chips were prepared using tion only (Additional file  1: Figure S1B, C). Since previ- the Ion Chef System. The unaligned bam files generated ous reports [13] suggest that IDH mutant cells have a by the Proton sequencer were mapped against the human reduced capability of producing GSH, GSH levels were reference genome (GRCh37/hg19) using the TMAP 5.0.7 measured 1  h after radiation in combination with AGI- software with default parameters (https ://githu b.com/ 5198, for the mutant cell lines or D-2HG (oncometabo- ionto rrent /TS). The Torrent Variant Caller (TVC)-5.0.2 lite resulting from an IDH mutation) for the wild type was used for variant calling and variant interpretation cell line. The most radioresistant SW1353 cells displayed was done using Geneticist Assistant (http://softg eneti the highest baseline levels of GSH, followed by CH2879 cs.com/Genet icist Assis tant_2.php) as described. Chro- and JJ012, however no differences were observed mosomal gains and deletions were assessed by calculat- between different treatment conditions, indicating that ing the median base coverage per amplicon, which was GSH levels are not influenced by D-2HG inhibition normalized using the median value of all amplicons in (Additional file  1: Figure S1D). These results suggest that that sample. IDH and D-2HG do not play a role in chondrosarcoma radiosensitivity. Results Chondrosarcoma cell lines are variably resistant Chondrosarcoma samples that are more resistant to γ‑radiation towards radiotherapy have an increased incidence Clonogenic assay  SF2 values (surviving fraction of cells of mutations in cell cycle regulators after 2  Gy radiation) suggested that JJ012 cells (SF2 Chondrosarcoma patient samples showed an increase in 0.55) were more radiosensitive compared to SW1353 γ-H2AX foci after 5 Gy radiation treatment. In Tables  1, (SF2 0.88) (Fig.  1a). JJ012 cells showed a high α/β ratio 2 and Additional file  1: Figure S2 quantified results are of 38.47, in contrast to SW1353 with a very low α/β shown for nine chondrosarcoma explant tissue samples ratio of − 0.75, suggesting that SW1353  cells may ben- analyzed 2 or 24 h after treatment with 5 gy of radiation. efit from hypofractionation. CH2879 cells were unable Radiation response was heterogeneous across the sam- to form colonies and were therefore not included in this ples; the largest difference between control and radiated assay. In addition, dose response curves were made to samples was 11.2 foci (sample L5213, Fig.  2a, b), while assess viability after 72  h. JJ012 cells were most sensi- the smallest difference was 0.9 foci. Samples analyzed tive followed by the CH2879 and SW1353 cells (Fig. 1b). 24 h after radiation treatment in general showed a lower An X-CELLigence assay was performed to determine amount of foci compared to samples analyzed 2  h after the effect on proliferation in real time after 2 or 4  Gy radiation treatment. Samples that showed a difference of of radiation. All cell lines showed a reduction in prolif- 4 or more foci were subjected to the less radioresistant eration after radiation. However, CH2879 cells showed group, while samples that showed < 4 foci difference were a similar response when irradiated to 2 or 4  Gy, while designated as radioresistant for further analysis. This JJ012 and SW1353 showed a dose response relation- cutoff value was taken for both the 2  h and 24  h groups ship (Fig.  1c). The amount of double strand breaks was and both time points were analyzed together since sam- determined by quantifying γ-H2AX  foci 2 and 24  h ple size impeded separate analyses. The division between after radiation treatment. JJ012, SW1353 and CH2879 more or less radioresistant was made prior to mutation cells exhibited an increase in foci after 2  h, which was analysis. No correlation was observed between histo- decreased again after 24 h in SW1353 and CH2879 cells logical grade or IDH mutation status and the amount of (Fig.  1d–e). Conversely JJ012 cells still showed a sub- γ-H2AX  foci, consistent with the results obtained in the stantial significant amount of foci compared to non- chondrosarcoma cell lines (Table  1). Mutation analysis radiated cells, indicating this cell line is less proficient was performed on 50 known cancer related genes and Dose (Gy) Dose (Gy) CH2879 SW1353 CH2879 JJ012 Time-Interval (Hour) JJ012 Time-Interval (Hour) SW1353 Time-Interval (Hour) de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 6 of 11 ab 0.1 SW1353 SW1353 SF2: 0.88, α/β ratio: -0.75 JJ012 JJ012 SF2: 0.55, α/β ratio: 38.47 CH2879 01 2 56 0 10 c d 10 250 250 * 250 ns 0 Gy **** **** **** **** 2 Gy 200 200 200 4 Gy 150 ns 150 150 **** **** 100 100 100 50 50 50 0 0 0 0255075 100 125 0 Gy 5 Gy 5 Gy 0 Gy 5 Gy 5 Gy 0 Gy 5 Gy 5 Gy 0 h 2 h 24 h 0 h 2 h 24 h 0 h 2 h 24 h e CH2879 JJ012 SW1353 0 Gy 2 Gy 4 Gy 0 Gy 0 h 0255075 100 125 5 Gy 2 h 0 Gy 2 Gy 4 Gy 5 Gy 24 h 0255075 100 125 Fig. 1 Chondrosarcoma cell lines exhibit variable γ‑radioresistance. a Colony forming assay of SW1353 and JJ012 chondrosarcoma cells. b Viability of JJ012, SW1353 and CH2879 cells, 72 h after treatment with increasing doses of γ‑radiation measured using presto blue viability reagent. c Normalized cell index of CH2879, JJ012 and SW1353 cells irradiated to 0 (black), 2 (blue) or 4 Gy (green) radiation. d, e γH2AX foci/cell in CH2879, JJ012 and SW1353 cells 2 and 24 h after 5 Gy radiation.****P values < 0.0001, *P values < 0.05 Normalized Cell index Normalized Cell index Normalized Cell index Surviving fraction Foci/cell Viability (%) de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 7 of 11 expression of Bcl-2, Bcl-xl, Survivin, P-S6, LC3B and P53, mutant IDH1 using AGI-5198 did not lead to any changes previously identified to play a role in chondrosarcoma in radiosensitivity, nor in altered GSH levels. Also, no dif- [26–30], was determined using immunohistochemistry. ference was observed in γ-H2AX foci formation between No significant difference was observed in protein expres - IDH mutant and IDH wild type chondrosarcoma explants. sion of selected markers (Fig. 2c, Additional file  1: Figure This indicates that, unlike the observations in gliomas, S3), however three CDKN2A deletions (3/4) and one RB1 IDH1 or IDH2 mutations in chondrosarcoma do not cor- deletion (1/4) were found in the highly radioresistant relate with radiosensitivity. This is in line with our previ - group and none in the less radioresistant group indicat- ous results in which we also did not find any correlation ing that a defective Rb pathway may be able to impair between IDH1 or -2  mutation status in sensitivity for the response to γ-radiation in chondrosarcoma (Fig.  2d, inhibitors of glutaminolysis [34], NAD synthesis [35] or Tables 1 and 2). Interestingly a mutation in CDKN2A was Bcl-2 family members in chondrosarcoma cells, while in also identified in the SW1353 cell line (Additional file  1: other tumor types there was a clear difference in sensitiv - Table S2), which is the most radioresistant cell line. ity [36–38]. This difference indicates that IDH mutations may have a tissue specific effect rather than a more gen - Discussion eral effect in different tumor types. Chondrosarcomas are relatively radioresistant tumors Our results suggest that deletions in cell cycle regu- and therefore, after multidisciplinary discussions, very lators CDKN2A and RB1  are associated with increased few of these patients are offered radiotherapy [10]. In this radioresistance in chondrosarcoma explant tissue. In study we investigated the sensitivity of chondrosarcoma line with this, the most resistant SW1353 cells harbored cells and tumor explants to γ-radiation. In addition, we a mutation in the splice site region of CDKN2A  in screened for biomarkers that could select patients that addition to mutations in TP53 and a kRAS mutation might benefit from γ-radiation. Chondrosarcoma cell [39]. CDKN2A is a gene encoding p16(INK4A) and lines showed a heterogeneous response to radiotherapy p14(ARF), which are two tumor suppressor proteins with relatively high SF2 values [31]. JJ012 cells presented controlling the cell cycle. P16 inhibits CDK4 and CDK6, a SF2 value of 0.55, while SW1353 were more resistant two inhibitors of Rb1 phosphorylation, while p14- showing a SF2 value of 0.88. Compared to cell lines of ARF protects p53 from being broken down by inhibit- other tumor types, these values are relatively high, con- ing MDM2. Alterations in the pRB pathway have been firming chondrosarcoma  radio resistance. A previous described in the majority of high grade conventional published study by Hamdi et  al. showed a SF2 value of chondrosarcomas [40–43]. Previous studies in chon- 0.64 for the SW1353 cell line [32]. This difference might drosarcoma cell lines (CS-7, CS-8, CS-9) showed that be attributed to the fact that different methods were restoring P16 expression, and thereby increasing Rb1 used to perform the clonogenic assay. In our study the phosphorylation, resulted in an increased radiosensi- cells were seeded prior to radiation, while in the study tivity [12], in line with our results. Although we see a of Hamdi et  al. the cells were subconfluently seeded in clear difference in the amount of foci in tumors with culture flasks and subsequently radiated and seeded. In and without deletions in CDKN2A or RB1, this study addition the differences in culture conditions (normoxic is based on a small heterogenous group of chondrosar- vs hypoxic) can very well influence how cells respond comas and measurements are taken after 2 or 24  h. In to radiation treatment. Quantification of γ-H2AX foci addition the threshold of 4 foci/cell is taken arbitrar- showed that SW1353 and CH2879 cells were able to ily. This complicates making firm conclusions based on repair double strand breaks within 24  h after radiation this data and further studies should focus on; includ- while JJ012 cells were less capable of doing so, in line with ing more patients and analyze amount of foci after 24 h the lower SF2 values observed in JJ012 cells. to determine the damage remaining after DNA repair. In contrast to published studies in glioma [13, 33], we In addition when radiotherapy is included in the treat- did not observe a correlation between radiosensitivity and ment plan, follow the response towards radiation and IDH mutation status in chondrosarcoma. Inhibition of correlate this towards mutation status. (See figure on next page.) Fig. 2 γ‑Radioresistance in chondrosarcoma tissues correlates with mutations in cell cycle related genes. a L5213 chondrosarcoma tissue sample showing a large induction of γH2AX foci after5 Gy(right) compared to controls (right). b Amount of foci/cell in sample L5213 after 5 Gy radiation compared to controls. c Protein expression of Bcl‑2, Bcl‑xl, Survivin, P ‑S6, LC3B and P53 in highly radioresistant (< 4 foci difference) compared to less radioresistant (> 4 foci difference) chondrosarcomas. d Difference between mean amount of γH2AX foci before and after radiation in chondrosarcomas with and without alterations in RB1 or CDKN2A. Each dot represents one sample. Samples in green indicate 2 h, while samples in blue indicate samples 24 h after radiation de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 8 of 11 L5213 P = 1.217e-008 0 Gy 5 Gy Bcl-xl Survivin Bcl-2 8 10 P=0.07331 P=0.5496 P=0.4825 4 4 2 2 0 0 P-S6 LC-3B P53 8 5 4 P=0.7128 P=0.6195 6 3 4 2 2 P=0.8383 1 0 0 -1 Mean amount of foci P=0.01768 >4 foci difference <4 foci difference >4 foci difference >4 foci difference <4 foci difference <4 foci difference >4 foci difference >4 foci difference <4 foci difference <4 foci difference no cell cycle alterations deletion of CDKN2A or RB >4 foci difference <4 foci difference difference between means IHC score IHC score IHC score foci/cell IHC score de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 9 of 11 Abbreviations In contrast to our findings in chondrosarcoma, deletion D‑2HG: d ‑2‑hydroxyglutarate; IDH1 or IDH2: isocitrate dehydrogenase 1 or ‑2; of RB1 has been described to enhance radiosensitivity in ROS: reactive oxygen species; SF: surviving fraction. breast, prostate and bladder cancer [44–47]. This observa - Acknowledgements tion might therefore be tissue and context specific; only a We would like to thank Matty Meijers and MaaikeVreeswijk for technical assis‑ limited number of cancer types have been investigated. In tance with the tumor culture model and Ben Floot ( The Netherlands Cancer addition, pRb has multiple functions, not only in the cell Institute, Amsterdam) for his help with the colony formation assay. In addition we would like to thank Liam McDonnell for his advice and discussions on cycle but also in chromatin organization, transcription the optimization of the glutathione measurements. Also we would like to patterns, metabolic pathways and the proteome [48]. Sev- thank Francois Chevalier and Yannick Saintigny for fruitful discussions. We are eral studies found that loss of pRB expression resulted in grateful to Dr. JA Block (Rush University Medical Centre, Chicago, IL, USA), who provided us with the JJ012 cell line and Professor A Llombart Bosch (University an increase in glutamine consumption and an increased of Valencia, Spain) for the CH2879 cell line. glutamine incorporation into GSH. Upon RB knock down cells increased the expression of glutamine transport- Authors’ contributions YDJ, RH and JB were involved in the design of the experiments and writing the ers and upregulated glutaminase activity, indicating that manuscript. YDJ, MI and IP performed the mass spectrometry experiments. the pRB pathway can regulate glutamine metabolism MI, IP and BH analysed the mass spectrometry experiments. IB, AK, SV and YDJ and that cells with inactivated pRB are potentially more performed all in vitro cell culture experiments. IB and AMCJ performed and analysed the sequencing experiments. All authors read and approved the final sensitive for targeted anti-glutamine treatment [49–51]. manuscript. Interestingly, we recently reported that interfering with glutamine metabolism can be a therapeutic target for Funding This work was financially supported by the Dutch Cancer Society (UL2010‑ high grade chondrosarcoma [34]. The SW1353 chondro - 4873 and UL2013‑6103) and EuroSarc under Grant Agreement 278742. sarcoma cell line was particularly sensitive for inhibition of glutaminase, which is in line with the high basal GSH Availability of data and materials All data generated or analysed during this study are included in this published levels measured in this cell line. We can hypothesize that article (and its additional files). the mutation in CDKN2A observed in this cell line might contribute to the glutamine dependence, however more Ethics approval and consent to participate All samples were coded according to the Dutch code of proper secondary use research is needed to further investigate this. of human material as accorded by the Dutch society of pathology (Federa), In addition to conventional radiotherapy, research has and as approved by the LUMC ethical board (B17.019). been focusing increasingly on proton [ H] and carbon Consent for publication [ C] therapy, which have several advantages compared Not applicable. to γ-radiation. Both these beam qualities have dose dis- tribution advantages, causing less damage to surrounding Competing interests The authors declare that they have no competing interests. healthy tissues, making it possible to deliver higher doses to the tumor. Chondrosarcomas of the skull base and Author details spine are increasingly treated with proton beam radiation, Department of Pathology, Leiden University Medical Center, Albinusdreef 2, 2333 ZA Leiden, The Netherlands. Center for Proteomics and Metabolomics, showing promising results [52]. Whether chondrosar- Leiden University Medical Center, Leiden, The Netherlands. Department coma radiosensitivity differs between photon (γ-) beams of Radiation Oncology, Leiden University Medical Center, Leiden, The Nether‑ and proton beams is subject for further investigation. lands. Department of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, The Netherlands. Received: 27 February 2019 Accepted: 20 May 2019 Conclusion In conclusion, this study confirms a heterogeneous radio - sensitivity of chondrosarcoma cell lines and fresh explant tissue. In addition, we identified alterations in CDKN2A/ References 1. Evans HL, Ayala AG, Romsdahl MM. Prognostic factors in chondrosarcoma RB1 as predictors of decreased double strand break for- of bone: a clinicopathologic analysis with emphasis on histologic grad‑ mation after radiotherapy in chondrosarcoma tissue, and ing. Cancer. 1977;40(2):818–31. although further research is necessary in a larger group, 2. Gelderblom H, Hogendoorn PCW, Dijkstra SD, van Rijswijk CS, Krol AD, Taminiau AH, et al. The clinical approach towards chondrosarcoma. this suggests that these patients might benefit less from Oncologist. 2008;13(3):320–9. radiation therapy. 3. Hogendoorn PCW, Bovee JVMG, Nielsen GP. Chondrosarcoma (grades I‑III), including primary and secondary variants and periosteal chon‑ drosarcoma. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO classification of tumours of soft tissue and bone. Lyon: IARC; Additional file 2013. p. 264–8. 4. Inwards C, Hogendoorn PCW. Dedifferentiated chondrosarcoma. In: Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO clas‑ Additional file 1. Supplementary material. sification of tumours of soft tissue and bone. Lyon: IARC; 2013. p. 269–70. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 10 of 11 5. McCarthy EF, Hogendoorn PCW. Clear cell chondrosarcoma. In: Mertens F, 25. van Riet J, Krol NMG, Atmodimedjo PN, Brosens E, Jansen M, et al. SNPitty: editor. WHO classification of tumours of soft tissue and bone. Lyon: IARC; an intuitive web application for interactive B‑allele frequency and copy 2013. p. 273–4. number visualization of next‑ generation sequencing data. J Mol Diagn. 6. Nakashima Y, de Pinieux G, Ladanyi M. Mesenchymal chondrosarcoma. In: 2018;20(2):166–76. Fletcher CDM, Bridge JA, Hogendoorn PCW, Mertens F, editors. WHO clas‑ 26. de Jong Y, van Oosterwijk JG, Kruisselbrink AB, Briaire‑ de Bruijn IH, Agrogi‑ sification of tumours of soft tissue and bone. Lyon: IARC; 2013. p. 271–2. annis G, Baranski Z, et al. Targeting survivin as a potential new treatment 7. Amary MF, Bacsi K, Maggiani F, Damato S, Halai D, Berisha F, et al. IDH1 for chondrosarcoma of bone. Oncogenesis. 2016;5:e222. and IDH2 mutations are frequent events in central chondrosarcoma 27. Perez J, Decouvelaere AV, Pointecouteau T, Pissaloux D, Michot JP, Besse and central and periosteal chondromas but not in other mesenchymal A, et al. Inhibition of chondrosarcoma growth by mTOR inhibitor in an tumours. J Pathol. 2011;224(3):334–43. in vivo syngeneic rat model. PLoS ONE. 2012;7(6):e32458. 8. Damato S, Alorjani M, Bonar F, McCarthy SW, Cannon SR, O’Donnell P, 28. Terek RM, Healey JH, Garin‑ Chesa P, Mak S, Huvos A, Albino AP. p53 muta‑ et al. IDH1 mutations are not found in cartilaginous tumours other than tions in chondrosarcoma. Diagn Mol Pathol. 1998;7(1):51–6. central and periosteal chondrosarcomas and enchondromas. Histopa‑ 29. van Oosterwijk JG, Herpers B, Meijer D, Briaire‑ de Bruijn IH, Cleton‑ Jansen thology. 2012;60(2):363–5. AM, Gelderblom H, et al. Restoration of chemosensitivity for doxorubicin 9. Pansuriya TC, d’Adamo P, van Ruler MA, Kuijjer ML, Oosting J, et al. and cisplatin in chondrosarcoma in vitro: BCL‑2 family members cause Somatic mosaic IDH1 and IDH2 mutations are associated with enchon‑ chemoresistance. Ann Oncol. 2012;23(6):1617–26. droma and spindle cell hemangioma in Ollier disease and Maffucci 30. Zhang YX, van Oosterwijk JG, Sicinska E, Moss S, Remillard SP, et al. Func‑ syndrome. Nat Genet. 2011;43(12):1256–61. tional profiling of receptor tyrosine kinases and downstream signaling 10. van Maldegem AM, Gelderblom H, Palmerini E, Dijkstra SD, Gambarotti in human chondrosarcomas identifies pathways for rational targeted M, Ruggieri P, et al. Outcome of advanced, unresectable conventional therapy. Clin Cancer Res. 2013;19:3796. central chondrosarcoma. Cancer. 2014;120(20):3159–64. 31. Fertil B, Malaise EP. Intrinsic radiosensitivity of human cell lines is cor‑ 11. Kim DW, Seo SW, Cho SK, Chang SS, Lee HW, Lee SE, et al. Targeting of cell related with radioresponsiveness of human tumors: analysis of 101 pub‑ survival genes using small interfering RNAs (siRNAs) enhances radiosensi‑ lished survival curves. Int J Radiat Oncol Biol Phys. 1985;11(9):1699–707. tivity of Grade II chondrosarcoma cells. J Orthop Res. 2007;25(6):820–8. 32. Hamdi DH, Barbieri S, Chevalier F, Groetz JE, Legendre F, Demoor M, et al. 12. Moussavi‑Harami F, Mollano A, Martin JA, Ayoob A, Domann FE, Gitelis In vitro engineering of human 3D chondrosarcoma: a preclinical model S, et al. Intrinsic radiation resistance in human chondrosarcoma cells. relevant for investigations of radiation quality impact. BMC Cancer. Biochem Biophys Res Commun. 2006;346(2):379–85. 2015;15:579. 13. Molenaar RJ, Botman D, Smits MA, Hira VV, van Lith SA, Stap J, et al. Radio‑ 33. Li S, Chou AP, Chen W, Chen R, Deng Y, Phillips HS, et al. Overexpression protection of IDH1‑mutated cancer cells by the IDH1‑Mutant Inhibitor of isocitrate dehydrogenase mutant proteins renders glioma cells more AGI‑5198. Cancer Res. 2015;75(22):4790–802. sensitive to radiation. Neuro Oncol. 2013;15(1):57–68. 14. Menegakis A, De Colle C, Yaromina A, Hennenlotter J, Stenzl A, Scharpf 34. Peterse EFP, Niessen B, Addie RD, de Jong Y, Cleven AHG, Kruisselbrink M, et al. Residual gamma H2AX foci after ex vivo irradiation of patient AB, et al. Targeting glutaminolysis in chondrosarcoma in context of the samples with known tumour‑type specific differences in radio ‑respon‑ IDH1/2 mutation. Br J Cancer. 2018;118:1074. siveness. Radiother Oncol. 2015;116(3):480–5. 35. Peterse EF, van den Akker B, Niessen B, Oosting J, Suijker J, de Jong Y, et al. 15. Menegakis A, von Neubeck C, Yaromina A, Thames H, Hering S, Hennen‑ NAD synthesis pathway interference is a viable therapeutic strategy for lotter J, et al. gammaH2AX assay in ex vivo irradiated tumour specimens: chondrosarcoma. Mol Cancer Res. 2017;15:1714–21. a novel method to determine tumour radiation sensitivity in patient‑ 36. Chan SM, Thomas D, Corces‑Zimmerman MR, Xavy S, Rastogi S, Hong WJ, derived material. Radiother Oncol. 2015;116(3):473–9. et al. Isocitrate dehydrogenase 1 and 2 mutations induce BCL‑2 depend‑ 16. Scully SP, Berend KR, Toth A, Qi WN, Qi Z, Block JA. Marshall Urist Award. ence in acute myeloid leukemia. Nat Med. 2015;21(2):178–84. Interstitial collagenase gene expression correlates with in vitro invasion in 37. Grassian AR, Parker SJ, Davidson SM, Divakaruni AS, Green CR, Zhang human chondrosarcoma. Clin Orthop Relat Res. 2000;376:291–303. X, et al. IDH1 mutations alter citric acid cycle metabolism and increase 17. Gil‑Benso R, Lopez‑ Gines C, Lopez‑ Guerrero JA, Carda C, Callaghan RC, dependence on oxidative mitochondrial metabolism. Cancer Res. Navarro S, et al. Establishment and characterization of a continuous 2014;74(12):3317–31. human chondrosarcoma cell line, ch‑2879: comparative histologic and 38. Tateishi K, Wakimoto H, Iafrate AJ, Tanaka S, Loebel F, Lelic N, et al. genetic studies with its tumor of origin. Lab Invest. 2003;83(6):877–87. Extreme vulnerability of IDH1 mutant cancers to NAD+ depletion. Can‑ 18. Suijker J, Oosting J, Koornneef A, Struys EA, Salomons GS, Schaap FG, cer Cell. 2015;28(6):773–84. et al. Inhibition of mutant IDH1 decreases D‑2‑HG levels without affect ‑ 39. Bernhard EJ, Stanbridge EJ, Gupta S, Gupta AK, Soto D, Bakanauskas VJ, ing tumorigenic properties of chondrosarcoma cell lines. Oncotarget. et al. Direct evidence for the contribution of activated N‑ras and K ‑ras 2015;6(14):12505–19. oncogenes to increased intrinsic radiation resistance in human tumor cell 19. Corver WE, Demmers J, Oosting J, Sahraeian S, Boot A, Ruano D, et al. lines. Cancer Res. 2000;60(23):6597–600. ROS‑induced near ‑homozygous genomes in thyroid cancer. Endocr Relat 40. Schrage YM, Lam S, Jochemsen AG, Cleton‑ Jansen AM, Taminiau AH, Cancer. 2018;25(1):83–97. Hogendoorn PC, et al. Central chondrosarcoma progression is associated 20. Lapytsko A, Kollarovic G, Ivanova L, Studencka M, Schaber J. FoCo: a sim‑ with pRb pathway alterations: CDK4 down‑regulation and p16 overex ‑ ple and robust quantification algorithm of nuclear foci. BMC Bioinform. pression inhibit cell growth in vitro. J Cell Mol Med. 2009;13(9A):2843–52. 2015;16:392. 41. Asp J, Brantsing C, Benassi MS, Inerot S, Sangiorgi L, Picci P, et al. Changes 21. Naipal KA, Verkaik NS, Ameziane N, van Deurzen CH, Ter Brugge P, Meijers in p14(ARF) do not play a primary role in human chondrosarcoma tissues. M, et al. Functional ex vivo assay to select homologous recombination‑ Int J Cancer. 2001;93(5):703–5. deficient breast tumors for PARP inhibitor treatment. Clin Cancer Res. 42. Asp J, Sangiorgi L, Inerot SE, Lindahl A, Molendini L, Benassi MS, et al. 2014;20(18):4816–26. Changes of the p16 gene but not the p53 gene in human chondrosar‑ 22. Baranski Z, Booij TH, Cleton‑ Jansen AM, Price LS, van de Water B, Bovee coma tissues. Int J Cancer. 2000;85(6):782–6. JV, et al. Aven‑mediated checkpoint kinase control regulates proliferation 43. van Beerendonk HM, Rozeman LB, Taminiau AH, Sciot R, Bovee JV, and resistance to chemotherapy in conventional osteosarcoma. J Pathol. Cleton‑ Jansen AM, et al. Molecular analysis of the INK4A/INK4A‑ARF 2015;236(3):348–59. gene locus in conventional (central) chondrosarcomas and enchondro‑ 23. van Oosterwijk JG, Meijer D, van Ruler MA, van den Akker BE, Oost‑ mas: indication of an important gene for tumour progression. J Pathol. ing J, Krenacs T, et al. Screening for potential targets for therapy in 2004;202(3):359–66. mesenchymal, clear cell, and dedifferentiated chondrosarcoma reveals 44. Bosco EE, Wang Y, Xu H, Zilfou JT, Knudsen KE, Aronow BJ, et al. The Bcl‑2 family members and TGFbeta as potential targets. Am J Pathol. retinoblastoma tumor suppressor modifies the therapeutic response of 2013;182(4):1347–56. breast cancer. J Clin Invest. 2007;117(1):218–28. 24. Sibinga Mulder BG, Mieog JS, Handgraaf HJ, Farina Sarasqueta A, Vasen 45. Ertel A, Dean JL, Rui H, Liu C, Witkiewicz AK, Knudsen KE, et al. RB‑ HF, Potjer TP, et al. Targeted next‑ generation sequencing of FNA‑ derived pathway disruption in breast cancer: differential association with disease DNA in pancreatic cancer. J Clin Pathol. 2017;70(2):174–8. de Jong et al. Clin Sarcoma Res (2019) 9:9 Page 11 of 11 subtypes, disease‑specific prognosis and therapeutic response. Cell Cycle. 50. Nicolay BN, Gameiro PA, Tschop K, Korenjak M, Heilmann AM, Asara 2010;9(20):4153–63. JM, et al. Loss of RBF1 changes glutamine catabolism. Genes Dev. 46. Pollack A, Wu CS, Czerniak B, Zagars GK, Benedict WF, McDonnell TJ. 2013;27(2):182–96. Abnormal bcl‑2 and pRb expression are independent correlates of 51. Reynolds MR, Lane AN, Robertson B, Kemp S, Liu Y, Hill BG, et al. Control radiation response in muscle‑invasive bladder cancer. Clin Cancer Res. of glutamine metabolism by the tumor suppressor Rb. Oncogene. 1997;3(10):1823–9. 2014;33(5):556–66. 47. Sharma A, Comstock CE, Knudsen ES, Cao KH, Hess‑ Wilson JK, Morey 52. Frisch S, Timmermann B. The evolving role of proton beam therapy for LM, et al. Retinoblastoma tumor suppressor status is a critical deter‑ sarcomas. Clin Oncol (R Coll Radiol). 2017;29(8):500–6. minant of therapeutic response in prostate cancer cells. Cancer Res. 2007;67(13):6192–203. Publisher’s Note 48. Dyson NJ. RB1: a prototype tumor suppressor and an enigma. Genes Dev. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 2016;30(13):1492–502. lished maps and institutional affiliations. 49. Clem B. RB in glutamine metabolism. Oncoscience. 2014;1(5):304–5. Ready to submit your research ? Choose BMC and benefit from: fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Clinical Sarcoma Research – Springer Journals

Published: May 28, 2019

There are no references for this article.