Exploitation of the Ewing sarcoma chemoresistance and stemness transcriptome to identify therapeutic target candidates and drug repurposing candidates. (2023)

Author(s): Elizabeth Ann Roundhill (corresponding author) [1,*]; Pantziarka [2]; Danielle E Liddle [1]; Lucy A Shaw [1]; Ghadir Albadrani [1]; Susan Ann Burchill (corresponding author)

1 Introduction

Ewing's sarcoma (SE) arises in bone or soft tissue [1] and is most common in young people between the ages of 10 and 25 [2]. Standard treatment, which includes a combination of chemotherapy, surgery, and multi-agent radiation therapy, has resulted in improved outcomes in some patients [1]. However, about 40% of these patients develop multidrug-resistant (MDR) metastasis [2,3,4,5], resulting in relapses and survival rates normally associated with metastasis (5-day survival rate).Years 10% [5 , 6, 7]). Late relapses and chemotherapy-induced morbidity are additional ongoing burdens for patients, their families, and caregivers. Therefore, there is an important unmet clinical need for the introduction of molecularly targeted therapy to minimize treatment-related toxicity and improve outcomes.

Progression and relapse are driven by subpopulations of cells capable of self-renewal and migration within tumors that are resistant to current treatments. These cancer stem cells (CSCs) have been identified in a variety of pediatric and adult solid tumors based on the expression of cell surface markers, most commonly CD133 (also known as Prominim-1) [8,9,10,11,12]. . Ewing's sarcoma cancer stem cells (ES-CSCs) have been isolated based on the expression of CD133 [13,14,15]. However, distinct CD133-negative CSCs are present in some cancers [8, 9, 10, 11, 12], including Ewing's sarcoma [15]. Therefore, additional approaches, including single-cell cloning and three-dimensional (3D) spheroid generation, have been used to improve the identification of ES-CSC [16,17] and CSC in other cancers [18,19,20].

In this study, we examined the integrity of CD133 as a cell surface marker of ES-CSCs and identified ABCG1 as an additional marker of these cells using 3D spheroids and single-cell self-renewal. The expression and prognostic potential of this ABC transporter protein were assessed in patient-derived cells and tumors in the online dataset GSE17618. Genes regulating pluripotency, stemness, and MDR-ABC transporter proteins were identified by comparing the transcriptomes of substrate-adherent two-dimensional (2D) ES-CSCs and 3D spheroid-derived ES-CSCs. We combined genes differentially expressed in 3D spheroid-cultured ES cells and in 2D-cultured cells with previously reported genes in patient-derived ES-CSCs [17] to identify candidate molecular targets derived from ES-CSCs be expressed. We then developed a tailored pipeline to identify small molecule inhibitors of these targets for further preclinical studies. If these drugs are effective in preclinical models of SE, they can be accelerated to clinical trials to be evaluated in patients in the future. This reuse strategy aims to reuse already approved drugs to treat new indications and is a complementary strategy to de novo drug development [21,22].

2. Materials and Methods

2.1. cell culture

ES cell lines (A673, RD-ES, SKES-1, SK-N-MC, TC-32 and TTC-466) were grown as previously described [23] and purchased from the American Type Culture Collection, Manassas, VA. except for the following cells which were kind gifts: TC-32 cells from Dr. J. Toretsky (Division of Pediatrics, University of Maryland, Baltimore, MD, USA), TTC-466 cells from Dr. P. Sorenson (British Columbia Children's Hospital, Vancouver, BC, Canada). Primary ES cell cultures and daughter ES-CSCs were grown as previously described [17]. SHEF-4 Embryonic Stem Cell (ESC) Culture (RRID:CVCL_9791) was a gift from Professor P. Andrews, Center for Stem Cell Biology, University of Sheffield, Sheffield, UK [24] and was used as a positive control for CD133 and member 1 of the ATP-binding cassette subfamily G (ABCG1). The glioblastoma cell line (T98G) was a gift from Professor M. Knowles, University of Leeds and was a positive control for Multidrug Resistance-Associated Protein 1 (MRP1, gene name = ABCC1) [23]. Human embryonic fibroblast line KMST-6 (grown in MEM with 10% FBS and 2mM glutamine) and Jurkat cells (grown in RPMI 1640 with 10% FBS and 2mM glutamine) were gifts from Dr. E. Morrison, University of Leeds, and used as a positive control or to construct the calibration curve for the measurement of telomere length.

2.2. western transmission

Western blotting (WB) was performed as previously described [23]. Equal protein loading was confirmed using α-tubulin [23] or β-actin (0.4 µg/ml, A5441, Sigma-Aldrich, Paisley, UK). For CD133 detection, protein extracts were heated to 95°C for 5 min before being chilled on ice and WB (CD133, 1 µg/ml, W6B3C1, Miltenyi Biotech, Surrey, UK). After incubation with primary antibodies (MRP1 [23] or ABCG1 (1:1000, ab36969; Abcam Plc., Cambridge, UK)) and secondary antibodies [23], the proteins were visualized by imaging (at different exposure times) depending on exposure to intensity ). signal) and quantified using the Li-cor Odyssey infrared imaging system (Li-cor Biosciences, Lincoln, NE, USA).

23. Flow cytometry

2.3.1. Cell surface expression of CD133

Cells (5 × 10[sup.5]) were cultured in FcR blocking buffer (Miltenyi Biotech) and anti-CD133/2 (4.5 µg/ml, clone 293C3, Miltenyi Biotech) or the IgG2b isotype control -PE (4, 5 µg/mL, Miltenyi Biotech) antibodies in the dark at 4°C for 10 min. Cells (10,000 per sample) were then analyzed by flow cytometry using FACSCalibur (BD Biosciences, Berkshire, UK).

2.3.2. Co-expression of ABCG1 and CD133

SK-N-MC cells were incubated in normal goat serum (1:10, Dako, Agilent Technologies, Santa Clara, CA, USA) in BD Perm/Wash™ Buffer for 30 min at 4°C. The cells were then treated with CD133 (anti-CD133/2, 4.5 µg/ml, clone 293C3) and/or rabbit polyclonal antibody ABCG1 (100 µg/ml, PA5-13462, Thermo-Fisher Scientific, Paisley, UK) in BD Perm/Wash™ Buffer for 1 h at 4 °C in the dark. Control cells were incubated with IgG2b-PE (4.5 µg/ml, Miltenyi Biotech) or rabbit IgG isotype control antibody (100 µg/ml, Dako). The cells were then incubated with the secondary antibody (2 µg/ml goat anti-rabbit IgG FITC, A31556, Thermo-Fisher Scientific) in BD Perm/Wash™ buffer for 30 min at 4°C in the dark. Cells (10,000 per sample) were analyzed by flow cytometry using CytoFLEX (Beckman Coulter, High Wycombe, UK).

2.4. self-renewal capacity

The growth of single cell progeny as an adherent culture was determined as previously described [17, 18]. A single cell (Poisson distribution probability of λ<1=0.9) was seeded into each well of 10 96-well primary plates (Corning) and the number of wells containing = 5 cells was recorded after 21 days light. Microscopy (Olympus CKX41). Whenever possible, self-renewing cell populations were expanded from single cells to establish daughter cell cultures; these are hereinafter referred to as ES-CSC. To study the efficiency of colony formation on soft agar, a suspension of single cells (1.8 × 105 cells) in cell-specific medium containing 0.3% agar (Aldrich, Poole, UK) was plated onto a solid agar bed (0.5% agar in Media). After 21 days, colonies were stained with 8 mM iodonitrotetrazolium chloride (prepared in ddH2O (w/v); Sigma-Aldrich) for 16 h and the number of colonies counted by light microscopy. Colony forming efficiency = [number of colonies formed in field of view/number of cells seeded] × 100.

For spheroid formation, a single cell was seeded into each well of an ultra-low fixation plate (Corning, UK) in cell line specific medium. The number of spheroids after 21 days was recorded to calculate the spheroid formation efficiency (SFE); SFE=[number of wells containing a spheroid/number of wells populated with a single cell]×100. The spheroids were imaged by light microscopy (Olympus CKX41). Spheroids formed from single cells were collected after 21 days for quantitative reverse transcriptase-polymerase chain reaction (RTqPCR), Western blotting, immunocytochemistry (ICC), and flow cytometry.

2.5. Sorting of magnetically activated cells to enrich for CD133 positive cells

ES cells (1 x 10 [sup.8]) were incubated with 300 µl CD133 microbeads and 100 µl FcR blocking buffer (Miltenyi Biotech). CD133 positive cells were isolated using LS columns (Miltenyi Biotech). CD133 negative cells were cleared of labeled cells (CD133 positive) by passing the cells through two LD columns (Miltenyi Biotech). CD133 expression on the cell surface was confirmed by flow cytometry immediately after separation: >90% of the CD133-positive selection expressed CD133, and CD133 was expressed on <5% of the CD133-negative selection cells. CD133-positive A673 and TC-32 cells remained positive for 15 passages.

2.6. RNA expression of pluripotency and differentiation markers, the Wnt signaling pathway, and ABC transporter proteins

RNA was extracted with the RNeasy Micro Kit (Qiagen, Düsseldorf, Germany) and RNA quantity and quality were measured with the Nanodrop Bioanalyzer and Agilent 2100. RNA (with a RIN > 8) was converted into cDNA by reverse transcription [2, 3] . mRNA expression was measured using the TaqMan[sup.®] Human Stem Cell Pluripotency Array, the TaqMan[sup.®] Human Wnt Pathway, and the TaqMan[sup.®] Human ABC Transporter Array (Applied Biosystems, Waltham, MA, United States) rated. United States) [23]. To allow a direct comparison between the 3 array platforms, the target Ct values ​​were normalized using the overall mean [25,26,27]. ABC transporter and pluripotency mRNAs (n = 140, excluding endogenous control mRNAs) expressed more than other mRNAs (Ct values ​​< 25), expressed (Ct values ​​25-35) and unexpressed (Ct values ​​> 35 , [18,28, 29]) in 3D spheroids from TTC 466 and SK-N-MC cells. Unique and common mRNAs in each cluster were analyzed using the Interacting Genes/Proteins Database (STRING) retrieval search tool (http://string-db.org, [17,30]) to identify the GO terms. Significant differences in mRNA expression were determined using linear models for microarray data (LIMMA) [18]. Target mRNAs were validated by individual RTqPCR assays if the Q value was < 0.1, there was at least a mean Ct difference of > 2 between the compared populations, and the Ct values ​​were < 35 [18]. For mRNA validation, total RNA (20 ng) was reverse transcribed and the PCR mix, the sequence-specific forward and reverse primer, and a probe for PPIA (the housekeeping gene) or ABCG1 (Thermo Fisher Scientific; ABCG1 Hs00245154_m1) and 1× cDNA added to TaqMan[sup.®] Universal PCR Master Mix (Thermo Fisher Scientific). RNA expression was calculated using the comparative Ct method [31].

ABCG1 transcripts were characterized in total RNA extracted from SK-N-MC cells grown as 2D cultures or 3D spheroids, sequenced and aligned as previously described [17]. Normalized read counts were identified in total RNA sequencing data using DESeq2 [32], adjusted p-values ​​of <0.01 were considered significant.

2.7. Immunohistochemistry (IHC)

Sections of formalin-fixed paraffin-embedded (FFPE) spheroids (4 µm) were deparaffinized in xylene (2 min) and resuspended in citric acid buffer (10 mM in ddH[sub.2]O, pH 6, [33]) prior to antigen recovery. . Endogenous peroxidases were blocked with 3% hydrogen peroxide for 5 min, and endogenous biotin, biotin receptors, and avidin binding sites were blocked with an avidin/biotin blocking kit (Invitrogen, Waltham, MA, EE, USA). Sections were incubated for 1 h with primary antibody ABCG1 (10 µg/mL, PA5-13462, Thermo-Fisher Scientific) or rabbit IgG control (10 µg/mL, Dako) at room temperature, followed by incubation with the secondary antibody. Sections were then incubated with streptavidin peroxidase (Abcam Plc.) followed by DAB substrate (Dako) for 10 min and nuclei counterstained with hematoxylin.

2.8. siRNA-Knockdown von ABCG1

Cells (5 x 10 [sup.4]) were seeded and allowed to attach overnight. Media were delivered with Accell siRNA Delivery Media (B-005000-500, Dharmacon, Lafayette, CO, USA) alone or with ABCG1 siRNA (1.5 µM, E-008615-00-0005, SMARTpool: Accell ABCG1 siRNA containing 4 exon 6) replaced targeting of siRNAs present in all canonical and novel transcripts (Dharmacon) or untargeted control siRNA (1.5 µM, D-001910-10-20, Dharmacon) for 72 hr. ABCG1 was confirmed by RTqPCR.

2.9. Apoptosis

Cells (5 × 10[sup.4]) were harvested after treatment with ABCG1 and control siRNA for 24–72 h and apoptosis was measured by labeling the cells with Annexin V and propidium iodide (PI) (Kit for detecting apoptosis by Annexin V-FITC, BD Biosciences, [34]).

2.10. Statistical analysis

Differences in proliferation and number of viable cells were log-transformed and analyzed by linear regression. Differences in the gradients of each plot were compared using the additional sum-of-squares F-test. For all other experiments, data were analyzed by analysis of variance (ANOVA followed by a Bonferroni or Tukey post hoc test) or a Mann-Whitney nonparametric or unpaired two-tailed t-test. Correlations were determined using Pearson's correlation coefficient (r). Values ​​of p<0.05 were considered significantly different. Statistical analyzes were performed with GraphPad PRISM 7.03 (GraphPad Software, San Diego, CA, USA).

2.11. Identification of drug candidates for priority molecular targets

Gene lists were analyzed using the STRING database. To identify drugs reported to target these molecular candidates, we surveyed the DrugBank (version 5.1.9 [35]) and the DGI database (DGIdb, version v4.2.0 [36]) to Access information about FDA-approved drugs and their molecular targets. Additional information on the drug target was derived from the literature-based database Reuse of Drugs in Oncology (ReDO) [37]. A polypharmacological approach identified targets for each drug candidate. Targets previously associated with ES via the Open Targets Platform [38] and DisGeNET [39] were used as the source of identification. Data from DGIdb and Open Targets characterize the strength of target-disease and drug-target associations. Using DGIdb, the interaction score is calculated as: (number of publications + number of data sources) × (mean known genetic partners for all drugs/known genetic partners for drug candidates) × (mean known pharmaceutical partners for all known genes/pharmaceutical partners for the target gene). For example, the Dir DGI score includes data on the number of molecular drug targets, the number of drugs targeting a gene, and the number of publications and data sources supporting the association. The Dir Assoc score is generated from the Open Targets database and is a measure of the relationship between the molecular target and the disease term cancer.

Approved anticancer drugs were identified from the list of drug candidates that address >1 ES molecule target using the anticancer drug database [40]. ES trials were identified using clinicaltrials.gov, the EU clinical trials registry, and the WHO International Clinical Trials Registry Platform. For non-oncology drugs previously identified as oncology buyback candidates (ReDO database), clinical trial activity in all cancer types was extracted using the ReDO_Trials database for active oncology buyback trials [41]. For these repurpose candidates, a support score was calculated from the ReDO database data to characterize the range of available data illustrating the anticancer effects of drugs (e.g., in vitro, in vivo, case reports, observational data, and data from clinical trials). studies).

3. Results

3.1. ES cell lines produce spheroids and clones from a single cell

All ES cell lines (6/6) formed clones or spheroids from a single cell when grown on soft agar or ultra-low binding plates (Table 1, Figure 1), consistent with previous reports that cultures of ES -Cells have the ability to self-renew. cell population [8,9,10,11,12,13,14,17].

The formation of clones and spheroids derived from single cells in soft agar (29 ± 2%, p < 0.05) and extremely low adhesion conditions (75 ± 7%, p < 0.05; Table 1, Figure 1) was more efficient in the SKES-1 cell line. Spheroid formation was 100% in all cell lines when 100 cells or more were pooled (Table 1). Larger spheroids (>400 µm) produced a cell-line specific morphology (Figure 1). SKES-1 spheroids produced a relatively uniform sphere of disseminated or dissociated cells, consistent with spheroid formation in a variety of solid tumor cell types [42,43,44,45,46]. For the first time, we identified 3D projections evolving from the central nucleus in five of the six ES cell lines (Figure 1). The biological significance of these projections requires further investigation.

3.2. CD133 identifies some drug-resistant ES-CSCs that are self-renewing

CD133 protein was detected in all ES cell lines except SK-N-MC cells (Figure 2A and Supplementary Data S1). CD133-positive cells A673 and TC-32 formed significantly more colonies on soft agar than CD133-negative SK-N-MC cells (Supplementary Data S1-S3). However, there were no differences in proliferation, cell cycle status, or telomere length in the two cell lines (Supplementary Data S1-S3), phenotypes commonly associated with CSC, and self-renewal capacity [47]. In addition, the expression and activity of the multidrug-resistant protein MRP1 was increased in the CD133-positive TC-32 cells but not in the CD133-positive A673 population (Supplementary Data S1-S3). The common CSC phenotype of the CD133-positive and CD133-negative ES populations indicates that CD133 expression alone is not sufficient to enrich the entire ES-CSC population.

Consistent with the cellular origin of the ES and the high level of stem markers expressed by these tumor cells [48,49,50], there are no significant differentially expressed genes involving pluripotency, ABC transporters and the Wnt signaling pathways in CD133-positive and CD133. -negative cells (Figure 2B (TC-32) and Figure 2C (A673), Supplementary Data S4). These observations are also consistent with the premise that CD133 can be used to identify some cells with ES-CSC features, although signaling pathways classically associated with the CSC phenotype in other cancers are also found in CD133-negative ES- cells are expressed. To examine the phenotype of CD133-independent ES-CSCs, we examined ES-CSCs enriched by spheroid formation in ES cells with no or low expression of CD133.

3.3. Gene expression profile of 3D spheroids and 2D cultures CD133 low or negative ES

Because TTC 466 and SK-N-MC cells had low or undetectable CD133 protein (Figure 2A), we compared the transcriptome of TTC 466 and SK-N-MC 3D spheroids and 2D cultures (Figure 3A ). Consistent with the premise that spheroid formation is a feature of CSCs, expression of the stem markers LEFTB and LIN28 was significantly increased in SK-N-MC spheroids compared to cells in 2D culture, while expression of LAMA1, COL2A1, ACTC , GCG, SEMA3A and PODXL were significantly reduced (Q-value < 0.1, Figure 3B,D, Supplementary Data S5A). In TTC 466 spheroids, no marker was significantly increased; However, expression of FLT1 and GATA6 was significantly reduced compared to cells in 2D culture (Figure 3C, E, Supplementary Data S5A).

The expression of 20 ABC transporter genes was significantly different in SK-N-MC spheroids compared to cells in 2D cultures, although only ABCG1 and CFTR were more than twice as highly differentially expressed (Figure 4A-C). However, ABC transporter genes, including ABCG1, were not significantly differentially expressed in TTC 466 cells grown in 2D or as 3D spheroids (Figure 4D-F). A comparison of mRNAs in SK-N-MC and TTC-466 spheroids revealed that 35% of the genes were undetected in both populations and an additional seven (5%) and six (4%) unique mRNAs in TTC. 466 and SK -N-MC spheroids were not detected (Figure 4G). Fifty-six percent of the mRNAs were detected in both populations of spheroids, and 8% of these were highly expressed (Figure 4G). All six mRNAs, which are only expressed in TTC 466 spheroids, are involved in cell differentiation (GO:0030154), which could explain why increased pluripotency mRNA expression was not observed in TTC 466 spheroids ES cells [51]. It remains to be seen whether the lack of common pluripotency mRNA was increased in TTC 466 and SK-N-MC spheroids and the different transcriptional activators of TTC 466 (EWSR1-ERG) and SK-N-MC (EWSR1-FLI1) -Reflecting cells.

The decrease in CFTR gene expression was not validated by RTqPCR or Western blot (results not shown). However, the increase in ABCG1 expression was confirmed in SK-N-MC spheroids at the mRNA (p=0.0015, Figure 4H) and protein (Figure 4I) level. ABCG1 is heterogeneously expressed in all ES cell lines (6/6; Supplementary Data S5B). ABCG1 expression was confirmed by IHC in the outer proliferating region of 3D SK-N-MC spheroids but not in the hypoxic region or inner necrotic core (Figure 4J; [34]), suggesting that the Expression of ABCG1 can be regulated by hypoxia and under these conditions have a functional role in ES cells.

3.4. Functional role and characterization of ABCG1 in ES-CSC

Knockdown of ABCG1 with siRNA had no effect on the viability or apoptosis of SK-N-MC cells in culture (10 ± 0.8% in untargeted cells and 10 ± 1.1% in ABCG1-siRNA treated cells, p > 0.05, Figure 5A). There was also no effect on proliferation (0.93 ± 0.1 and 0.94 ± 0.05 in untargeted and ABCG1-siRNA-treated cells, respectively, p > 0.05) or self-renewal capacity of a single cell on agar, bland (60% in untargeted and ABCG1 siRNA -treated cells). 58% in cells treated with siRNA ABCG1). However, after ABCG1 knockdown, spheroid production was significantly reduced; 65 ± 3% in non-target cells and 33 ± 5% in ABCG1 siRNA treated cells (p<0.0001, n=10). These data suggest that while ABCG1 plays no role in 2D ES cell homeostasis, it can affect cell-cell interactions and components of the tumor microenvironment important for the development of 3D spheroids and possibly tumors. This hypothesis needs further investigation.

Analysis of ABCG1 RNA in SK-N-MC spheroids and 2D cultures by total RNA sequencing revealed the expression of 11 previously reported transcripts (www.ensembl.org, accessed April 27, 2016; Figure 5B,C ) and two new ABCG1 RNA species (Figure 5C). Canonical ABCG1 generates the multiple transcripts through alternative splicing (Figure 5C). The new transcript 1 is more similar to transcript 4 (ENST00000450121.5), both sequences lacking exon 5 (Figure 5C). However, the new transcript 1 encompassed exons 8–15, and an extended 3'UTR region was predicted to produce a higher molecular weight protein than transcript 4. The RNA sequence of the new transcript 2 is more similar to that of the new transcript ( ENST00000361802.6) the additional 3'UTR sequence is unlikely to produce a unique protein product. The expression of both novel transcripts at the RNA level was increased over the canonical protein-producing sequences (transcripts 1, 3-8) in SK-N-MC cells grown as spheroids compared to 2D cultures, this was more significant for the new transcript 1 (Figure 5D). The increased expression of the ABCG1 gene in SK-N-MC spheroid cells (Figure 4G, H) could reflect the expression of the novel transcript 1. Further studies are needed to elucidate this novel ABCG1 transcript and its role in ES-CSC investigate.

Since ABCG1 mRNA was increased in ES-CSC of SK-N-MC spheroids, we examined its co-expression with CD133. There was no significant difference in the percentage of cells expressing ABCG1 or CD133 in SK-N-MC cells from 3D spheroids or 2D cultures (Figure 6A,B). Consistent with the increase in ABCG1 in protein extracts from 3D spheroids (Figure 4H), the ABCG1 protein level per cell was higher in cells from 3D spheroids than in 2D cultures (2.5 ± 0.8-fold, p < 0 .05, Figure 6C). CD133 levels per cell were also increased (6.3±1-fold increase, p<0.05, Figure 6C), consistent with the higher expression of ABCG1 protein in TC-32 positive cells compared to CD133- matches cells negative for CD133 (Figure 6D). . ABCG1 RNA expression was confirmed at the protein level in patient-derived primary ES cells (Figure 6E). However, there was no correlation between the percentage of CD133-positive cell lines or patient-derived cells and the ability to produce offspring as measured by the soft agar colony formation assay (R[sup.2]<0.1). Renewal capacity of a single cell plated on poorly adherent or adherent plates (R[sup.2]=0.002; Fig. 6F), suggesting that CD133 in patient-derived cells may not identify cell populations with a self-renewing phenotype.

Querying the publicly available GSE17618 RNA dataset revealed that high expression of CD133, but not ABCG1, was associated with a greater than three-fold risk of event and poor outcome (Supplementary Data S6), consistent with the hypothesis, that high expression of CD133 ES identifies -CSCs that may be responsible for progression and relapse in some patients (Figure 6G).

3.5. Identification of drug candidates for ES-CSC molecular targets

Small molecule inhibitors on 62% (13/21) of the unique molecular targets were expressed in 3D spheroids from TTC 466 cells and SK-N-MC and ES-CSC (Figure 7A, B) using our custom pipeline (Figure 7A, B) 7C). Of the 279 anti-targeting drugs, 66 (24%) are approved anticancer drugs and 213 (76%) are noncancer drugs (Figure 7C and Supplementary Data S7). Most drugs (202/279; 72%) target one of the two ATP-binding cassette (ABC) drug precursor proteins, p-glycoprotein (n=168, including n=10, targeting p-glycoprotein and additional proteins​ ​target) or MRP1 (n=13), while 21 drugs targeted both p-glycoprotein and MRP1. Forty-eight of the 279 (17%) drugs that met targets were evaluated in oncology studies, while the majority of drugs were evaluated in non-oncology drug studies (n = 986; Figure 7C, Table 2, Supplementary Data S8).

POU5F1/OCT-4 interacted with 10/13 of the identified strain and chemoresistance genes (Figure 7D), suggesting common functional protein-protein association networks. Five of these 10 gene products (POU5F1/OCT4, c-KIT, CAV1, ITGB1, and CD44) interact (Figure 7D) and regulate vital cellular processes (Supplementary Data S9). All of these proteins are highly expressed in ES-CSCs and therefore may represent potential prognostic biomarkers and/or therapeutic targets in ES. We are currently examining these possibilities. Using our custom pipeline, we identified two FDA-approved small molecule inhibitors, allopurinol and phenytoin, that target POU5F1/OCT4 and are used to treat gout and control seizures in epilepsy, respectively (Table 2, https://www.dgidb .org/genes/POU5F1#_interactions, accessed 3 January 2023). These two approved drugs have the highest DGIdb interaction scores of the three compounds, which interact with POU5F1/OCT4, inhibit a variety of other cancer-related molecular targets, and are also currently in trials as anti-cancer therapies, making them attractive drug candidates for additional drugs in vitro -Studies. Analysis for ES. Several inhibitors targeting multiple receptor tyrosine kinases (TKIs) involved in sarcoma pathogenesis have also been identified, including SE [52,53], including regorafenib, which is being studied in combination with chemotherapy (NCT02085148/2013-003579- 36; NCT04055220 /REGOSTA;2021-005061-41/INTER-EWING-1;Table 2).

4. Discussion

We have identified for the first time the expression of the membrane-associated ABC transporter protein ABCG1 on the surface of human embryonic stem cells. ABCG1 levels were increased in SK-N-MC cells forming 3D spheroids compared to cells in 2D culture. ABCG1 expression was particularly high at the outer edge of the spheroids, suggesting a structural or transport function between cells inside and outside the spheroid. Consistent with this hypothesis, ABCG1 knockdown reduced the ability of ES cells to bind to each other and produce 3D spheroids, reminiscent of ABCG1-dependent regulation across cell and intracellular membranes [54]. ABCG1 protein was detected at lower levels in hypoxic cells surrounding the necrotic center of the spheroids, consistent with observations in mouse intestinal adenocarcinoma spheroids where ABCG1 expression and hypoxia-inducible factor 1a are inversely correlated [55 ]. In contrast to studies in lung cancer [56] and normal hematopoietic stem cells [57], ABCG1 downregulation had no direct impact on ES cell proliferation or apoptosis. Decreased ABCG1 is reported to promote apoptosis through increased expression of endoplasmic reticulum (ER) stress proteins GRP78 and CHOP [58]. The ability to tolerate high levels of ER stress conferred by the EWSR1-ETS oncogene [59] may explain why ABCG1 knockdown does not induce apoptosis in these ES cells. ABCG1 expression is higher in several cancers compared to normal tissues [29,56] and expression is associated with higher grade tumors [60], metastasis [61] and poor response to chemotherapy [62]. Consistent with these observations, ABCG1 expression is increased in drug-resistant and self-renewing osteosarcoma cells [18], a second bone cancer characterized by recurrent disease and acquired MDR. Further studies are needed to determine the effect of this lipid ABC transporter protein on plasma membrane organization and recruitment of signaling processes that regulate cell fate and may contribute to tumor development, evasion and metastasis. Single nucleotide polymorphisms (SNPs) of ABCG1 in the first cytoplasmic domain (within intron 2) are associated with survival of patients with non-small cell lung cancer [63]. Mutagenesis studies have highlighted the importance of this region for efficient transport of ABCG1 to the plasma membrane and regulation of cholesterol efflux [64,65]. The clinical relevance of the 17 SNPs identified within the cytoplasmic region of ABCG1 (amino acids 178-195, exon 5, www.nvbi.nlm.nih.gov/snp, accessed 13 July 2022). In ES 3D spheroids, we identified two new ABCG1 transcripts, the most abundant transcript (transcript 1) missing exon 5 from the cytoplasmic region. The clinical and cellular functional significance of this deletion and these transcripts warrants further investigation. These hypotheses and data are consistent with the premise that ABCG1 identifies ES cells capable of surviving chemotherapy and may be responsible for progression and relapse in some patients. We are currently investigating the expression and functional role of ABCG1 mRNAs and proteins in ES.

Consistent with the premise that CD133 can identify some cells with a CSC phenotype [14,66], the number of CD133-positive cells and the level of CD133 expression was higher in 3D spheroid ES cells than in ES cells that were grown in 2D. The CD133-positive cells shared common characteristics of cancer stem cells, including increased colony formation from a single cell. However, in patient-derived cells, there was no correlation between CD133 expression and the self-renewal capacity of a single cell. This underlines the difference between established ES cell lines and newer cultures from patient tumors and supports the premise that CD133 does not identify all ES cell populations with a self-renewing phenotype. We are currently investigating the difference in genotype and phenotype of ES cell lines and patient-derived cells. Although CD133 is reported to be a stem cell biomarker, this is controversial, which may be due to heterogeneity and uncertainty about its physiological function(s) [67]. In contrast to studies in metastatic melanoma [68], we found no enrichment of canonical Wnt signaling in CD133-positive ES cells compared to CD133-negative cells [69] and CD133 levels did not predict outcomes. The increased drug resistance of TC-32 CD133 positive cells compared to CD133 negative cells could instead be brought about by increased levels of the drug exit protein MRP1, whose overexpression induces resistance to chemotherapy [23] and high membrane expression worsens predicts clinical picture. Result [27]. This points to the need for further studies on the intrinsic and acquired mechanisms of resistance in SE.

There is increasing evidence that SEs can occur in neurally derived mesenchymal stem cells (MSCs; [70]) or in neural crest cells [71]. The interaction between the permissive cellular environment and the tumor-specific EWSR1-ETS chimeric transcription factors leading to cell transformation and SE is poorly understood, although it is certainly important, illustrated by the high incidence of SE in Europeans compared to Africans [72, 73, 74]. Transcriptome and epigenome wiring are also known contributing factors, along with rare oncogenic events such as STAG2, TP53, and CDKN2A [75]. Interestingly, five of the six cell lines produced 3D progeny with protrusions similar to those displayed by tissue organoids [76], which have not been previously reported in bone cancer. In contrast, SKES-1 cells generated a nucleus with surrounding dissociated cells resembling the ES spheroids of CD133-positive STA-ET 8.2 cells [15]. The heterogeneity of spheroid compactness, growth, and stability over time may reflect the amount of extracellular matrix components produced by the different cell lines [77], suggesting that these models are useful tools to assess the efficacy of therapeutic can be candidates. In this study, we found that the ES-CSC phenotype was associated with the expression of ABC transporter proteins and strain traits. However, due to the bystander stem pathways associated with the cell of origin, further investigation is needed to unravel the functional molecular mechanisms leading to Ewing sarcogenesis.

We identified POU5F1 as a promising therapeutic target candidate that interacts with 10 of the prioritized targets. The protein product of this gene (octamer-binding transcription factor 4, OCT4) regulates multiple functional properties of CSCs, including self-renewal, survival, drug resistance, epithelial-mesenchymal junction, and metastasis [78] . This is consistent with the emerging role of this protein in adult cancer tumorigenesis and the stimulation of expression by the EWSR1 proto-oncogene in various human cancers, including SE [79]. Using our custom pipeline, we identified two off-patent, FDA-approved drugs targeting POU5F1/OCT4, allopurinol and phenytoin. The efficacy and safety of phenytoin are currently being evaluated in a phase 2 clinical trial as part of combination maintenance therapy in patients with clinically advanced sarcoma including SE [80] and in patients with metastatic pancreatic cancer [81]. A trial combining allopurinol and mycophenolate mofetil with chemotherapy in patients with relapsed small cell lung cancer is expected to begin enrollment later this year (the CLAMP trial, NCT05049863). We are currently investigating the effect of allopurinol and phenytoin alone and in combination with other new drugs and standard chemotherapy [82] in ES.

Most of the drugs we have identified target the ABC transporter proteins p-glycoprotein and/or MRP1, which contribute to MDR by reducing drug levels in cells. Several ABC transporter proteins have been described in SE, including MRP1, p-glycoprotein, ABCG1, ABCF1, ABCA6, and ABCA7 [83]. P-glycoprotein has been studied most frequently, although it does not predict outcome [27,84,85,86]. To date, despite the development of third-generation P-glycoprotein inhibitors, these drugs have had limited or no clinical value, and most studies have been discontinued due to unacceptable toxicity. In contrast, high membrane MRP1 expression in tumors at the time of diagnosis predicts reduced event-free and overall survival for patients [27]. Interestingly, low ABCF1 levels combined with high IGF2BP3 levels also predict poor treatment outcome [87]. Despite the clinical failure of inhibitors of ABC transporter proteins, several targeting small molecules, including some TKIs, interact with one or more ABC transporters, suggesting that inhibitors may be beneficial in some cases [88]. A further understanding of the functional role of ABC transporter proteins, including ABCG1, in cells responsible for progression and relapse is needed to determine which therapeutic target candidates, if any, are worthy therapeutic targets to overcome MDR.

5. Conclusions

In conclusion, we have evidence that ABCG1 could be a candidate biomarker that could be used to select SS patients for treatment. This hypothesis requires validation. We have identified proteins expressed by ES-CSCs that could be therapeutic targets and are using our pipeline of tailored projects to identify drug candidates for reuse that have the potential to inhibit these targets and eradicate ES-CSCs. These drugs include FDA-approved small molecule ABC transporter proteins and two inhibitors of POU5F1/OCT-4.

author contributions

Conceptualization, S.A.B.; Methodology, E.A.R. And sat.; Software, E.A.R.; Validation, E.A.R., P.P., D.E.L., L.A.S., G.A. And sat.; formal analysis, E.A.R., P.P., D.E.L., L.A.S., G.A. And sat.; Investigation, E.A.R., D.E.L., L.A.S. and G.A.; Resources, P.P. And sat.; Data maintenance, E.A.R., P.P., D.E.L., L.A.S., G.A. And sat.; Writing - original draft preparation, E.A.R. And sat.; Writing – proofreading and editing, E.A.R., P.P. And sat.; Visualization, E.A.R. And sat.; Supervision, S.A.B.; Project Management, S.A.B.; Fund acquisition, S.A.B. All authors have read and accepted the published version of the manuscript.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki. The study was approved and monitored by the Integrated Research Application System (IRAS167880) and the CRN Yorkshire & Humber NIHR Clinical Research Network (EDGE 79301).

declaration of consent

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

FASTQ files of sequenced ES cell lines, 2D and 3D SK-N-MC cells are in the Leeds Research Data Repository (University of Leeds), Burchill Susan and Roundhill Elizabeth (2022): Total RNA Sequencing of Lines ES cells, TTC 466, available and 2D and 3D SK-N-MC cultures. [Record]. https://doi.org/10.5518/1210.

conflicts of interest

The authors declare no conflict of interest.

Disclaimer/Editor's Note: The statements, opinions and data contained in all publications are solely those of the individual authors and contributors and not of MDPI or the editors. MDPI and/or the publisher(s) disclaim any responsibility for any damage to persons or property resulting from any ideas, methods, instructions or products mentioned in the content.

expression of gratitude

The authors thank Kim Cass, Tom Maisey, Samantha Brownhill, and Andrea Berry for technical assistance.

additional materials

The following supporting information can be downloaded from https://www.mdpi.com/article/10.3390/cancers15030769/s1. Complementary Methods. Methods related to the data described in Supplemental Data S1 are described. Additional data S1. Results of experiments analyzing populations of TC-32 and A673 CD133 positive and CD133 negative. Supplementary Data S2 (A) Colony formation, (B) proliferation, (C) cell growth, (D) cell cycle, (E) telomere length analysis, and (F) migration in CD133-positive and CD133-negative A673 cells CD133 . (G) Protein expression of MRP1 in CD133-positive and -negative A673 cells. Equal protein loading was confirmed by α-tubulin expression. (H). MRP1 drug efflux activity was quantified by calcein-F efflux versus baseline fluorescence after calcein-AM loading. Results are shown as mean ± SEM of three independent experiments. Calcein F efflux was compared using a nonparametric Mann-Whitney two-tailed t-test. (I) The effect of doxorubicin (3.5-224 nm) on the number of viable CD133 positive and negative cells of A673 after 48 h was quantified by trypan blue exclusion assay. The number of viable cells is presented as a percentage of cells after doxorubicin treatment relative to vehicle control (DMSO); results are shown as mean ± SEM of three independent experiments. EC[sub.50] values ​​were calculated using linear regression and compared using the additional sum of squares F-test. Supplementary Data S3. (A) Colony formation, (B) proliferation, (C) cell growth, (D) cell cycle, (E) telomere length analysis, and (F) migration in CD133-positive and CD133-negative TC-32 cells. (G) Protein expression of MRP1 in TC-32-CD133 positive and negative cells. (H) The drug efflux activity of MRP1 was quantified by calcein F efflux. (I) The effect of doxorubicin (3.5-224 nm) on the number of TC-32-CD133 positive and negative viable cells in 48 h was quantified using trypan blue exclusion test. Supplementary data S4. Expression of genes associated with differentiation, pluripotency and stemness, ABC transporter protein mRNA and those associated with Wnt signaling in A673 and TC-32 CD133 positive and negative cells. Supplementary data S5. (A) Expression of genes at RNA level related to differentiation, pluripotency and stem cells in SK-N-MC and TTC-466 cells in 2D and 3D cultures. (B) ABCG1 protein expression in ES cell lines (n=6) was assessed by Western blotting. Equal protein loading was confirmed by examining the blots for GRP75. Supplementary data S6. Prognostic importance of CD133 and ABCG1 RNA in the diagnosis of ES tissue. Additional data S7. Complete list of drugs (n=279) targeting the common therapeutic targets of trunking and chemoresistance. Additional data S8. Drugs directed against therapeutic targets with common strain and chemoresistance and corresponding clinical trial identifiers. Additional data S9. Statistically significant GO terms (by false detection rate) and biological processes driven by POU5F1/OCT4, c-KIT, CAV1, ITGB1 and CD44. References [18,23,31,34,89,90,91] are cited in the Supplementary Materials.

references

1. New Jersey Balamuth; RB Womer Sarkom Ewing. 184-1 DOI: https://doi.org/10.1016/S1470-2045(09)70286-4. PMID: https://www.ncbi.nlm.nih.gov/pubmed/20152770.

2. SJ Cotterill; S. Ahrens; M. Paulussen; H. F. Jürgens; PENNSYLVANIA. Abstimmung; H. Gadner; A. W. Artisanal Prognostic Factors in Ewing Tumor of Bone: Analysis of 975 Patients from the European Intergroup Cooperative Ewing Sarcoma Study Group, 2000, 18, S. 3108-3114. DOI: https://doi.org/10.1200/JCO.2000.18.17.3108. PMID: https://www.ncbi.nlm.nih.gov/pubmed/10963639.

3. S. Ahrens; C. Hoffman; S. Jabar; G. Braun-Münzinger; M Paulussen; J Dunst; C. Rubio; W. Winkelman; A. Heinecke; U. Goebel et al. Evaluation of prognostic factors in a tumor volume-adapted treatment strategy for localized Ewing's sarcoma of bone: The CESS experience 86. Ewing Sarcoma Cooperative Study., 1999, 32, pp. 186-195. DOI: https://doi.org/10.1002/(SICI)1096-911X(199903)32:3<186::AID-MPO5>3.0.CO;2-D.

4. C. Rodríguez-Galindo; F. Navid; T. Liu; KALIFORNIEN. Rechnungen; BN Rao; M.J. Krasin Prognostic Factors for Local and Distant Control in the Ewing Sarcoma Family of Tumors., 2007, 19, S. 814-820. DOI: https://doi.org/10.1093/annonc/mdm521. PMID: https://www.ncbi.nlm.nih.gov/pubmed/17998282.

5.M.Steel; A.Raft; M Paulussen; T Bolling; V.Vieth; S. Bielack; I. Gortitz; G. Braun-Münzinger; J Hardes; H. Juergens et al. Risk of recurrence and survival after recurrence in patients with Ewing's sarcoma., 2011, 57,pp. 549-553. DOI: https://doi.org/10.1002/pbc.23040.

6.LM Barker; T.W. Kraut von Pender; J.E. Sanders; survival of D.S. Hawkins on recurrence of the Ewing sarcoma family of tumors., 2005, 23, pp. 4354-4362. DOI: https://doi.org/10.1200/JCO.2005.05.105.

7.M.Huang; K. Lucas Current treatment approaches in metastatic and recurrent Ewing's sarcoma., 2011, 2011, p. 863210. DOI: https://doi.org/10.1155/2011/863210.

8. K.M. Joo; S.Y. kim; Xjin; S.Y. Song; DS Kong; JI Lee; JW Jeon; M.H. Kim; B.G. kang; Y Jung et al. Clinical and biological implications of CD133 positive and CD133 negative cells in glioblastoma., 2008, 88, pp. 808-815. DOI: https://doi.org/10.1038/labinvest.2008.57.

9. A. T. Ogden; AE Waziri; RA. Seekopf; D. Fusco; K. Lopez; JA Ellis; J. Kang; M. Assana; G. M. McKhann; MEGABYTE. Sisti et al. Identification of A2B5+CD133- tumor-initiating cells in adult human gliomas., 2008, 62, pp. 505-514. DOI: https://doi.org/10.1227/01.neu.0000316019.28421.95.

10. V. Tyrone; V. Desiderio; R. d'Aquino; F. by Francesco; G Pirozzi; A. Graziano; U. Galderisi; C. Cavaliers; A. DeRosa; G. Papaccio et al. Detection and characterization of CD133+ cancer stem cells in human solid tumors., 2008, 3, e3469. DOI: https://doi.org/10.1371/annotation/6475ab87-5c24-4ee6-9989-0762186dd073.

11. D. Walter; S. Satheesha; S. Albert; BEFORE CHRIST. Houses of birth; V. D'Alessandro; SM Oesch; H. Rehrauer; I. Leuschner; E. Koscielniak; C. Gengleret al. The CD133-positive embryonic rhabdomyosarcoma stem cell population is enriched in rhabdospheres., 2011, 6, e19506. DOI: https://doi.org/10.1371/journal.pone.0019506.

12.J.Wang; MAIL. Zechariah; O. Tsinkalovsky; H. Everfull; SO. Estate; A Svendsen; L. Presbytery; G.Rosland; F Thorsen; L.Stuhr et al. CD133-negative glioma cells form tumors in nude rats and give rise to CD133-positive cells., 2008, 122, pp. 761-768. DOI: https://doi.org/10.1002/ijc.23130.

13. S. Cornaz Offices; N.Riggi; C. De Vito; A Sarre; I. Letovanec; S. Provero; I. Stamenkovic Targeting of stem-like cancer cells as a method to combat cellular heterogeneity in Ewing's sarcoma., 2014, 74, pp. 6610-6622. DOI: https://doi.org/10.1158/0008-5472.CAN-14-1106.

14. ML Suva; N.Riggi; JC McCarthy awaken; K Baumer; S. Tertiary; JM Joseph; D.Suva; V. Clemens; S. Provero; L. Cironie et al. Identification of cancer stem cells in Ewing's sarcoma., 2009, 69, pp. 1776-1781. DOI: https://doi.org/10.1158/0008-5472.CAN-08-2242.

15th X Jiang; Y.Gwye; D Russell; C.Cao; D Douglas; L. hung up; H.Kovar; TJ trikes; ER Lawlor CD133 Expression in Chemoresistant Ewing Sarcoma Cells., 2010, 10, 116. DOI: https://doi.org/10.1186/1471-2407-10-116.

16. H. Fujii; K.Honoki; T. Tsujiuchi; A. Kido; K. Yoshitani; Y. Takakura Populations of sphere-forming stem-like cells with drug resistance in human sarcoma cell lines., 2009, 34, pp. 1381-1386.

17. E.A. round hill; M. Chicon-Bosch; L.Jeys; M. parry; KANSAS. Classification; A. fall; RNA sequencing by S.A. Cell function studies derived from Burchill and patients indicate that neurexin-1 and regulators of this pathway are associated with poor outcomes in Ewing's sarcoma., 2021, 44, pp. 1065-1085. DOI: https://doi.org/10.1007/s13402-021-00619-8.

18. E.A. round hill; S.Jabri; SA Burchill ABCG1 and Pgp identify drug-resistant, self-renewing osteosarcoma cells., 2019, 453, pp. 142-157. DOI: https://doi.org/10.1016/j.canlet.2019.03.011.

19. E. Charafe-Jaufret; C. Ginestier; F.Iovino; J.Wicinski; N.cervera; P.Finetti; M.H. Hur; ME Diebel; F.Monville; J. Dutcher et al. Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a unique molecular signature., 2009, 69, pp. 1302-1313. DOI: https://doi.org/10.1158/0008-5472.CAN-08-2741.

20. E. Suzuki; T.Chiba; Y.Zen; S.Miyagi; M.Tada; F.Kanai; F. Imazeki; M.Miyazaki; A.Iwama; O. Yokosuka Aldehyde dehydrogenase 1 is associated with recurrence-free survival but not with stem cell-like properties in hepatocellular carcinoma., 2012, 42, pp. 1100-1111. DOI: https://doi.org/10.1111/j.1872-034X.2012.01028.x.

21. F. Bertolini; Vice President Sujatmé; G. Bouche Drug reuse in oncology: Opportunities for patients and healthcare systems., 2015, 12, pp. 732-742. DOI: https://doi.org/10.1038/nrclinonc.2015.169. PMID: https://www.ncbi.nlm.nih.gov/pubmed/26483297.

22. P. Pantziarka; C. Verbanderd; I. Hughes; G. Bouche; L. Meheus Drug Reuse in Oncology: from Candidate Selection to Clinical Adoption., 2021, 68, S. 186-1 DOI: https://doi.org/10.1016/j.semcancer.2020.01.008.

23. E. A. runder Hügel; S.A. Burchill Detection and characterization of multidrug resistance protein 1 (MRP-1) in human mitochondria., 2012, 106, pp. 1224-1233. DOI: https://doi.org/10.1038/bjc.2012.40. PMID: https://www.ncbi.nlm.nih.gov/pubmed/22353810.

24. B. Aflatoonian; L. Ruban; S. Schamsuddin; D. Bäcker; P. Andrews; H. Moore Generation of Sheffield (Shef) Human Embryonic Stem Cell Lines Using a Droplet Culture System., 2010, 46, S. 236-241. DOI: https://doi.org/10.1007/s11626-010-9294-2. PMID: https://www.ncbi.nlm.nih.gov/pubmed/20224972.

25. M. Fischer; M.Skowron; F. Berthold Reliable quantification of transcripts by real-time reverse transcriptase polymerase chain reaction in primary neuroblastoma using normalization to mean expression levels of the control genes HPRT1 and SDHA., 2005, 7, pp. 89-96. DOI: https://doi.org/10.1016/S1525-1578(10)60013-X.

26.M.Lastowska; V. Viprey; M. Santibanets-Koref; I. Wappler; H Peters; C. cullinane; P Roberts; AG lounge; DA chatter; AD Pearson et al. Identification of candidate genes involved in neuroblastoma progression by combining expression and genomic microarrays with survival data., 2007, 26, pp. 7432-7444. DOI: https://doi.org/10.1038/sj.onc.1210552.

27. E Roundhill; S. Burchill Membrane MRP-1 expression but not MRP-1 splicing or Pgp expression predicts survival in patients with TFSE., 2013, 109, pp. 195-206. DOI: https://doi.org/10.1038/bjc.2013.168.

28. M. Kodani; G.Yang; L.M. Conklin; TC Travis; CG Whitney; LJ Anderson; SJ Schrag; T. H. Taylor; B.W. I know everything; High frequency Breiman et al. Application of TaqMan low-density arrays for simultaneous detection of multiple respiratory pathogens., 2011, 49, pp. 2175-2182. DOI: https://doi.org/10.1128/JCM.02270-10.

29. R. Demidenko; D. Razanauskas; K. daniunaite; JR Lazutka; F. Jankevicius; S. Jarmalaite Frequent negative regulation of ABC transporter genes in prostate cancer.

30. D. Szklarczyk; AL Giebel; D. Lyons; A. Junge; S.Wyder; J. Huerta-Cepas; M. Simonovic; NEUES TESTAMENT. doncheva; JH Morris; P. Bork et al. STRING v11: Wide Coverage Protein-Protein Association Networks, Supporting Functional Discovery in Genome-Wide Experimental Datasets., 2019, 47, S. D607-D613. DOI: https://doi.org/10.1093/nar/gky1131.

31.SC Brownhill; C.Taylor; S.A. gene copy number Burchill chromosome 9p21 and prognostic importance of p16 in ESFT., 2007, 96, pp. 1914-1923. DOI: https://doi.org/10.1038/sj.bjc.6603819.

32. S. Anders; W. Huber Differential Expression Analysis for Sequence Count Data., 2010, 11, p. R106. DOI: https://doi.org/10.1186/gb-2010-11-10-r106.

33. S. Brownhill; D Cohen; S. Burchill proliferation index: a continuous model to predict prognosis in patients with tumors of the Ewing sarcoma family., 2014, 9, e104106. DOI: https://doi.org/10.1371/journal.pone.0104106.

34th SS Myatt; CP red fern; SA Burchill p38MAPK: Ewing's sarcoma family with tumor-dependent susceptibility to fenretinide-induced cell death., 2005, 11, pp. 3136-3148. DOI: https://doi.org/10.1158/1078-0432.CCR-04-2050.

35.DS Wishart; COURT. Feunang; AC Guo; EJ It; A.Marcu; JR grant; T. Sayed; D.Johnson; C.Li; Z. Sayeeda et al. DrugBank 5.0: A Major DrugBank Database Update for 2018., 2018, 46, pp. D1074-D1082. DOI: https://doi.org/10.1093/nar/gkx1037.

36. S.L. Freshness; S. daughter; KC cotton; AC Coffman; JF McMichael; JJ song; M Griffith; O.L. Griffith; AH. Wagner Integration of the Drug-Gene Interaction Database (DGIdb 4.0) with open collective collaboration., 2021, 49, pp. D1144-D1 DOI: https://doi.org/10.1093/nar/gkaa1084.

37. P. Pantziarka; C. Verbander; V. Sujatme; I. Rico Capistrano; S. Crispin; B. Wo war es; I. Roman; BIN. Van Nuffel; L. Meheus; Vice President Nach Sukhatme et al. ReDO_DB: The Oncology Drug Reuse Database, 2018, 12, p. 886. DOI: https://doi.org/10.3332/ecancer.2018.886.

38. D. Ochoa; A. Herkules; M.Carmona; D. Vorschläge; A. González-Uriarte; C. Malangone; A. Miranda; L. Fumis; D. Carvalho-Silva; M.Spitzer et al. Open Targeting Platform: Supporting Systematic Identification and Prioritization of Drug Targets., 2021, 49, S. D1302-D1310. DOI: https://doi.org/10.1093/nar/gkaa1027.

39. J. Pinero; JM Ramirez-Anguita; J. Sauch-Pitarch; F. Ronzano; E. Roggen; F. Sanz; LI Furlong The DisGeNET Knowledge Platform for Disease Genomics: 2019 Update., 2020, 48, S. D845-D8

40. P. Pantziarka; I. R. Capistrano; A. DePotter; L. Vandeborne; G. Bouche An Open Access Database of Licensed Cancer Drugs, 2021, 12, p. https://doi.org/10.3389/fphar.2021.627574.

41. P. Pantziarka; L. Vandeborne; G. Bouche A database of clinical trials of drug reuse in oncology., 2021, 12, p. https://doi.org/10.3389/fphar.2021.790952. PMID: https://www.ncbi.nlm.nih.gov/pubmed/34867425.

42.W.Shi; J.Kwon; Y.Huang; J Tan; CG Oh; A. He; C.Zhou; Y. Liu Easy formation of tumor spheroids in large numbers with controllable size and high uniformity., 2018, 8, p. 6837. DOI: https://doi.org/10.1038/s41598-018-25203-3. PMID: https://www.ncbi.nlm.nih.gov/pubmed/29717201.

43.P.Zhang; S.Yu; MMT Justice; TCW spoon; L. sun; Y Jiao; AST Wong; L.T.O. Lee angiotensin II promotes spheroid formation and ovarian cancer metastasis by upregulating lipid saturation and suppressing endoplasmic reticulum stress., 2019, 38, p. 116.DOI: https://doi.org/10.1186/s13046-019-1127-x. PMID: https://www.ncbi.nlm.nih.gov/pubmed/30845964.

44.JW Leeward; JS Sung; YS Park; S.Chung; YH Kim Isolation of single spheroid-forming cells from gastric cancer cell lines: Enrichment of stem cell-like cancer cells., 2018, 65, pp. 197-203. DOI: https://doi.org/10.2144/btn-2018-0046. PMID: https://www.ncbi.nlm.nih.gov/pubmed/30284938.

45. A. Ivascu; M. Kubbies Rapid Generation of Single Tumor Spheroids for High Throughput Toxicity and Cell Function Analysis., 2006, 11, S. 922-932. DOI: https://doi.org/10.1177/1087057106292763.

46. ​​​​C. Lesen; C.Lee; A. Angriff; A.Siu; D. M. Formation of differentielle Ramos spheroids by oral cancer cells., 2014, 34, pp. 1-14. 6945-6

47.A.Z. ajob; T. S. Ramasamy's cancer stem cells as key drivers of tumor progression., 2018, 25, p. 20.DOI: https://doi.org/10.1186/s12929-018-0426-4.

48. MDC Ribeiro-Dantas; D. Oliveira Imparato; M.G.S. dalmoline; c.b. from Farias; EN Brunetto; M. by Cunha Jaeger; R Roesler; M. Sinigaglia; R. J. Siqueira Dalmolin Reverse engineering of the Ewing sarcoma regulator network reveals PAX7 and RUNX3 as major regulators associated with good prognosis., 2021, 13, 1860. DOI: https://doi.org/10.3390/cancers13081860.

49. F. Cider Pineapple; J Lee; TLB Holting; MF East; R. Imle; S. Kutschmann; G.Ammirati; K Ceranski; MJ Carreno Gonzalez; M. Kazan et al. Integrative gene network and functional analyzes identify a key regulator of prognostic metastasis in Ewing's sarcoma., 2022, 21, p. 1.DOI: https://doi.org/10.1186/s12943-021-01470-z.

50. S. E. Molinero; A. Gorthi; N. Bassani; LA Lawrence; BS Iskra; AJR Bishop Reconstruction of Ewing Sarcoma Developmental Context from Mass-Scale Transcriptomics Reveals Characteristics of EWSR1-FLI1 Permissibility., 2020, 12, 948. DOI: https://doi.org/10.3390/cancers12040948.

51.R.Khogar; Flee; Y Chen; M. Ignacio; Leylor emergencies; K.Kitagawa; TH Huang; BE. Phelps; PJ McCarthy Houghton Single Cell RNA Profiling identifies distinct cellular responses to EWSR1/FLI1 downregulation in Ewing's sarcoma cells., 2022, 45, pp. 19-4 DOI: https://doi.org/10.1007/s13402-021 -00640-x. PMID: https://www.ncbi.nlm.nih.gov/pubmed/34997546.

52. T.G. Gruenewald; Sr Alonso; S. Avnet; A. Banito; S Burdach; F. Cider Peanut; G. The pump; M. thistle; H. Dorado-Garcia; J. Garcia-Castro et al. Sarcoma Treatment in the Age of Molecular Medicine., 2020, 12, p. e11131. DOI: https://doi.org/10.15252/emmm. PMID: https://www.ncbi.nlm.nih.gov/pubmed/33047515.

53. C.P. wild; ML-Ulmen; I. Judson; AC Braunen; R. L. Jones; PH Huang The landscape of tyrosine kinase inhibitors in sarcomas: looking beyond pazopanib. 971-9 DOI: https://doi.org/10.1080/14737140.2019.1686979. PMID: https://www.ncbi.nlm.nih.gov/pubmed/31665941.

54.A.Wu; K. Wojtowicz; S. Savary; Y. Hamon; T. Trombik Regulieren ABC-Transporter die Organization der Plasmamembran?, 2020, 25, p. 37. DOI: https://doi.org/10.1186/s11658-020-00224-x.

55. Number Y.; C. Sogawa; Y. Okusa; H.Kawai; Mr. Itagaki; K. This; J Murakami; E. Aoyama; K Ohyama; JI Asumi et al. Depletion of ABCG1 lipid efflux pump triggers intracellular accumulation of extracellular vesicles and reduces aggregation and tumorigenesis of metastatic cancer cells., 2018, 8, p. 376. DOI: https://doi.org/10.3389/fonc.2018.00376.

56. C. Tian; D.Huang; Y.Yu; J Zhang; P. Fang; C. Xie ABCG1 as a potential oncogene in lung cancer., 2017, 13, pp. 3189-3194. DOI: https://doi.org/10.3892/etm.2017.4393.

57. L. YvanCharvet; RESERVE ARMY. pagler; RESERVE ARMY. Seimon; E. Thorp; CL Welch; JL McCarthy witticism; I. Tabas; ARKANSAS. Tall ABCA1 and ABCG1 protect against oxidative stress-induced macrophage apoptosis during efferocytosis. 1861-1869. DOI: https://doi.org/10.1161/CIRCUSSAHA.110.217281.

58. J.Xue; J. Wei; X.Dong; C.Zhu; Y.Li; A song; Z. Liu ABCG1 deficiency promotes endothelial apoptosis through the stress-dependent endoplasmic reticulum signaling pathway., 2013, 63, pp. 101–111. 435-4 DOI: https://doi.org/10.1007/s12576-013-0281-8

59. A.A. apple tree; EN wren; ER Lawlor The importance of fusion protein activity in Ewing's sarcoma and the intrinsic and extrinsic cellular factors that regulate it: a review., 2022, 12, p. 1044707. DOI: https://doi.org/10.3389/fonc.2022.1044707.

60.K.Elsnerova; B. Mohelnikova-Duchonova; E.Cerovska; M. Ehrlichova; I. good; LRob; P.Skapa; M.Hruda; A. Bartakova; J Bouda et al. Gene expression of membrane transporters: importance for the prognosis and course of ovarian cancer., 2016, 35, pp. 1-13. 2159-2 DOI: https://doi.org/10.3892/or.2016.4599.

61.NVlitviakov; E. Garbukov; E.M. Slonimskaia; M.M. Zyganow; EV Denisov; S.V. Wtorushin; K. Kristenko; MV Zav'ialova; NV Cherdyntseva Correlation of metastasis-free survival in breast cancer patients and changes in the direction of expression of multiple drug resistance genes during neoadjuvant chemotherapy., 2013, 59, pp. 1-14. 334-340.

62.NVlitviakov; NV Cherdyntseva; M.M. Zyganow; E.M. Slonimskaya; MK Ibragimova; PV Kazantseva; J. Kzhyshkowska; HE. Choinzonov deletions of multidrug resistance gene loci in breast cancer downregulate their expression and predict tumor response to neoadjuvant chemotherapy., 2016, 7, pp. 7829-7841. DOI: https://doi.org/10.18632/oncotarget.6953.

63.Y.Wang; H.Liu; NORTHEAST. Ready; L.Su; Y. Wei; DC Christiani; Q. Wei Genetic variants in ABCG1 are associated with survival of patients with non-small cell lung cancer., 2016, 138, pp. 2592-2601. DOI: https://doi.org/10.1002/ijc.29991.

64. H.M. gu; G. Li; X.Gao; LG Berthiaume; D. W. Zhang Characterization of palmitoylation of the ATP-binding-cassette transporter G1: effect on protein trafficking and function., 2013, 1831, pp. 1067-1078. DOI: https://doi.org/10.1016/j.bbalip.2013.01.019.

65. H. M. gu; CF Wang; D. W. Zhang Caveolin-1 interacts with the ATP-binding cassette transporter G1 (ABCG1) and regulates ABCG1-mediated cholesterol efflux., 2014, 1841, pp. 847-858. DOI: https://doi.org/10.1016/j.bbalip.2014.02.002.

66. N. Riggi; ML Suva; I. Stamenkovic The Cancer Stem Cell Paradigm in Ewing Sarcoma: What Can We Learn About These Rare Cells From a Rare Tumor?, 2011, 11, S. 143-145. DOI: https://doi.org/10.1586/era.10.235.

67. Uhr Glumaka; BIN. LeBeau The role of CD133 in cancer: a concise review., 2018, 7, p. 18. DOI: https://doi.org/10.1186/s40169-018-0198-1.

68. G. Rappa; O. Fodstad; A. Lorico Stem cell-associated antigen CD133 (prominin-1) is a molecular therapeutic target for metastatic melanoma., 2008, 26, pp. 3008-3017. DOI: https://doi.org/10.1634/stemcells.2008-0601.

69. G. Daniel; S. Morice; F.Redini; F. Verrecchia; B.B. Royer New Insights about the Wnt/beta-Catenin Signalling Pathway in Primary Bone Tumors and Their Microenvironment: A Promising Target to Develop Therapeutic Strategies?, 2019, 20, 3751. DOI: https://doi.org/10.3390/ijms20153751 .

70. F. Tyrode; K. West-Duval; A.anterior; B. Delorme; P Charbord; O. Delattre Mesenchymal Stem Cell Characteristics of Ewing's Tumors., 2007, 11, pp. 421-429. DOI: https://doi.org/10.1016/j.ccr.2007.02.027.

71. C. von Levetzow; XJiang; Y.Gwye; G. von Levetzow; L. hung up; A cooper; JH Hsu; ER Lawlor Modeling of Ewing sarcoma onset in human neural crest cells., 2011, 6, e19305. DOI: https://doi.org/10.1371/journal.pone.0019305. PMID: https://www.ncbi.nlm.nih.gov/pubmed/21559395.

72. U.M. Yawad; MC Cheung; It IS minimal; M.M. Schneiderbauer; LG coniaris; SP Scully-Ewing sarcoma shows racial differences in incidence and sex-related outcome differences: an analysis of 1631 cases from the SEER database, 1973-2005., 2009, 115, pp. 3526-3536. DOI: https://doi.org/10.1002/cncr.24388. PMID: https://www.ncbi.nlm.nih.gov/pubmed/19548262.

73. M.J. Machiela; T.G.P. Gruenewald; D Surdez; S.Reynaud; O.Mirabeau; E. Karlins; RA. Blond; S. Zaidi; S. Grossetete-Lalami; S Ballet et al. A genome-wide association study identifies several novel loci associated with susceptibility to Ewing's sarcoma., 2018, 9, p. 3184. DOI: https://doi.org/10.1038/s41467-018-05537-2. PMID: https://www.ncbi.nlm.nih.gov/pubmed/30093639.

74. R.L. Randall; SL lessnick; K.B. Jones; LG wow; JE Cummings; L. Cannon-Albright; J.D. McCarthy Schiffman Is there a predisposition gene for Ewing's sarcoma?, 2010, 2010, p. 397632. DOI: https://doi.org/10.1155/2010/397632.

75th TGP Grünewald; F. Cider Peanut; D. deafness; E.M. Tomazou; E. of Alava; H.Kovar; PH Sorensen; O. Delattre; Sarcoma by U. Dirksen Ewing., 2018, 4, p. 5.DOI: https://doi.org/10.1038/s41572-018-0003-x.

76. S. V. kumar; P.X. Äh; K. T. Leylor; A. Motazedian; M. Scurr; I. Ghobrial; A. Kämme; L. Zappia; A. Oschlack; Z.B. Stanleyet al. Kidney Microorganoids in Suspension Culture as a Scalable Source of Human Pluripotent Stem Cell-derived Kidney Cells., 2019, 146, p. dev172361. DOI: https://doi.org/10.1242/dev.172361.

77. F. Cellar; R. Ruediger; H. MAthias Toward optimized 3D spheroid monoculture and coculture models of breast cancer for pharmacological research and screening., 2019, 5, pp. 89-101. DOI: https://doi.org/10.3233/JCB-199001.

78. YJWang; M. Herlyn The Emerging Functions of Oct4 in Tumor-Initiating Cells., 2015, 309, S. C709-C718. DOI: https://doi.org/10.1152/ajpcell.00212.2015.

79.J.Kim; H.S. kim; JJ Calce; J Lee; YES. kim; J. Kim Critical role of fibroblast growth factor signaling pathway in Ewing's sarcoma octamer-binding transcription factor 4-mediated cell proliferation and tumorigenesis., 2019, 286, pp. 4443-4472. DOI: https://doi.org/10.1111/febs.14946.

80. V. Chua; SP Chawla; K Zheng; T.Kim; G Del Priore; Phase II study by S. Kim of SM-88 in Ewing's sarcoma and other sarcomas.,

81.V.Chung; SP Chawla; H.Dong; Overfly; R.L. Korn; K.-H.Lim; EM. Christmas; AM. Midday; PHYSICAL EDUCATION. Oberstein; A.J. Oceano et al. Phase II/III study of SM-88 in patients with metastatic pancreatic cancer.,

82. B. Brennan; L Kirton; P. Marec-Bérard; J. Martin-Broto; H. Gelderblom; N. Kaspar; S Strauss; A. Schneider Urgelles; J Anderton; V. Laurence et al. Comparison of two chemotherapy regimens in Ewing's sarcoma (ES): overall and subgroup results of the randomized study Euro Ewing 2012 (EE2012), p. 11500.

83. M. Serra; CM. Hutmacher; M.Pasello; C. Casotti; L. Fantoni; C. Riganti; MC Manara Impact of ABC Transporters in Osteosarcoma and Ewing's Sarcoma: 2021, 10, 2461. https://doi.org/10.3390/cells10092461.

84. T.Perry; M. Fögel; S.Mor; G. Horev; I.Meller; D. Geliebte; J. Issakov; Y. Kollender; A.Smirnov; R. Zaizov et al. Effect of P-glycoprotein expression on outout in the Ewing family of tumors., 2001, 18, pp. 325-334. DOI: https://doi.org/10.1080/088800101300312591.

85. K. Scotlandi; MC Vor; CM. hat maker; S.Benin; S.Perdichizzi; M. Pasello; G. Bacci; L.zanella; F. Bertoni; P.Picci et al. Prognostic and therapeutic relevance of HER2 expression in osteosarcoma and Ewing's sarcoma., 2005, 41, pp. 101-116. 1349-1 DOI: https://doi.org/10.1016/j.ejca.2005.03.015.

86. A. Rößner; Y. Ueda; B. Bockhorn-Dworniczak; S. Blasio; A.Peters; P. Wisman; J.Ritter; M. Paulussen; H. Jürgens; W. Bocker Prognostic Implication of P-glycoprotein Immunodetection in Ewing Sarcoma., 1993, 119, pp. 185-189. DOI: https://doi.org/10.1007/BF01624429.

87. C. Mancarella; M. Pasello; S. Ventura; A. crickets; L.Calzolari; L.Toraccio; P.L. Lollini; MD donates; P. Picci; S.Ferrari et al. Insulin-like growth factor 2 mRNA-binding protein 3 is a novel post-transcriptional regulator of Ewing's sarcoma malignancy., 2018, 24, pp. 3704-3716. DOI: https://doi.org/10.1158/1078-0432.CCR-17-2602.

88. E. D. Eisenmann; Z.Talebi; A. Sparreboom; SÜDDAKOTA. Baker Boosting Oral Bioavailability of Anticancer Drugs Through Intentional Drug-Drug Interactions, 2022, 130, S. 23-35. DOI: https://doi.org/10.1111/bcpt.13623.

89. Y. Kang Analysis of Cancer Stem Cell Metastasis in Animal Models of Xenografts., 2009, 568, S. 7-19. DOI: https://doi.org/10.1007/978-1-59745-280-9_2.

90. K. Sampieri; R. Fodde Cancer Stem Cells and Metastases., 2012, 22, pp. 187-193. DOI: https://doi.org/10.1016/j.semcancer.2012.03.002.

91. N. O’Callaghan; V. Dillon; Pater Thomas; M. Fenech A quantitative real-time PCR method for absolute telomere length., 2008, 44, pp. 807-809. DOI: https://doi.org/10.2144/000112761.

figures and tables

Figure 1: Self-renewal capacity of ES cell lines. Single ES cell suspensions were prepared in 0.3% soft agar or 1 to 10,000 cells were seeded into each well of an ultra-low fixation plate (96 wells). Images show colonies formed after 21 days and are representative of three independent experiments. [Please download the PDF to view the image]

Figure 2: Characterization of CD133 cells in ES cell lines. (A) CD133 protein expression in ES cell lines (n=6) was assessed by Western blotting. Equal protein loading was confirmed by examining the blots for alpha-tubulin. The differentially expressed major genes when comparing (B)TC-32 and (C)A673 CD133 positive and negative cells using LIMMA were not significant. The difference in ?Ct between the two groups and the corresponding Q values ​​are displayed. Genes were considered to be significantly differentially expressed if the Q value was < 0.1 and the difference in ΔCt > 2. Genes with Ct values ​​> 35 in both cell populations were excluded. Dark gray = ABC transporter proteins, gray = pluripotency-associated genes, white = Wnt signaling pathway genes. Results from two independent experiments. SHEF-4 ES cells were included at all times as a positive control for CD133 expression. [Please download the PDF to view the image]

Figure 3: Expression of pluripotency genes in SK-N-MC and TTC-466 cells in 2D and 3D spheroid cultures. (A) Summary of the strategy to identify common strain and MDR markers in ES cell lines. RNA (1 µg) from (B) SK-N-MC and (C) TTC 466 cells in 2D and 3D cultures was analyzed by RTqPCR using the TaqMan® Human Stem Cell Pluripotency Matrix. The mRNA expression level is reported by the comparative Ct method after normalizing the target Ct values ​​to the overall mean of all mRNAs. A volcano graph summarizing the differentially expressed mRNAs in 2D and 3D spheroids is shown, showing the Q value (level of importance) and the difference in ?Ct between cells in 2D and 3D spheroids. The vertical dashed lines show ΔCt thresholds of ± ΔCt > 2 (x-axis) and the horizontal line a significant value of Q < 0.1 (y-axis). Black squares = mRNAs were significantly decreased in 3D SK-N-MC spheroids and with change of ?Ct > 2, black diamonds = mRNAs were significantly decreased in 3D TTC 466 spheroids and with change of ?Ct > 2, Black triangles = significantly increased mRNA in 3D spheroids and with a change of ?Ct >2, black circles = mRNA with a ?Ct of ±<2 and Q value > 0.1. Genes with Ct values ​​>35 in 2D cells were excluded. Comparing (D) SK-N-MC and (E) TTC 466 cells in 2D and 3D cultures with LIMMA shows important differentially expressed genes with a Q value <0.2. The difference in ?Ct between the two groups and the corresponding Q values ​​are displayed. Genes were considered to be significantly differentially expressed if the Q value was < 0.1 and the difference in ?Ct > 2. [Download PDF to view image]

Figure 4: Expression of ABC transporter proteins by SK-N-MC and TTC-466 cells in 2D and 3D spheroid cultures. RNA (1 µg) from (A) SK-N-MC and (D) TTC 466 cells in 2D and 3D cultures was analyzed by RTqPCR using the TaqMan® Human ABC Transporter Array. The mRNA expression level was expressed as the mean ± SEM after normalizing the target Ct values ​​to the overall mean of all mRNAs using the comparative Ct method. The results are the mean of two independent experiments. * = genes with change of ?Ct > 2 and Q value < 0.1 when comparing SK-N-MC cells in 2D and 3D cultures. Genes with Ct values ​​>35 in 2D cultured SK-N-MC cells were excluded. Volcano plot summarizing differentially expressed ABC transporter mRNAs in (B) SK-N-MC and (E) TTC 466 2D and 3D spheroids and the Q value (level of significance) and the difference in ?Ct between 2D cells and 3D spheroids shows. The vertical dashed lines indicate ΔCt thresholds of +/- ΔCt > 2 (x-axis) and the horizontal line a significant value of Q < 0.1 (y-axis). Black square = CFTR, which was significantly decreased, and black triangle = ABCG1, which was significantly increased in 3D spheroids compared to 2D cultures. Black circles = ABC carrier proteins with a ?Ct of ±<2. All differentially expressed genes have a Q value < 0.1. The mean Ct values ​​of ABC transporter mRNAs in (C) SK-N-MC and (F) TTC 466 cells grown in 2D and 3D, the difference in ΔCt between the two groups and the corresponding Q values ​​are shown. Gray = genes with change in ?Ct > 2 and Q value < 0.1. (G) ABC transporters and pluripotency mRNA (n=140) detected (Ct values ​​< 35), highly expressed (Ct values ​​< 25), and undetected (Ct values ​​> 35) in 3D spheroids of TTC 466 and SK-N-MC cells. Venn diagrams show the number of undetected, unique, or common mRNAs for each cell line. (H) Validation of ABCG1 mRNA expression by RTqPCR and reported as 2[sup.-αCt] in SK-N-MC cells grown in 2D and 3D culture; ABCG1 expression is normalized to the endogenous control gene PPIA and the control cell line SHEF-4. RNA expression was compared between 2D and 3D SK-N-MC cells using a two-tailed t-test. The results are representative of 2 independent experiments. (I) Increased ABCG1 protein expression in SK-N-MC cells grown as 3D spheroids compared to 2D cultures was validated by Western blotting. Equal loading of each protein was confirmed by probing the Western blot for β-actin. The results are representative of 2 independent experiments. (J) High expression of ABCG1 protein in the outer 50 µm region of SK-N-MC spheroids detected by IHC; the nuclei are labeled with hematoxylin. Black scale bar = 100 µm. IgG control = SK-N-MC spheroid section incubated with the isotype control antibody (4 µg/ml, negative control mouse IgG1, X0931 (Dako) and 20 µg/µl, serum control Ig-Mix normal rabbit, 086199 (Life Technologies), stained with hematoxylin [download PDF to view image]

Figure 5: Functional role and characterization of ABCG1 in ES. (A) Representative scatterplot of Annexin V and PI labeling of SK-N-MC cells analyzed by flow cytometry after ABCG1 knockdown using SMARTpool:Accell ABCG1 siRNA and untargeted control siRNA. Upper left quadrant shows mean percentage of necrotic cells, upper right = apoptotic cells, lower left = viable cells and lower right early apoptotic cells. The percentage of apoptotic cells was compared using a non-parametric Mann-Whitney two-tailed t-test. (B) Canonical ABCG1 RNA transcripts were downloaded from Ensembl.gov and annotated with the ENST prefix and a unique transcript number. * = No protein produced from this transcript (www.ensembl.org, accessed 27 April 2016), # = no protein produced from this transcript, intron retained (www.ensembl.org, accessed 27 April 2016) (C ) Canonical Transcripts 1 to 11 were detected in SK-N-MC cells in 2D and 3D spheroids. In addition, two new transcripts (transcript 1 and 2) were identified in 3D spheroids. Orange box = exon, gray arrows = direction of transcription, dashed box = missing exon 5. (D) The novel transcript 1 was the ABCG1 transcript, which was more differentially expressed in 3D spheroids compared to 2D cultured cells. [Please download the PDF to view the image]

Figure 6: Expression profile of ABCG1, CD133 and MRP1 in ES cell lines and patient-derived cells. (A) The percentage of SK-N-MC cells grown in 2D and 3D cultures expressing ABCG1 and CD133 was quantified by flow cytometry; 10,000 events were examined for each condition. The percentage of positive cells is presented as mean ± SEM (n=6). The expression of ABCG1 and CD133 in SK-N-MC cells grown in 2D and 3D cultures was compared using a two-tailed Mann-Whitney nonparametric t-test. (B) Representative scatter plot of ABCG1 and CD133 expression of SK-N-MC cells grown in 2D and 3D disaggregated cultures analyzed by flow cytometry. Quadrants represent cells with expression levels ABCG1 low CD133 high (top left), ABCG1 high CD133 high (top right), ABCG1 low CD133 low (bottom left), and ABCG1 high CD133 low (bottom right). Red dots = cells in 3D culture, blue dots = cells in 2D culture. (C) Protein expression of ABCG1 and CD133 in SK-N-MC cells grown in 2D and 3D culture was quantified by flow cytometry. Data are presented as a fold change in fluorescence expressing the mean fluorescence of the target protein in SK-N-MC cells grown in 3D compared to cells grown in 2D. Results show the mean ± SEM for two independent experiments, 3 replicates per experiment (n=6). The expression of ABCG1 and CD133 in 2D and 3D cultures was compared using an unpaired two-tailed t-test. (D) Expression of ABCG1 protein in CD133-positive and CD133-negative TC-32 cells detected by Western blotting. Equal protein loading was confirmed by examining the blots for β-actin. The results are representative of 2 independent experiments. (E) ABCG1 protein expression in patient-derived ES cells determined by Western blot. Equal protein loading was confirmed by examining the blots for β-actin. The results are representative of 2 independent experiments. (F) There was no correlation between the expression level of CD133 and the ability to produce progeny from a single cell. The expression level of CD133 was quantified by flow cytometry and the ability to produce progeny from a single cell was assessed after 21 days in 2D adherent culture. The correlation was examined by linear regression. Open circles = cell lines, filled circles = patient-derived cells. (G) Summary of ABCG1 and MRP1 protein expression in ES cells grown in 2D and 3D culture. Salmon circle = cultured 2D ES cells, orange circle = cultured 3D ES cells (spheroids), blue circle = ABCG1 protein, green circle = CD133 protein, black line = plasma membrane. [Please download the PDF to view the image]

Figure 7: Identification of candidate therapeutic targets with common strain and chemoresistance. (A) Summary of the strategy to identify common strain and MDR markers in preclinical ES models. (B) Genes associated with stemness and MDR were upregulated in 3D spheroids and patient-derived ES-CSCs. Full and alternate gene names, the Ensembl gene abbreviation (gene), and the data from which the genes were identified (gene source) are displayed. The gene source is given as 1 or 2, where 1 = target gene identified by transcriptome analysis of patient-derived ES-CSCs [17] and 2 = target gene identified by comparison of the transcriptome of SK-N-CSCs, MC and TTC 466 grown in 3D and 2D (Figure 3 and Figure 4). (C) Channeling to identify drug candidates based on 21 stem cell and chemoresistance genes. Green ovals = drug reuse data sources, pink ovals = public data sources, blue squares = number of targets or drugs in each step. (D) Predicted interaction between the 13 common stemness and chemoresistance genes (excluding p-glycoprotein and MRP1) that have corresponding drug candidates using STRING [30]. Five of the genes are known to bind to and regulate vital biological processes and may be candidates for prognostic biomarkers and/or therapeutic targets in SE. Pink lines = known interaction determined experimentally, taken from the BIND, DIP, GRID, HPRD, IntAct, MINT, and PID databases. Blue lines = known interaction based on data from selected Biocarta, BioCyc, GO, KEGG and Reactome databases. Gray lines = proteins reported as co-expressed. Green lines = interactions identified through text mining. Filled nodes = known or predicted 3D structure. Colored nodes = first layer of interacting proteins. White nodes = second layer of interacting proteins. Parental cells = patient-derived bulk population of ES cells from which the ES-CSCs were derived [14]. [Please download the PDF to view the image]

Table 1: Self-renewal capacity of ES cell lines. A single cell suspension of each cell line was seeded onto soft agar or a 96-well ultra-low adherence plate at a density of 1 to 10,000 cells. The number of colonies was counted after 21 days. For studies on soft agar, the mean percentage (± SEM) of colony formation is expressed as the number of colonies counted after 21 days relative to the number of cells plated. For cells seeded onto an extremely low adherence plate, the mean percentage (±SEM) of colony formation is expressed as the number of wells containing a colony of >5 cells relative to the total number of wells seeded with cells. The results are representative of three independent experiments. Percent colony formation was compared by ANOVA and the Bonferroni post hoc test.

Table 2: Drugs targeting the common therapeutic targets of trunking and chemoresistance. Drugs targeting common ES strain and chemoresistance targets used in human clinical trials identified with our in-house portfolio. Drug names, direct molecular targets, whether the drug has already been used to treat cancer patients and whether it is patented or not are displayed. The study number, phase and status, cancer type and detailed study information from clinical studies are listed. NA = No phase 3 or phase 4 trials have been registered for the drug. BNFC = British National Formulary - Children, a directory of all medicines authorized for use in children. Y = yes, N = no. * = direct targets of fostamatinib ES (KIT, STK10, SRC, SIK1, RPS6KA1, ROCK2, RET, PTK2B, PTK2, PRKG2, PRKCD, PLK4, PLK3, PLK1, PIM3, PIK3CG, PIK3CD, PDGFRB, PDGFRA, PAK3, PAK1, NTRK3 , NTRK2, NTRK1, NEK2, MST1R, MAPK14, MAP3K3, MAP2K2, LYN, LIMK1, KDR, ITK, INSRR, INSR, GSK3A, MTOR, FLT4, FLT3, FLT1, FGFR1, ERN1, ERBB4, ERBB2, EPHB4, EPHB2, EPHA3 , EPHA2, EIF2AK4, EGFR, DYRK1B, DCLK3, DCLK1, CSK, CSF1R, CHEK2, CHEK1, CDK4, BTK, BRAF, AXL, AURKB, AURKA, ALK, ZAP70, WEE1, TNK2, TNK1, TIE1, TGFBR2, TEK, TAOK3 , STK36, STK33, MET, JAK1, ABL1, SYK).

author links):

[1] Childhood Cancer Research Group, Leeds Institute of Medical Research, St James's University Hospital, Beckett Street, Leeds LS9 7TF, UK

[2] Cancer Fund, Brusselsesteenweg 11, 1860 Meise, Belgium

Author's note(s):

[*] Korrespondenz: e.a.roundhill@leeds.ac.uk (E.A.R.); s.a.burchill@leeds.ac.uk (S.A.B.)

DOI: 10.3390/Krebs15030769