Purpose: SLFN11 was identified as a critical determinant of response to DNA-targeted therapies by analyzing gene expression and drug sensitivity of NCI-60 and CCLE datasets. However, how SLFN11 is regulated in cancer cells remained unknown. Ewing sarcoma, which is characterized by the chimeric transcription factor EWS-FLI1, has notably high SLFN11 expression, leading us to investigate whether EWS-FLI1 drives SLFN11 expression and the role of SLFN11 in the drug response of Ewing sarcoma cells.
Experimental Design: Binding sites of EWS-FLI1 on the SLFN11 promoter were analyzed by chromatin immunoprecipitation sequencing and promoter-luciferase reporter analyses. The relationship between SLFN11 and EWS-FLI1 were further examined in EWS-FLI1-knockdown or -overexpressing cells and in clinical tumor samples.
Results: EWS-FLI1 binds near the transcription start site of SLFN11 promoter and acts as a positive regulator of SLFN11 expression in Ewing sarcoma cells. EWS-FLI1–mediated SLFN11 expression is responsible for high sensitivity of Ewing sarcoma to camptothecin and combinations of PARP inhibitors with temozolomide. Importantly, Ewing sarcoma patients with higher SLFN11 expression showed better tumor-free survival rate. The correlated expression between SLFN11 and FLI1 extends to leukemia, pediatric, colon, breast, and prostate cancers. In addition, expression of other ETS members correlates with SLFN11 in NCI-60 and CCLE datasets, and molecular experiments demonstrate that ETS1 acts as a positive regulator for SLFN11 expression in breast cancer cells.
Conclusions: Our results imply the emerging relevance of SLFN11 as an ETS transcription factor response gene and for therapeutic response to topoisomerase I inhibitors and temozolomide–PARP inhibitor combinations in ETS-activated cancers. Clin Cancer Res; 21(18); 4184–93. ©2015 AACR.
See related commentary by Kovar, p. 4033
DNA-damaging agents, such as topoisomerase I inhibitors, are widely used for the treatment of human cancers. Emerging studies suggest the synergistic effects of inhibitors of PARP combined with temozolomide for treating Ewing sarcoma. Recently, SLFN11 has been suggested as a predictor for the sensitivity of cancer cells, and importantly SLFN11 is capable of sensitizing cancer cells to DNA-damaging agents. Here, we show that SLFN11 expression is transcriptionally activated by ETS transcription factors EWS-FLI1 and ETS1. SLFN11 may be a prognostic marker for the tumor-free survival of Ewing sarcoma patients. Our results further show that EWS-FLI1–activated SLFN11 expression sensitizes Ewing sarcoma cells to camptothecin and PARP inhibitors plus temozolomide combinations, suggesting the emerging relevance of SLFN11 for the sensitivity of ETS-overexpressing cancer to DNA-damaging agents.
The family of Schlafen genes (SLFN), which is only found in mammals, has been reported to regulate several key biologic functions including cell-cycle arrest, differentiation, and cancer cell invasion (1–3). In addition, SLFN11 was recently discovered as a dominant response factor of cancer cells to topoisomerase I inhibitors (4, 5). Knockdown of SLFN11 increases chemoresistance of cancer cells to a broad range of DNA-damaging agents (4, 6), and ectopic expression of SLFN11 sensitizes colon cancer cells to topoisomerase I inhibitors (7), consistent with the involvement of SLFN11 in the DNA damage response (4). Concurrently, David and colleagues demonstrated an antihuman immunodeficiency virus-1 (HIV-1) function of SLFN11 due to replication inhibition by selective suppression of viral protein synthesis in a codon usage–dependent manner (8). The emerging relevance of SLFN11 in cancer biology and therapeutic responses incited us to investigate how SLFN11 expression is regulated in cancer cells.
Ewing sarcoma is a malignant tumor primarily occurring in bones or soft tissues of children and young adults. It readily metastasizes to other organs including the lungs, bones, and bone marrow (9). The overall survival of Ewing sarcoma patients remains poor; 25% of patients with localized tumor and 75% of patients with metastasis do not have durable therapeutic responses (9, 10). Approximately 85% of Ewing sarcoma are characterized by the chromosome translocation (11;22)(q24;q12) encoding the chimeric transcription factor EWS-FLI1 comprising the amino-terminal transactivation domain of EWSR1 fused to the carboxy-terminal ETS DNA-binding domain of FLI1 (11). FLI1 is a member of the family of ETS transcription factors containing a highly conserved domain that recognizes ETS core consensus sites (GGAA/T); the flanking DNA sequence providing the affinity and specificity (12, 13). EWS-FLI1 is oncogenic by regulating multiple target genes through binding to typical ETS core consensus sites or GGAA microsatellites (14, 15). Moreover, EWS-FLI1 is able to colocalize with E2F3 on proximal promoters to activate target genes (16).
Ewing sarcoma cells expressing EWS-FLI1 have been shown to be remarkably sensitive to the inhibitors of PARP (17). Emerging studies suggest that PARP inhibitors combined with temozolomide show significant synergism in Ewing sarcoma cells (18, 19), with the combination inducing more cytotoxic PARP–DNA complex than either single agent (20, 21). In addition, Barretina and colleagues showed that Ewing sarcoma cell lines with high SLFN11 are highly sensitive to topoisomerase I inhibitors (5). Motivated by these findings, we hypothesized that EWS-FLI1 plays a role in the expression of SLFN11 and the response of Ewing sarcoma cells to camptothecin and the combinations of PARP inhibitor plus temozolomide. In the current study, we demonstrate a direct role of EWS-FLI1 as a positive transcriptional regulator of SLFN11, and implicate SLFN11 in the sensitivity to topoisomerase I inhibitor and the synergistic effects of PARP inhibitors plus temozolomide in relationship with EWS-FLI1 expression. We further extend SLFN11 regulation by another ETS transcription factor ETS1 in breast and other cancers.
Materials and Methods
Cells, plasmids, and drugs
293T and A673 cells were maintained in DMEM supplemented with 10% FBS, 2 mmol/L l-glutamine and antibiotics in 5% CO2 at 37°C. HT1080/GFP, HT1080/EWS-FLI1, and Hs 343.T cells were maintained in RPMI supplemented with 10% FBS, 2 mmol/L l-glutamine, 4 μg/mL balsticidin, and antibiotics in 5% CO2 at 37°C. A673 cells with doxycycline-inducible shRNA targeting EWS-FLI1 (ASP14) were kindly provided from Dr. Heinrich Kovar (22) and maintained in DMEM with 10% FBS, 2 mmol/L l-glutamine, 2 μg/mL balsticidin, 50 μmol/L Zeocin, and antibiotics in 5% CO2 at 37°C. pCB6/EWS-FLI1–expressing plasmid was a gift from Dr. Suzanne Baker. pCMV6-ETS1–expressing plasmid was obtained from OriGene. Drugs were obtained from the NCI Drug Developmental Therapeutics Program (DTP).
Chromatin immunoprecipitation sequencing data analysis
The ChIP-Seq dataset for the chromosome binding region of EWS-FLI1 in A673 Ewing sarcoma cell line were obtained as described in a previous study (16). The tag data from ChIP-Seq analysis were uploaded to the Integrative Genomics Viewer from UCSC (http://www.genome.ucsc.edu) and the tag density of EWS-FLI1 binding near the transcription start site (TSS) of SLFN11 was analyzed.
ETS-binding motif prediction
The potential ETS-binding sites in EWS-FLI1 binding region of SLFN11 promoter (33,700,532–33,700,847) were predicted using the JASPAR database, including the DNA-binding patterns of transcription factors and other sequence-specific DNA binding proteins (http://jaspar.genereg.net; refs. 23, 24). The predicted sites with a relative score more than 0.9 were considered as the potential ETS-binding sites.
Construction of human SLFN11 promoter
The SLFN11 promoter between −840 bp upstream and +460 bp downstream of TSS) was amplified by PCR with the primers (forward, 5′-TTTCTCTATCGATAGGTACCAGTGCGGCATTAACCGCTGCT-3′; reverse, 5′-ACGCGTAAGAGCTCGGTACCCGGACAGGGGAGAAAAGCACA-3′), and cloned into pGL3-basic luciferase reporter vector following the manufacturer's instructions (In-Fusion HD Cloning Kit; Clontech Laboratories, Inc.). The three potential EWS-FLI1 binding sites were mutated using QuikChange II XL Site-Directed Mutagenesis Kit according to the manufacturer's instructions (Stratagene). The primers used to introduce mutations were: mt+91: 5′-TCGCGGGCTTAGCAGACCTATACATTGGCTCTTGCATCTCC-3′; mt+181: 5′-ACCTGGGCGCCTCCAGCATGACGCTAAGGGGGCTTC-3′; mt+201, 5′-ACGCTAAGGGGGCTTCCATGGCGCTGGAGCTTGAGAG-3′. All constructs were verified by nucleotide sequencing.
Promoter-luciferase reporter assay
Cells were cotransfected with pGL3-SLFN11 promoter plasmid and pCMV-β-galactosidase plasmid using Lipofectamine 2000 Transfection Reagent according to the manufacturer's instructions (Life Technologies). After transfection for 24 hours, luciferase activity was assayed by the luciferase assay kit (Promega) and β-galactosidase activity was determined by the Tropix Galacto-Star chemiluminescent reporter gene assay (Applied Biosystems) following the manufacturer's instructions. The luminescent signal was measured by the EnVision 2104 Multilabel Reader (PerkinElmer). β-Galactosidase activity was the internal control of transfection efficiency for the normalization of luciferase activity.
Quantitative real-time PCR
The cells were washed with PBS twice, and total RNAs were extracted using TRIzol reagent (Invitrogen) and reverse transcribed to complementary DNA by SuperScript II Reverse Transcriptase kit (Invitrogen) according to the manufacturer's instructions. The primers used to amplify specific gene were as follows: SLFN11 (forward 5′-GGCCCAGACCAAGCCTTAAT-3′ and reverse 5′-CACTGAAAGCCAGGGCAAAC-3′), FLI1 (forward 5′-CCAAAGTGCACGGCAAAAGA-3′ and reverse 5′-GGCATGGTAGGAAGGCATGT-3′) and GAPDH (forward 5′-TCAACGACCACTTTGTCAAGCT-3′ and reverse 5′-GTGAGGGTCTCTCTCTTCCTCTTGT-3′). Quantitative real-time PCR was carried out with FastStart Universal SYBR Green Master (Roche Applied Science) by the 7900HT Fast Real-Time PCR System (Applied Biosystems) according to the manufacturer's instructions. The melting curve was generated to confirm the amplification specificity. GAPDH was used as the internal control. The relative level of gene expression was determined using the 2(−ΔΔCt) method.
Cells were washed with PBS twice and then lysed in 50 mmol/L Tris-HCl, pH 7.5, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% Triton X-100, 0.1% SDS supplemented with protease inhibitor cocktail (Roche). The lysates were separated in 8% of SDS-PAGE and transferred to polyvinylidene fluoride membranes (Millipore). Immunoblotting was performed with antibodies specific to SLFN11 (sc-374339, Santa Cruz Biotechnology), FLI1 (sc-356, Santa Cruz Biotechnology), GSK3β (610201, BD Transduction Laboratories), or actin (MAB1501, Chemicon International), followed by horseradish peroxidase–conjugated secondary antibodies (GE Healthcare Life Sciences). Actin was used as the loading control.
Microarray and RNA-seq data
The whole-genome expression profile analyses of the NCI-60 (CellMiner tools: http://discover.nci.nih.gov/cellminer) and CCLE (http://www.broadinstitute.org/ccle/home) cell lines were recently described (5, 25). The gene expression profiles of 44 Ewing sarcoma tumor samples and 18 normal skeletal muscle tissues by microarray analysis (26) were downloaded from the Gene Expression Omnibus website (GSE17674, http://www.ncbi.nlm.nih.gov/geo; ref. 27). The raw signal intensities were normalized by Robust Multichip Average (RMA) method and Log2-transformed. The median-normalized expression of SLFN11 and FLI1 in the dataset of 163 pediatric cancers was obtained from Pediatric Preclinical Testing Program (http://home.ccr.cancer.gov/oncology/oncogenomics; ref. 28). The RNA-Seq data for gene expression in the datasets of colon, breast, and prostate cancers were retrieved from TCGA portal (http://cancergenome.nih.gov).
siRNA transfection and cell viability assay
Cells were transfected with 10 nmol/L of SLFN11-targeting, ETS1-targeting, or nontargeting siRNAs (Thermo Scientific Dharmacon) using Lipofectamine 2000 Transfection Reagent according to the manufacturer's instructions (Life Technologies). After transfection for 1 day, cells were treated as indicated. Cell viability was measured by MTS assay (Promega). The cell viability was calculated by setting untreated cells as 100%.
The two-tailed independent samples t test was applied to determine the statistical significance of the differences between the two experimental groups for the promoter-luciferase reporter assay, quantitative real-time PCR, and the cell viability assay. Nonlinear regression was used to calculate the inhibitory concentration 50% (IC50) of camptothecin. The association between the expression levels of two individual genes in a dataset was evaluated by the Pearson correlation, and where multiple comparisons were made, the P values were adjusted using the Bonferroni method. The Kaplan–Meier curves based on SLFN11 expression higher or lower than the median in the cohort of 44 Ewing sarcoma patients (26) were analyzed using the Mantel–Cox log-rank test. The test results with P < 0.05 were considered to be statistically significant unless otherwise stated.
EWS-FLI1 binds to and activates the SLFN11 promoter
To determine whether EWS-FLI1 regulates SLFN11, we analyzed the results of ChIP-Seq analysis from Ewing sarcoma A673 cells (16) by focusing on the SLFN11 promoter region. Notably, preferential binding of EWS-FLI1 was found at the TSS of SLFN11 (Fig. 1A). To determine whether EWS-FLI1 directly activates SLFN11 transcription, we cloned the promoter region of SLFN11 (from −840 to +460 bp from the TSS), and performed promoter-luciferase reporter assays. For these experiments, we used the EWS-FLI1–overexpressing HT1080 cells (HT1080/EWS-FLI1) and GFP-expressing cells (HT1080/GFP) as reference control (29). The results showed that the induction of SLFN11 promoter activity was selectively enhanced in HT1080/EWS-FLI1 cells, but not in the HT1080/GFP control cells (Fig. 1B). In addition, transient expression of EWS-FLI1 in 293T cells significantly enhanced SLFN11 promoter activity (Fig. 1C). These results indicate that EWS-FLI1 binds to and activates the SLFN11 promoter.
Next, we analyzed the cis-acting sequence responsible for EWS-FLI1–mediated regulation of the SLFN11 promoter. The EWS-FLI1 ChIP-Seq data showed the highest sequence tag density in EWS-FLI1 binding sites on the SLFN11 promoter, approximately at genomic position 33,700,733 on chromosome 17, which is 91 bp downstream from the SLFN11 TSS (Fig. 1A, +91 bp). In addition, two potential ETS core consensus sites (+181 bp, +201 bp from the TSS) were annotated by JASPAR, the open-access database of transcription factor binding motif (23, 24). To determine the relative contribution of these putative FLI1 binding site(s) to the EWS-FLI1-induced SLFN11 promoter activity, we tested 3 SLFN11 promoter mutations by site-directed mutagenesis (mt+91, mt+181, mt+201). Luciferase reporter assays showed that individual mutations at positions +91 and +201 reduced SLFN11 promoter activity by more than 80%, whereas the +181 mutation had no significant effect (Fig. 1D). SLFN11 promoter activity in 293T/EWS-FLI1 cells was also suppressed by approximately 90% for mt+91 and 50% for mt+201, whereas mt+181 did not affect SLFN11 promoter activity (Fig. 1E). These results indicate that consensus sequences +91 and +201 are critical for activation of the SLFN11 promoter by EWS-FLI1.
EWS-FLI1 regulates the expression of SLFN11
To demonstrate that EWS-FLI1 activates endogenous SLFN11, SLFN11 mRNA and protein levels were examined in HT1080/EWS-FLI1 and HT1080/GFP cells by quantitative real-time PCR and Western blotting analyses (Fig. 2A and B). HT1080/EWS-FLI1 cells displayed significant higher SLFN11 both at the mRNA and protein levels than HT1080/GFP cells. To confirm the regulation of SLFN11 expression by EWS-FLI1, we used the ASP14 cell line, which is derived from the A673 cells, but contain a doxycycline-inducible shRNA targeting EWS-FLI1 (22). The results showed that doxycycline significantly decreased both EWS-FLI1 and SLFN11 expression in ASP14 cells, but not in the parent A673 cells not expressing the EWS-FLI1 shRNA (Fig. 2C and D). On the basis of these results, we conclude that EWS-FLI1 transcriptionally regulates the expression of SLFN11.
SLFN11 expression is correlated with FLI1 expression and may predict tumor-free survival in Ewing sarcoma patients
Next, we tested the correlations between FLI1 and SLFN11 expression using gene expression microarray data of Ewing sarcoma tumor samples and normal skeletal muscle tissues (26, 30). As shown in Fig. 3A, FLI1 and SLFN11 expressions were significantly correlated (r = 0.83, P < 0.0001). We also examined the correlation between FLI1 and SLFN11 expression in a cohort of pediatric cancer samples including primary tumors, xenograft, and cell line samples of Ewing sarcoma, rhabdomyosarcoma, osteogenic sarcoma, acute lymphoblastic leukemia (ALL), brain cancers, and Wilms' tumor from the NCI Pediatric Preclinical Testing Program (http://home.ccr.cancer.gov/oncology/oncogenomics; ref. 28). FLI1 and SLFN11 expression levels were highly correlated in 163 samples of pediatric cancers (r = 0.68, P < 0.0001; Fig. 3B). Ewing sarcoma and ALL, which express high EWS-FLI1 and FLI1, respectively, showed the strongest SLFN11 expression (Fig. 3B). Brain cancers and rhabdomyosarcoma displayed a strong correlated expression of FLI1 and SLFN11 (Fig. 3B).
We also evaluated whether SLFN11 expression might be a prognostic marker for the tumor-free survival of Ewing sarcoma patients. The patients were divided into two groups by SLFN11 higher or lower than median expression and analyzed by log-rank test. Ewing sarcoma patients with higher SLFN11 expression exhibited better prognosis than those with lower SLFN11 expression (P = 0.0046, HR = 3.17, 95% CI, 1.43–7.05; Fig. 3C).
EWS-FLI1–mediated SLFN11 expression determines DNA damage response
Because SLFN11 expression sensitizes cells to DNA-damaging agents (4, 5, 7), we tested the impact of EWS-FLI1–mediated SLFN11 expression on the sensitivity of Ewing sarcoma cells to camptothecin, a specific topoisomerase I inhibitor whose derivatives irinotecan and topotecan are widely used in the anticancer armamentarium (31). We found that doxycycline-induced EWS-FLI1 downregulaton significantly reduced the sensitivity of ASP14 cells to camptothecin (IC50 = 94 nmol/L vs. 17 nmol/L in the presence and absence of doxycycline, respectively; Fig. 4A). Similarly, transfection of A673 cells with SLFN11 siRNA increased resistance to camptothecin (IC50 = 75 nmol/L vs. 10 nmol/L in the presence and absence of SLFN11 siRNA, respectively; Fig. 4B). Conversely, HT1080/EWS-FLI1 cells were more sensitive to camptothecin than HT1080/GFP cells (IC50 = 0.29 μmol/L vs. 0.85 μmol/L, respectively; Fig. 4C), and SLFN11 knockdown in HT1080/EWS-FLI1 cells counteracted EWS-FLI1–induced sensitivity to camptothecin in comparison with control siRNA-transfected HT1080/EWS-FLI1 cells (IC50 = 0.62 μmol/L vs. 0.19 μmol/L, respectively; Fig. 4C).
We further evaluated the role of SLFN11 in the synergistic effects of the combination of PARP inhibitors plus temozolomide. ASP14 and A673 cells were relatively insensitive to single agent temozolomide (50–200 μmol/L) or niraparib (1 μmol/L), while the combination of niraparib and temozolomide killed ASP14 and A673 cells efficiently (Fig. 4D and E). ASP14 cells with EWS-FLI1 knockdown and A673 cells with SLFN11 knockdown exhibited resistance to the combination of niraparib plus temozolomide (Fig. 4D and E). These findings demonstrate that the upregulation of SLFN11 by EWS-FLI1 enhances the sensitivity of Ewing sarcoma cells to camptothecin and plays a role in the synergistic effects of PARP inhibitors with temozolomide.
FLI1 and SLFN11 coexpression in other types of cancers
We next examined the correlation of FLI1 and SLFN11 in other types of cancers. By analyzing the gene expression dataset of 1,036 cancer cell lines from the Broad Insitute (the Cancer Cell Line Encyclopedia, CCLE; ref. 5), we observed that the majority of ALL, acute myeloid leukemia, and chronic myeloid leukemia cell lines showed a strong correlated expression between FLI1 and SLFN11 (Fig. 5A). We also evaluated the expression of FLI1 and SLFN11 in tumor samples of colon, breast, and prostate cancers obtained from the gene expression datasets of The Cancer Genome Atlas (TCGA). The results showed that FLI1 and SLFN11 expression are highly correlated across 233 colon cancers (Fig. 5B; r = 0.62, P < 0.0001), 994 breast cancers (Fig. 5C; r = 0.45, P < 0.0001), and in 195 prostate cancers (Fig. 5D; r = 0.19, P = 0.007). Together, these results imply that FLI1 regulates SLFN11 expression not only in Ewing sarcoma but also in other pediatric cancers, leukemia, colon, breast, and prostate cancers.
ETS1 and SLFN11 coexpression in breast cancer and prostate cancer
To examine whether other ETS transcription factors (12) might regulate SLFN11 expression, we tested the correlation between the expression levels of 27 ETS members and SLFN11 in the NCI-60 cancer cell lines (CellMiner tools: http://discover.nci.nih.gov/cellminer; refs. 25, 32). ETS1, FLI1, ETV4, and EHF showed a significant correlation with SLFN11 expression in the gene expression dataset of the NCI-60 cancer cell line panel (Table 1; |r| > 0.43, P < 0.05). Extension of correlation analysis of ETS members with SLFN11 in the CCLE dataset revealed significant correlation of FLI1, ETS1, ERG, SPI1, ELF3, ETV4, and EHF with SLFN11 (Table 1; |r| > 0.20, P < 0.01). ETS1 also exhibited a strong positive correlation with SLFN11 in the TCGA datasets of breast cancers (Fig. 5E; r = 0.43, P < 0.0001), and prostate cancers (Fig. 5F; r = 0.54, P < 0.0001).
|NCI60a .||CCLEa .|
|Gene symbol .||Correlation coefficient .||Gene symbol .||Correlation coefficient .|
|NCI60a .||CCLEa .|
|Gene symbol .||Correlation coefficient .||Gene symbol .||Correlation coefficient .|
aETS genes are ranked by decreasing Pearson correlation coefficient.
bPearson |r| > 0.43, P < 0.05 (Bonferroni-adjusted).
cPearson |r| > 0.20, P < 0.01 (Bonferroni-adjusted).
Promoter-luciferase reporter analyses were performed to test the causality between ETS1 and SLFN11 expression. Figure 6 shows that ETS1 was able to activate the SLFN11 promoter in 293T cells (Fig. 6A). Moreover, SLFN11 promoter activity was decreased by mutating the ETS consensus sequences within the SLFN11 promoter (see Fig. 1). The mutant mt+91 showed an approximately 80% reduction and the mt+201 mutant approximately 60% reduction (Fig. 6B), indicating that positions +91 and +201 in the SLFN11 promoter are important for ETS1-mediated activation of SLFN11. Finally, siRNA-mediated ETS1 knockdown suppressed SLFN11 expression at the RNA and protein levels and resulted in resistance to camptothecin treatment in breast cancer cell line Hs 343.T (Fig. 6C–E). These results are consistent with a role of ETS1 in the regulation of SLFN11 expression and sensitivity to topoisomerase I inhibitors in breast cancer.
Recent studies revealed that high SLFN11 expression enhances the response of cancer cells to a broad range of DNA-damaging agents (4, 5, 7). Yet, until the current study, there was no information on the regulation of SLFN11 in cancer cells (33). Our study implicates EWS-FLI1 as a causative regulator for SLFN11 expression in Ewing sarcoma. Unlike SLFN11, we found that the other 3 human Schlafen genes SLFN5, SLFN12, and SLFN13 are not strongly correlated with FLI1 expression in the NCI-60 and CCLE datasets, indicating that FLI1 specifically regulates the expression of SLFN11, but not the expression of the other SLFN genes in spite of their common location on human chromosome 17q12 (2).
Our study also demonstrates the involvement of EWS-FLI1–induced SLFN11 expression in the response of Ewing sarcoma cells to camptothecin and to the combination of PARP inhibitors with temozolomide, suggesting that cancers with higher SLFN11 expression might be more sensitive to DNA-damaging agents. Ewing sarcoma tumors initially respond to chemotherapy; however, 30% of patients relapse with relatively less sensitivity to chemotherapies and less than 20% long-term survival rate (34). Thus, further studies are warranted to investigate whether the expression or function of SLFN11 is suppressed in relapsed Ewing sarcoma tumors.
EWS-FLI1 has been considered as a therapeutic target for the treatment of Ewing sarcoma. Knockdown of EWS-FLI1 by antisense cDNA or siRNA suppresses the growth and invasiveness of Ewing sarcoma cells (35, 36). Recently, trabectedin, a marine alkaloid that alkylates DNA at guanine N2 and poisons transcription-coupled repair (37, 38) has been demonstrated to interfere with the activity of EWS-FLI1 and to reverse EWS-FLI1–mediated gene expression signatures (29). In addition, trabectedin showed a synergic effect with the clinical camptothecin derivative SN-38 in Ewing sarcoma cells by inhibiting EWS-FLI1–mediated expression of Werner syndrome helicase, whose deficiency causes cellular hypersensitivity to camptothecins (39, 40). Our experiments demonstrate that knockdown of EWS-FLI1 downregulates SLFN11 expression, and consequently reduces the sensitivity of cancer cells to camptothecin and the combinations of PARP inhibitor temozolomide. Hence, developing SLFN11-activating agents could be viewed as a rationale for combinatorial therapy to enhance the efficacy of DNA-damaging agents. According to our unpublished data, the demethylating agent azacytidine was able to reactivate SLFN11 expression in cancer cells with methylated SLFN11 promoter, providing an avenue to activate SLFN11 expression in the clinical setting.
Besides EWS-FLI1, chromosome translocations (21;22)(q22;q12), which fuse EWSR1 with another ETS transcription factor ERG1 (EWS-ERG), were found in about 5%–10% of Ewing sarcoma (41–43). Both ERG and FLI1 belong to ERG subfamily in ETS transcription factor family with similar DNA-binding domains (12). To understand whether SLFN11 expression is also activated in EWS-ERG–expressing cells, we compared SLFN11 levels in 25 Ewing sarcoma cell lines with EWS-FLI1 and 4 Ewing sarcoma cell lines with EWS-ERG, and found that SLFN11 expression showed no difference between EWS-FLI1 and EWS-ERG–expressing cells (Supplementary Fig. S1). Another chromosome translocation, which encodes a fusion protein TMPRSS2-ERG, was found in approximately 40% of prostate cancer (44). Interestingly, when analyzing the TCGA prostate cancer dataset, we did not observe a significant correlation between SLFN11 and ERG (data not shown) indicating that activation of SLFN11 expression might require other proteins besides ETS transcription factors.
In this study, we also identified ETS1, like FLI1, as a dominant activator of SLFN11 expression in breast cancer. ETS1 has been shown to play an important role in the progression of breast cancer (45) and has been recently linked to the RAS/ERK pathway in carcinomas (46). Bonetti and colleagues recently demonstrated that FLI1 and ETS1 cooperatively contribute to the growth of diffuse large B-cell lymphoma, and influence gene expression related to germinal center differentiation (47). In addition, due to the chromosome gene region of FLI1 and ETS1 are closely adjacent, the 11q24.3 gain might result in FLI1 and ETS1 expression in diffuse large B-cell lymphoma (47). Our findings reveal that both FLI1 and ETS1 are highly correlated with SLFN11 expression in both datasets of the NCI-60 and CCLE, as well as the datasets of breast, colon, and prostate cancers from TCGA. We also observed a positive correlation between the expression levels of FLI1 and ETS1 in the datasets of breast and prostate cancers (Supplementary Fig. S2). Further studies are warranted to understand how FLI1 and ETS1 are coregulated or regulate each other.
In conclusion, our findings reveal an unsuspected connection between SLFN11 and the ETS transcription factors, which are commonly upregulated in cancer and autoimmune diseases. Our study also highlights the potential importance of SLFN11, FLI1 and ETS1 as predictive genomic biomarkers for DNA-damaging agents and combinations of PARP inhibitors with temozolomide.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: S.-W. Tang, L. Cao, J. Murai, M. Yamade, L.J. Helman, Y. Pommier
Development of methodology: F.G. Sousa, Y. Pommier
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S.-W. Tang, F.G. Sousa, J. Khan, Y. Pommier
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S.-W. Tang, S. Bilke, L. Cao, F.G. Sousa, M. Yamade, V. Rajapakse, S. Varma, L.J. Helman, P.S. Meltzer, Y. Pommier
Writing, review, and/or revision of the manuscript: S.-W. Tang, S. Bilke, L. Cao, J. Murai, J. Khan, P.S. Meltzer, Y. Pommier
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y. Pommier
Study supervision: Y. Pommier
The authors thank Dr. Heinrich Kovar (Children's Cancer Research Institute, St. Anna Kinderkrebsforschung, Vienna, Austria) for providing the A673 cells with doxycycline-inducible shRNA targeting EWS-FLI1, and Dr. Suzanne Baker (Department of Developmental Neurobiology, St. Jude Children's Research Hospital, Tennessee) for the EWS-FLI1-expressing plasmid.
This work was supported by the Center for Cancer Research, Intramural NCI program (Z01 BC 006150).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.