Hedgehog signaling is aberrantly activated in hematologic malignancies and solid tumors, and targeting it is a promising therapeutic strategy against these cancers. Resistance to clinically available hedgehog-targeted Smoothened inhibitor (SMOi) drugs has become a critical issue in hedgehog-driven cancer treatment. Our previous studies identified inhibition of BET and CDK7 as two epigenetic/transcriptional-targeted therapeutic strategies for overcoming SMOi resistance, providing a promising direction for anti-hedgehog drug development. To uncover additional strategies for inhibiting aberrant hedgehog activity, here we performed CRISPR-Cas9 screening with an single-guide RNA library targeting epigenetic and transcriptional modulators in hedgehog-driven medulloblastoma cells, combined with tumor dataset analyses. Structure specific recognition protein 1 (SSRP1), a subunit of facilitates chromatin transcription (FACT) complex, was identified as a hedgehog-induced essential oncogene and therapeutic target in hedgehog-driven cancer. The FACT inhibitor CBL0137, which has entered clinical trials for cancer, effectively suppressed in vitro and in vivo growth of multiple SMOi-responsive and SMOi-resistant hedgehog-driven cancer models. Mechanistically, CBL0137 exerted anti-hedgehog activity by targeting transcription of GLI1 and GLI2, which are core transcription factors of the hedgehog pathway. SSRP1 bound the promoter regions of GLI1 and GLI2, while CBL0137 treatment substantially disrupted these interactions. Moreover, CBL0137 synergized with BET or CDK7 inhibitors to antagonize aberrant hedgehog pathway and growth of hedgehog-driven cancer models. Taken together, these results identify FACT inhibition as a promising epigenetic/transcriptional-targeted therapeutic strategy for treating hedgehog-driven cancers and overcoming SMOi resistance.

Significance:

This study identifies FACT inhibition as an anti-hedgehog therapeutic strategy for overcoming resistance to Smoothened inhibitors and provides preclinical support for initiating clinical trials of FACT-targeted drug CBL0137 against hedgehog-driven cancers.

Hedgehog (Hh) pathway is a highly conserved signaling pathway in governing intricate embryonic development and postnatal tissue homeostasis (1). Its activation process involves Hh ligand–induced derepression of transmembrane protein smoothened (SMO) from another transmembrane receptor patched (PTCH1) on cell surface and subsequent emancipation of core transcription factor GLI2 from its negative regulator suppressor of fused (SUFU) in cytoplasm. Then the activated GLI2 will be translocated into nucleus and induce transcription of another core transcription factor GLI1 as well as other Hh target genes (2–4). Aberrantly activated Hh pathway is associated with initiation and maintenance of several hematologic malignancies and solid tumors, such as basal cell carcinoma, SHH-subtype medulloblastoma (SHH-MB), rhabdomyosarcoma, SHH-subtype atypical teratoid rhabdoid tumor (SHH-ATRT), and others (5–9). As targeting Hh pathway proves to be a promising therapeutic strategy against these Hh-driven cancers, tremendous efforts have been dedicated to developing targeted anti-Hh drugs. So far, the biggest success has been achieved on targeting SMO, as multiple SMO inhibitor (SMOi) drugs, which are designed on the basis of a natural compound cyclopamine, have been approved by FDA for treating Hh-driven cancers (10–12). However, primary and acquired resistance to SMOi therapy have been emerging. SMOi resistance could arise from SMO mutations that decrease affinity to current cyclopamine-based SMOi drugs (12), or SMO-downstream activation of Hh pathway due to SUFU loss-of-function mutation and GLI2 or MYCN amplification (13–15). SMOi therapy could also be challenged by noncanonical Hh activation due to SMARCB1 deficiency or compensatory activation of alternative oncogenic pathways (16, 17). Therefore, it is critical to identify new anti-Hh therapeutic strategies that can overcome SMOi resistance.

Epigenetic/transcriptional modulators have been revealed to be critical regulators and effective therapeutic targets in cancer. The efficacy of epigenetic/transcriptional-targeted therapy alone or combined with other therapies have been increasingly recognized in treating various cancer models (18, 19). Of note, we and others have recently demonstrated multiple epigenetic/transcriptional-targeted therapeutic strategies, including BET inhibition, HDAC inhibition, and CDK7 inhibition, could effectively treat Hh-driven cancers in preclinical models and overcome both primary and acquired SMOi resistance via blocking transcriptional activation of GLI and other downstream Hh target genes, thus opening a promising avenue for novel anti-Hh drug development (20–23).

The human facilitates chromatin transcription (FACT) complex, which is composed of two protein subunits called structure specific recognition protein 1 (SSRP1) and suppressor of Ty 16 (SUPT16H), is a histone chaperon that can bind both histones and DNA. It plays important roles in nucleosome remodeling–related biological processes like DNA replication, repair, and transcription (24, 25). It interacts with helicase to reassemble nucleosome during DNA replication and facilitates DNA repair by recognizing and reorganizing chromatin at sites of DNA damage. It initiates transcription by promoting the dissociation of one histone H2A-H2B dimer from the nucleosome, and subsequently destabilizes and restores nucleosome structure to facilitate the passage of RNA polymerase II and the proper nucleosome recovery during transcription elongation (26, 27). FACT complex is barely detectable in well-differentiated cells but highly upregulated in stem cells and tumor cells (28). It has been shown to carry crucial oncogenic functions and may serve as a promising therapeutic target in many cancers, including neuroblastoma, glioblastoma, melanoma, small cell lung cancer, mixed lineage leukemia–rearranged leukemia (29–33). Worth to note, CBL0137, a FACT-targeted inhibitor drug, has entered multiple phase I clinical trials of treating patients with metastatic or unresectable advanced solid neoplasm (NCT01905228), with extremity melanoma or sarcoma (NCT03727789), or with hematologic malignancies (NCT02931110). As a small molecule of curaxins, CBL0137 could intercalate into DNA and change DNA architecture in tumor cells, which results in trapping of FACT complex and its redistribution in chromatin (34, 35). It could further disrupt the spatial organization of chromatin and interactions between enhancers and promoters, resulting in inhibition of FACT-dependent transcription of many essential oncogenes in various caner types, such as MYCN in neuroblastoma as well as SOX2, OCT4, NANOG, and OLIG2 in glioblastoma (36).

To systemically access the therapeutic potential of epigenetic/transcriptional modulators in treating Hh-driven cancer and overcoming SMOi resistance, we performed CRISPR-Cas9 screening of epigenetic/transcriptional-targeted single-guide RNA (sgRNA) library in SHH-MB cells. Through integrative analyses of medulloblastoma datasets and the functional genomics screening results, we identified SSRP1 as a novel Hh-induced essential oncogene that contributes to maintaining Hh-activated oncogenic transcriptional program in Hh-driven cancers. We further demonstrated that the FACT inhibitor drug CBL0137 worked effectively alone or in combination with BET inhibitor or CDK7 inhibitor on targeting aberrant Hh pathway and overcoming SMOi resistance of Hh-driven cancers.

CRISPR-Cas9 screening

CRISPR-Cas9 screening was applied as described previously (37). Briefly, mouse SHH-MB cells SmoWT was first transduced with lentiCas9-Blast (Addgene, 52962) at a low multiplicity of infection (MOI <0.7) to generate cell lines with stably expressed Cas9 components. Then, a pooled lentiviral sgRNA library targeting epigenetic/transcriptional factor genes was transduced in Cas9-expressing cells at a low MOI (<0.3). Our sgRNA library contains two sublibraries, and 1# sublibrary contains 5,200 sgRNAs [5,100 sgRNAs targeting 850 epigenetic/transcriptional factor genes and 100 nontargeting control (NTC) sgRNAs], and 2# sublibrary contains 950 sgRNAs (930 sgRNAs targeting 186 epigenetic/transcriptional factor genes and 20 NTC sgRNAs; detailed genes and sequences information of these sgRNA libraries are shown in Supplementary Table S1). Two days after sgRNA library transduction, cells were selected with 1.0 μg/mL puromycin for 5 days. Considering only 10% initial cell could survive after allograft in vivo and to ensure 500× coverage of the sgRNA library, 5 × 106 cells were subcutaneously injected into each side of dorsal flanks of 6 mice for 1# sublibrary, and 2.5 × 106 cells were subcutaneously injected into 2 mice for 2# sublibrary. Moreover, transduced cells were cultured in vitro simultaneously. Four weeks after inoculation, tumor cells were isolated from subcutaneous tumor and in vitro cultured cells were collected at the same time. At least 5 × 106 and 1 × 106 cells were collected for genomic DNA extraction to ensure over 500× coverage of two sublibrary, respectively. The sgRNA sequences were amplified using 2× KAPA HiFi HotStart ReadyMix (Roche, KK2602) and then PCR products were subjected to next-generation sequencing by Novogene Technology. The sgRNA read counts were acquired by Python2.7, pseudocount of 1 was added to the read count of each sgRNA and then normalized by the total number of read counts. To obtain the sgRNA fold change, the value of normalized sgRNA count was divided by the control and take the base 2 logarithm. Then, score, rank and P value of every gene in the library were obtained by RIGER algorithm with GENE-E software (37). The screening results of candidate genes are shown in Supplementary Table S2.

Cell culture

Mouse SHH-MB cell lines SmoWT, SmoWT-sgSufu, SmoD477G, human Hh-driven ATRT cell line ATRT-03 and primary mouse cerebellum-derived neural progenitor cells (cNPC) were cultured as described previously (22).

Mycn-MB cells were derived from Mycn-transformed p53−/− granule neuron precursors (GNP; ref. 22) and cultured in NeuroCult NS-A Proliferation Kit (Mouse; Stem Cell Technology, 05702) supplemented with murine EGF (20 ng/mL; PeproTech, 315-09), murine-FGF (20 ng/mL; PeproTech, 450-33), and heparin (10 ng/mL; Stem Cell Technology, 07980). GNPs from p53−/− C57 mice were infected with Mycn-overexpressing lentivirus and then injected into the cerebellum of nude mice. When those mice developed tumor-associated neurologic symptoms, their cerebellums were harvested and subjected to tumor cell isolation.

Establishment of patient-derived xenograft models

Two SHH-MB patient-derived xenograft (PDX) models used in this study (SHH-MB_PDX1 and SHH-MB_PDX2) were established by Shanghai LIDE Biotech Co., Ltd. Human SHH-MB tumor tissue samples for establishment of PDX models were obtained from children patients with written informed consent signed by their parents at Xin Hua Hospital affiliated to Shanghai Jiao Tong University School of Medicine (Shanghai, P.R. China). All procedures were approved by the Medical Experimental Animal Administrative Committee of Shanghai. Briefly, the patient tumor tissue was injected into female CB17-SCID mice. When grew steadily, PDX tissue was then engrafted into BALB/c Nude mice for the subsequent experiment. SHH-MB PDX-derived cells (named PDC in many articles) were established from freshly isolated subcutaneous tumors. In brief, subcutaneous tumors were cut into slurry and filtered through 100 μm cell strainer (Falcon, 352360). Then ACK Lysing Buffer (Thermo Fisher Scientific, A1049201) and HISTOPAQUE-1077 (Sigma, 10771-100ML) were applied to remove hemocyte and dead cells respectively from the isolated cells. The purified cells were cultured as neurospheres in Tumor Stem Media consisting of DMEM/F12 (Gibco, 11330032), Neurobasal(-A) (Gibco, 10888022), plus B27(-A) (Gibco, 12587010), human LIF (10 ng/mL; PeproTech, 300-05), human FGF-basic (20 ng/mL; PeproTech, 100-18B), human EGF (20 ng/mL; PeproTech, AF-100-15), 0.2% Heparin Solution (10 ng/mL; Stem Cell Technology, 07980), and 1× Antibiotic-Antimycotic (Gibco, 15240062). Short tandem repeat sequencing was applied in cell line retrieval of these two human SHH-MB PDX models, and no matching cell lines and no cross contamination by other cell lines were identified by Shanghai BIOWING Biotech Co., Ltd (Supplementary Table S3).

Virus production and infection

Lentiviral short hairpin (shRNA) or sgRNA plasmids were constructed by cloning target oligonucleotides into pLKO.1 (Addgene plasmid, 10878) or lentiCRISPR v2 plasmids (Addgene plasmid, 52961) respectively. Lentivirus was generated by transfecting objective plasmids with packaging vectors (psPAX and pMD2.G) in HEK293T cells. Then, virus was concentrated with PEG6000 (Sigma) for subsequent infection. Forty-eight hours after infection, positive cells were selected with puromycin (1 μg/mL; YEASEN, 60210ES25). The target oligonucleotides are listed in Supplementary Table S4.

In the rescue experiments, wildtype or wobble mutant CDS of Ssrp1 gene were cloned into vector plenti-EF1α-MCS-IRES-Neo-WPRE (EcoRI, BamHI) by homologous recombination. As shown in Supplementary Fig. S4A, four primer pairs containing mutation sites were used to amplify mutant Ssrp1 CDS, and four mutant amplicons were homologously recombinated into linearized vector. Lentivirus was generated and infected into mouse SHH-MB (SmoWT), and G418 was applied to screen stably Ssrp1-overexpressing SHH-MB cells. Then, the shSsrp1 lentivirus was infected into normal or mutant cells following puromycin selection. The primers for plasmids construction are listed in Supplementary Table S5.

Western blot assay

Protein samples were separated by SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Millipore). The membranes were blocked in 5% fat-free milk (BD Biosciences, 232100) and then incubated with primary antibodies against SSRP1 (1:1,000; Cell Signaling Technology, 13421S), SUPT16H (1:1,000; Cell Signaling Technology, 12191), GLI1 (1:2,000; Proteintech, 66905-1), MYCN (1:1,000; Cell Signaling Technology, 9405S), MYC (1:1,000; Cell Signaling Technology, 13987), SUFU (1:1,000; Cell Signaling Technology, 2522), β-Tubulin (1:5,000; Abcam, ab6046) or H3 (1:5,000; Abcam, ab1791). Secondary antibodies were horseradish peroxidase–conjugate goat anti-rabbit/mouse IgG (0.2 μg/mL; Pierce, 31460 or 31430). The chemical fluorescence images of proteins were visualized by Luminescent Image Analyzer (Fujifilm, LAS-4000) after incubation with enhanced chemiluminescence reagents (Millipore, WBKLS0500).

Quantitative real-time PCR assay

Total RNA was extracted with TRIzol Reagent (Thermo Fisher Scientific, 15596-026) according to the manufacturers' protocols. RNA was reversely transcripted into cDNA with High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, 4368813). Quantitative PCR was applied on Applied Biosystems QuantStudio Real-Time PCR System with SYBR Green Master (ROX; Roche, 24759100). All assays were performed in triplicate, and mRNA levels were normalized to GAPDH. qRT-PCR primer sequences are listed in Supplementary Table S5.

Cell viability and FACS assay

Cells were plated in 96-well plate in triplicate and treated with CBL0137 (MedChem Express, HY-18935A), GDC0449 (SelleckChem, S1082), THZ1 (MedChem Express, HY-80013) or JQ1 (MedChem Express, HY-13030) at indicated concentrations for 72 hours, then cell viability assays were performed using CellTiter-Glo Luminescent Cell Viability Assay (Promega, G7573) according to the manufacturer's protocol. For drugs combinatory assays, the synergistic effects were presented with combination index (CI) or BLISS analysis (38). CI value was calculated using CalcuSyn software (Biosoft). CI value < 1.0 indicates synergistic effect. BLISS analysis performed by R studio using R package. BLISS score > 0 indicates synergistic effect.

The proliferation, apoptosis, and cell cycle were analyzed by FACS assay with Click-iT Plus EdU Flow Cytometry Assay Kit (Invitrogen, C10632), FITC Annexin V Apoptosis Detection Kit (BD Biosciences, 556547) and Cell Cycle Staining Kit (Metasciences, CCS012) according to the manufacturer's instructions, respectively. FACS were performed by BD Fortessa FACS instrument (BD Biosciences) and the data were analyzed with FlowJo software (FlowJo).

Animal experiments

BALB/c nude female mice (6–8 weeks) were purchased from Shanghai Experimental Animal Center of the Chinese Academy of Sciences (Beijing, P.R. China). All animal experiments were reviewed and approved by the Medical Experimental Animal Administrative Committee of Shanghai. For subcutaneous allograft or xenograft model, 1.5 × 106 mouse medulloblastoma cells or small pieces SHH-MB PDX tissue were subcutaneously injected on each side of dorsal flanks of mice. For intracranial allograft or xenograft model, 1 × 105 mouse SHH-MB cells (SmoWT) or 7.5 × 105 human SHH-MB cells (SHH-MB_PDX1) were injected stereotactically into the cerebellum (Posterior fontanelle: AP, −2.0 mm; ML, −1.5 mm; DV, −2.65 mm) with Hamilton microinjector in 3 μL total volume at rate of 0.3 μL/minute, and the microinjector was left in place for an additional 10 minutes before being withdrawn at a rate of 0.5 mm/minute.

For CBL0137 treatment in vivo, stock solution was diluted with 10% cyclodextrin (Sigma, H107) and administrated at 60 mg/kg-BW (body-weight) every 4 days by intravenous injection or 30 mg/kg-BW every 2 days in oral gavage. For GDC0449 treatment in vivo, stock solution was diluted with 0.5% (m/V) methyl cellulose solution containing 0.15% Tween-80 (V/V) and administrated at 100 mg/kg-BW twice daily by oral gavage.

For subcutaneous allograft or xenograft model, when the tumor volume approached to 100 mm3, mice were divided randomly into three groups and treated with CBL0137, GDC0449, or vehicle control, respectively. After five times of CBL0137 treatment, then animals were euthanized. For intracranial allograft model, 3 days after intracranial transplantation, CBL0137 was administrated by intravenous injection or oral gavage. For mouse SHH-MB intracranial allograft model (SmoWT), CBL0137 or vehicle treatment was terminated when the first animal died. For human SHH-MB intracranial xenograft model (SHH-MB_PDX1), CBL0137 or vehicle treatment was terminated when the last animal died.

Hematoxylin and eosin and IHC staining

Hematoxylin and eosin (H&E) and IHC staining were performed by Shanghai Runnerbio Technology CO., Ltd. Primary antibodies against Ki67 (Abcam, ab15580), Cleaved caspase-3 (Cell Signaling Technology, 9661) were used for IHC staining. Immunoreactive cells were counted as the average of total positive cells in the five fields of view and expressed as cells/mm2 (39).

RNA sequencing data analysis

Trim galore was used to automatically detect and trim adapters. Reads were mapped to mm10 reference genome using Hisat2. Read count was generated using HTSeq (version 0.11.1). Differentially expressed genes were calculated using DESeq2 (version 1.26.0). Absolute value of fold change (FC) > 1.5 and FDR < 0.05 was used as cut-off value to select significant target genes. FPKM (fragments per kilobase million) was calculated from the number of reads that mapped to each particular gene sequence and the gene length and the sequencing depth were taken into account. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) functional annotation were analyzed with DAVID bioinformatics resources (https://david.ncifcrf.gov/). Top 10 hits were shown in the results for GO and KEGG analysis. Raw data are accessible at the NCBI Gene Expression Omnibus (GEO) accession code GSE171685 and GSE171689.

Gene set enrichment analysis

Gene set enrichment analysis (GSEA) was performed with GSEA 3.0 software according to the instructions on the website (http://www.broadinstitute.org/gsea/index.jsp.). The set of genes used in GSEA, named SmoWT_GDC_Down, were significantly downregulated genes (log2 FC < −0.6, FDR < 0.05) of SmoWT cells after GDC0449 treatment compared with DMSO treated counterpart for 24 hours.

ChIP-qPCR analysis

Chromatin immunoprecipitation (ChIP) coupled with quantitative PCR (ChIP-qPCR) was performed as described previously (20). Briefly, 1–2 × 107 cells were cross-linked with 1% formaldehyde for 10 minutes at room temperature before adding glycine to stop the cross-linking. Then, the harvested cells were digested with Micrococcal Nuclease (New England Biolabs, M0247S) and sonicated for ChIP experiment. Then 5–10 μg chromatin was immunoprecipitated with 5–10 μL of SSRP1 antibody (Cell Signaling Technology, 13421S) and 50 μL Pierce ChIP-grade Protein A/G Magnetic Beads (Thermo Fisher Scientific, 26162). qRT-PCR was performed using SYBR Green Master (ROX; Roche, 24759100). The size of ChIP-qPCR amplicon was 250 bp, and primers were designed between 2,000 bp upstream and 500 bp downstream of transcriptional start site (TSS). The primer sequences are listed in Supplementary Table S5.

Statistical analysis

GraphPad Prism 6.0 software was applied for the statistical analysis. Comparisons between two groups were analyzed by unpaired two-tailed Student t test. Two-way ANOVA was performed to compare in vivo tumor growth curves. The statistical significance of Kaplan–Meier survival curves was determined by log-rank (Mantel–Cox) test.

Identification of Ssrp1 as a potential therapeutic target for treating SHH-MB and overcoming SMOi resistance

To identify novel epigenetic/transcriptional modulators that may serve as potential therapeutic targets for inhibiting aberrant Hh pathway and overcoming SMOi resistance, we firstly analyzed multiple public human (40–42) and mouse (43) SHH-MB datasets containing normal cerebellum control samples to identify epigenetic/transcriptional genes that were significantly upregulated in both human and mouse SHH-MB (log2 FC > 1, FDR < 0.05). In addition, two SMOi-treated mouse SHH-MB datasets (44) were analyzed to identify SMOi-responsive epigenetic/transcriptional genes that were initially downregulated by SMOi treatment (log2 FC < −0.6, FDR < 0.05) but later restored expression levels after gaining SMOi resistance (log2 FC > 0.6, FDR < 0.05). Moreover, we performed in vitro and in vivo CRISPR-Cas9 screening of sgRNA library targeting 1,036 epigenetic/transcriptional modulator coding genes (five or six independent sgRNAs against each gene; Supplementary Table S1) in SmoWT, a SHH-MB cell line derived from spontaneous medulloblastoma of Ptch1+/−, Trp53−/− mice (45). On the basis of that sgRNAs targeting cancer-dependent genes would be significantly depleted after the screening, RIGER algorithm (P < 0.05) was used to identify the essential epigenetic/transcriptional modulator coding genes for both in vitro and in vivo growth (Fig. 1A; Supplementary Fig. S1A). Through combining all above datasets analyses and CRISPR-Cas9 screening results, 14 epigenetic/transcriptional genes (Brca1, Cenpa, Lsm2, Lsm3, Mcm4, Mcm6, Mcm7, Rad51, Rpa3, Smc2, Smc4, Snrpd1, Snrpf, Ssrp1), which were mainly involved in biological processes of cell cycle, DNA replication and DNA repair, were identified as aberrant Hh-induced essential regulators and potential therapeutic targets for treating SHH-MB and overcoming SMOi resistance (Fig. 1A; Supplementary Fig. S1B; Supplementary Fig. S2A–S2E; Supplementary Table S2). Among these genes, Ssrp1 was chosen for further investigation for it was the only one with a targeted inhibitor drug (CBL0137) that had entered human clinical trials for treating cancer (ClinicalTrials.gov Identifier: NCT02931110, NCT03727789, and NCT01905228).

Figure 1.

Ssrp1 is a potential therapeutic target gene for inhibiting aberrant Hh pathway and overcoming SMOi resistance. A, The streamlined workflow of determining potential therapeutic targets in SHH-MB through epigenetic/transcriptional-targeted CRISPR-Cas9 screening and tumor dataset analyses. B, Expression levels of Ssrp1 in mouse SHH-MB from datasets GSE11859 and GSE104633. C, Expression levels of Ssrp1 in mouse SHH-MB treated with SMOi (LDE) from datasets GSE19657 and GSE22005. Vehicle and LDE_48h represent SMOi (LDE)-sensitive group treated with vehicle or LDE for 48 hours, respectively; LDE_RES represents acquired SMOi (LDE)-resistant group. D and E, qRT-PCR and Western blot analysis of Ssrp1, Gli1, and Gli2 expression in normal cerebellum tissue at different postnatal age (C57-P7d∼C57-P11W) and in mouse SHH-MB cells (SmoWT and SmoD477G). P7d, 7 days postnatally; P8/9/11W, 8/9/11 weeks postnatally. F, qRT-PCR analysis of Ssrp1 and Gli1 expression in SmoWT cells treated with SMOi (GDC0449, 0.1 μmol/L) for 24 hours. Gand H, The sgRNA levels of Ssrp1 (green) and NTCs (red) in SmoWT screening in vitro (G) and in vivo (H). Rank, score, and P value in G or H were gained through RIGER algorithm. I and J, qRT-PCR and Western blot analysis of Ssrp1 expression in mouse SHH-MB cells expressing different shSsrp1 shRNAs. K, Cell viabilities were measured at day 0/2/4 in mouse SHH-MB lines expressing different shSsrp1 shRNAs and normalized to day 0 value. All qRT-PCR and cell viability assays were performed in triplicate and the data are presented as the means ± SD. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test.

Figure 1.

Ssrp1 is a potential therapeutic target gene for inhibiting aberrant Hh pathway and overcoming SMOi resistance. A, The streamlined workflow of determining potential therapeutic targets in SHH-MB through epigenetic/transcriptional-targeted CRISPR-Cas9 screening and tumor dataset analyses. B, Expression levels of Ssrp1 in mouse SHH-MB from datasets GSE11859 and GSE104633. C, Expression levels of Ssrp1 in mouse SHH-MB treated with SMOi (LDE) from datasets GSE19657 and GSE22005. Vehicle and LDE_48h represent SMOi (LDE)-sensitive group treated with vehicle or LDE for 48 hours, respectively; LDE_RES represents acquired SMOi (LDE)-resistant group. D and E, qRT-PCR and Western blot analysis of Ssrp1, Gli1, and Gli2 expression in normal cerebellum tissue at different postnatal age (C57-P7d∼C57-P11W) and in mouse SHH-MB cells (SmoWT and SmoD477G). P7d, 7 days postnatally; P8/9/11W, 8/9/11 weeks postnatally. F, qRT-PCR analysis of Ssrp1 and Gli1 expression in SmoWT cells treated with SMOi (GDC0449, 0.1 μmol/L) for 24 hours. Gand H, The sgRNA levels of Ssrp1 (green) and NTCs (red) in SmoWT screening in vitro (G) and in vivo (H). Rank, score, and P value in G or H were gained through RIGER algorithm. I and J, qRT-PCR and Western blot analysis of Ssrp1 expression in mouse SHH-MB cells expressing different shSsrp1 shRNAs. K, Cell viabilities were measured at day 0/2/4 in mouse SHH-MB lines expressing different shSsrp1 shRNAs and normalized to day 0 value. All qRT-PCR and cell viability assays were performed in triplicate and the data are presented as the means ± SD. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test.

Close modal

SSRP1 and SUPT16H (Ssrp1 and Supt16 in mouse) comprise the two-subunit FACT complex, which exhibits chromatin remodeling activity and has been shown to play important roles in many cancers (24, 46). Therefore, we compared datasets analyses and CRISPR-Cas9 screening results of Ssrp1 and Supt16 in parallel. In dataset analyses, only Ssrp1 but not Supt16 was consistently upregulated in all subtypes of mouse SHH-MB compared with normal cerebellum (Fig. 1B; Supplementary Fig. S3A). However, both genes were downregulated by SMOi and restored expression levels after gaining SMOi resistance (Fig. 1C; Supplementary Fig. S3B). These dataset analysis results were further validated by qRT-PCR and Western blot experiments (Fig. 1D and E; Supplementary Fig. S3E–S3G). As shown in Fig. 1D, similar expression patterns of Ssrp1, Gli1, and Gli2 were observed when comparing SmoWT cells with normal cerebellum of 7 days and 8/9/11 weeks postnatally. SMOi-resistant SmoD477G cells, which obtained SMOi resistance by D477G point mutation of Smo in SmoWT cells (45), exhibited almost the same Ssrp1 and Gli1 and Gli2 expression levels as their parental cells (Fig. 1D and E). In contrast, Supt16 mRNA and protein levels were similarly high in mouse SHH-MB cells and normal cerebellum (Supplementary Fig. S3E and S3F). Moreover, even though both Ssrp1 and Supt16 were significantly suppressed by SMOi GDC0449 in SmoWT cells as Gli1, the inhibitory effect was stronger on Ssrp1 than Supt16 (Fig. 1F; Supplementary Fig. S3G).

In CRISPR-Cas9 screening analyses, both Ssrp1 and Supt16 were identified as essential genes of SmoWT as their targeted sgRNAs were dramatically depleted compared with NTCs in vitro and in vivo (Fig. 1G and H; Supplementary Fig. S3C and S3D). RNAi was utilized to further confirm their dependencies in mouse SHH-MB cells. Our data showed that growth of SMOi-sensitive SmoWT or SMOi-resistant SmoD477G cells in vitro were dramatically disrupted when Ssrp1 or Supt16 was knocked down by three independent shRNAs individually (Fig. 1I–K; Supplementary Fig. S3H–S3J). Knockdown of Ssrp1 reduced cell proliferation, induced cell apoptosis and cell-cycle arrest at G2–M phase in SmoWT cells (Supplementary Fig S3K–S3M). To confirm the on-target inhibition by Ssrp1 knockdown, we constructed wobble mutant plasmids encoding Ssrp1 CDS resistant to all shRNA clones (Ssrp1-MUT; Supplementary Fig S4A–S4C, S4E and S4F) and proved that the inhibitory effects of shSsrp1 on cell growth could be substantially rescued in SmoWT cells stably expressing Ssrp1-MUT (Supplementary Fig. S4D and S4G). Taken together, these results demonstrated that Ssrp1 was a Hh-induced oncogene and potential therapeutic target for treating mouse SHH-MB and overcoming SMOi resistance.

CBL0137 suppresses growth of mouse SHH-MB in vitro and in vivo

As mentioned above, FACT-targeted small-molecule drug CBL0137 has already entered human clinical trials for cancer therapy. Therefore, we next tested the in vitro efficacy of CBL0137 in treating multiple mouse SHH-MB lines, including SMOi-sensitive SmoWT as well as SMOi-resistant SmoD477G, SmoWT-sgSufu, and Mycn-MB lines (Supplementary Fig. S5A). SmoWT-sgSufu line obtained SMOi resistance by knockout of Sufu with CRISPR-Cas9 approach in SmoWT cells (Supplementary Fig. S5B). Mycn-MB line, which represented MYCN-amplified and TP53-deficient human SHH-MB with very poor prognosis and primary SMOi resistance (13), was established by injecting primary GNP cells of Trp53−/− mice with stably expression of Mycn into cerebellum of nude mice (Supplementary Fig. S5B). As expected, both SmoWT-sgSufu and Mycn-MB cells exhibited resistance to GDC0449 but remained sensitive to BET inhibitor JQ1 and CDK7 inhibitor THZ1 (Supplementary Fig. S5A). As shown in Fig. 2A, chromatin fraction of Ssrp1 protein increased while soluble fraction reduced after CBL0137 treatment of SmoWT cells, confirming FACT complex was trapped in chromatin by CBL0137 as reported previously (34). We then treated the four mouse SHH-MB lines with increasing concentrations of CBL0137 and found that it inhibited their cell viability in dose-dependent manner with IC50 at approximately 270–360 nmol/L (Fig. 2B). Normal primary neural progenitor cells derived from mouse cerebellum (cNPC) were tested in parallel and the IC50 was approximately 640 nmol/L, indicating a selective inhibition of CBL0137 on mouse SHH-MB cells in vitro (Fig. 2B). Moreover, our data showed that the growth inhibition by CBL0137 was as effective in SMOi-resistant SmoD477G, SmoWT-sgSufu, Mycn-MB cells as in SMOi-sensitive SmoWT cells, where it could significantly inhibit cell proliferation and induce apoptosis and cell-cycle arrest at G2–M phase (Fig. 2C–F; Supplementary Fig. S5C).

Figure 2.

CBL0137 inhibits growth of SMOi-sensitive or SMOi-resistant mouse SHH-MB cells in vitro. A, Western blot of Ssrp1 expression in soluble and chromatin fraction from protein of mouse SHH-MB cells (SmoWT) treated with CBL0137 (1 μmol/L) for 2 hours. β-Tubulin and H3 are shown as loading controls for soluble and chromatin extracts, respectively. B, Dose–response curves of mouse SHH-MB cell lines, mouse cerebellum-derived primary neural progenitor cells (cNPC) treated with different dose of CBL0137 for 72 hours. IC50 of each line was listed. C, Mouse SHH-MB cell lines were treated with DMSO or CBL0137 as indicated. Cell viabilities were then measured at day 0/2/4 and normalized to day 0 value. D–F, Cell proliferation (D), apoptosis (E), and cell cycle (F) analyses of mouse SHH-MB cell lines treated with DMSO or CBL0137 (1 μmol/L) by FACS assay. All the cell viability assays were performed in triplicate and the data are presented as the means ± SD.

Figure 2.

CBL0137 inhibits growth of SMOi-sensitive or SMOi-resistant mouse SHH-MB cells in vitro. A, Western blot of Ssrp1 expression in soluble and chromatin fraction from protein of mouse SHH-MB cells (SmoWT) treated with CBL0137 (1 μmol/L) for 2 hours. β-Tubulin and H3 are shown as loading controls for soluble and chromatin extracts, respectively. B, Dose–response curves of mouse SHH-MB cell lines, mouse cerebellum-derived primary neural progenitor cells (cNPC) treated with different dose of CBL0137 for 72 hours. IC50 of each line was listed. C, Mouse SHH-MB cell lines were treated with DMSO or CBL0137 as indicated. Cell viabilities were then measured at day 0/2/4 and normalized to day 0 value. D–F, Cell proliferation (D), apoptosis (E), and cell cycle (F) analyses of mouse SHH-MB cell lines treated with DMSO or CBL0137 (1 μmol/L) by FACS assay. All the cell viability assays were performed in triplicate and the data are presented as the means ± SD.

Close modal

To demonstrate the in vivo therapeutic potential of CBL0137, we examined its efficacy against SMOi-sensitive SmoWT and SMOi-resistant Mycn-MB allografts in nude mice. GDC0449 treatment was performed in parallel as control. As expected, the growth of SmoWT, but not Mycn-MB subcutaneous allografts was effectively suppressed by oral gavage of GDC0449 at 100 mg/kg-BW twice daily (Fig. 3A and B). In contrast, intravenous injection of CBL0137 at 60 mg/kg-BW every 4 days significantly suppressed growth of both mouse SHH-MB allograft models without obviously affecting body weight of treated mice (Fig. 3A and B; Supplementary Fig. S5D). CBL0137 treatment dramatically reduced the mitotic activity and induced tumor cell death as shown in H&E and IHC staining results (Ki67+ for proliferating cells, cleaved caspase-3/CC3+ for apoptotic cells; Fig. 3C and D). Moreover, transcript levels of Ki67 and anti-apoptotic gene Bcl2 (Supplementary Fig. S5F), as well as protein levels of Ssrp1 and Supt16 were significantly downregulated by CBL0137 (Supplementary Fig. S5G).

Figure 3.

CBL0137 suppresses growth of SMOi-sensitive or SMOi-resistant mouse SHH-MB allograft models in vivo. A and B, SmoWT or Mycn-MB cells were subcutaneously injected into ventral flanks of nude mice. Tumors were treated with GDC0449 (100 mg/kg-BW; twice daily by gavage), CBL0137 (60 mg/kg-BW; every 4 days by intravenous injection), or vehicle (cyclodextrin). Tumor volume was measured by caliper after drug treatment and the data were presented as the means ± SEM (right). The image of three representing tumors in each group was shown at day of sacrifice (left). C, The representative H&E and IHC (Ki67+ for proliferation and cleaved caspase-3/CC3+ for apoptosis) staining images of tumor allografts of mouse SHH-MB treated with CBL0137 or vehicle. D, The immunoreactive positive cells were counted in five views and are shown as cells/mm2 for each condition. The data are presented as the means ± SD. ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. E, Mice survival was monitored and Kaplan–Meier survival curve and median survival are shown for nude mice with SmoWT intracranial allografts treated with CBL0137 or vehicle control. CBL0137 was administrated intravenously at 60 mg/kg-BW every 4 days or 30 mg/kg-BW by gavage (gav) every 2 days. Cyclodextrin was injected intravenously every 4 days as vehicle control. P value for tumor volume analysis or survival analysis was determined by two-way ANOVA or log-rank test (Mantel–Cox test).

Figure 3.

CBL0137 suppresses growth of SMOi-sensitive or SMOi-resistant mouse SHH-MB allograft models in vivo. A and B, SmoWT or Mycn-MB cells were subcutaneously injected into ventral flanks of nude mice. Tumors were treated with GDC0449 (100 mg/kg-BW; twice daily by gavage), CBL0137 (60 mg/kg-BW; every 4 days by intravenous injection), or vehicle (cyclodextrin). Tumor volume was measured by caliper after drug treatment and the data were presented as the means ± SEM (right). The image of three representing tumors in each group was shown at day of sacrifice (left). C, The representative H&E and IHC (Ki67+ for proliferation and cleaved caspase-3/CC3+ for apoptosis) staining images of tumor allografts of mouse SHH-MB treated with CBL0137 or vehicle. D, The immunoreactive positive cells were counted in five views and are shown as cells/mm2 for each condition. The data are presented as the means ± SD. ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. E, Mice survival was monitored and Kaplan–Meier survival curve and median survival are shown for nude mice with SmoWT intracranial allografts treated with CBL0137 or vehicle control. CBL0137 was administrated intravenously at 60 mg/kg-BW every 4 days or 30 mg/kg-BW by gavage (gav) every 2 days. Cyclodextrin was injected intravenously every 4 days as vehicle control. P value for tumor volume analysis or survival analysis was determined by two-way ANOVA or log-rank test (Mantel–Cox test).

Close modal

Because CBL0137 has been reported to be able to breach blood–brain barrier (32), we further tested its efficacy in treating orthotopic allograft model of mouse SHH-MB. SmoWT cells were orthotopically transplanted into cerebellum of nude mice and CBL0137 was administrated by oral gavage at 30 mg/kg-BW every 2 days or intravenous injection at 60 mg/kg-BW every 4 days (Fig. 3E). The results showed that both administration approaches could significantly improve overall survival of intracranial allografted mice without significantly altering their body weight, demonstrating the therapeutic potential of CBL0137 in treating intracranial Hh-driven cancers (Fig. 3E; Supplementary Fig. S5E).

CBL0137 suppresses aberrant Hh pathway through disrupting the binding of FACT to Gli1/2 promoters and blocking their transcription in mouse SHH-MB cells

To investigate whether CBL0137 could target the aberrant Hh pathway, we firstly performed RNA sequencing (RNA-seq) analyses of SmoWT cells treated by CBL0137 or GDC0449 (Fig. 4A; Supplementary Fig. S6A–C). Our results showed that there was a substantial overlap of significantly downregulated genes between GDC0449- and CBL0137-treated SmoWT cells, including Gli1 and Gli2, the core transcription factors of Hh pathway (Fig. 4B). We also found that there was a strong correlation between the inhibitory effects (calculated by log2 FC of drug vs. DMSO) of GDC0449 and CBL0137 on their commonly downregulated genes whereas no such correlations were detected between the inhibitory effects and the transcript levels of their target genes (Fig. 4C; Supplementary Fig. S6D). Moreover, GSEA showed that gene set “SmoWT_GDC_Down,” which contains GDC0449-downregulated genes of SmoWT cells, was significantly inhibited by CBL0137 in SmoWT cells (Fig. 4D).

Figure 4.

CBL0137 disrupts the binding of FACT to Gli1/2 promoters and blocks their transcription in mouse SHH-MB cells. Heatmap of gene expression values of active transcripts (mean FPKM ≥ 1 in either condition) in SmoWT treated with GDC0449 (0.1 μmol/L; right) or CBL0137 (1 μmol/L; left) versus DMSO for 24 hours. Rows show Z scores calculated for each condition. The rank positions of Gli1 and Gli2 are shown. B, Venn diagram of significantly downregulated genes (log2 FC < −0.6, FDR < 0.05) in GDC0449 (GDC_Down)- or CBL0137 (CBL_Down)-treated SmoWT cells. C, Correlation analysis of gene expression changes of the commonly downregulated genes (log2 FC < −0.6, FDR < 0.05) by GDC0449 or CBL0137 treatment. R and P values are shown. D, GSEA showing enrichment of Hh target genes in DMSO-treated SmoWT cells versus CBL0137-treated ones. Gene set “SmoWT_GDC_Down” is composed of significantly downregulated genes (log2 FC < −0.6, FDR < 0.05) in GDC0449-treated SmoWT cells. E, qRT-PCR analysis of Hh-targeted genes Gli1, Gli2, Hhip, and Gpr153, in mouse SHH-MB lines treated with CBL0137 (1 μmol/L), GDC0449 (0.1 μmol/L), or DMSO for 24 hours. F, Western blot of Gli1 in mouse SHH-MB lines treated with CBL0137 (1 μmol/L) or DMSO for 24 hours. G, qRT-PCR analysis of Gli1 and Gli2 in SmoWT subcutaneous allografts treated with CBL0137 or vehicle. H, Illustration of ChIP-qPCR primer design for Gli1 and Gli2. I and J, ChIP-qPCR analyses of Ssrp1 binding at promoters of Gli1 and Gli2 in SMOi-sensitive or -resistant mouse SHH-MB cells treated overnight with CBL0137 (1 μmol/L) or DMSO. All qRT-PCR and ChIP-qPCR assays were performed in triplicate and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. ns, nonsignificant.

Figure 4.

CBL0137 disrupts the binding of FACT to Gli1/2 promoters and blocks their transcription in mouse SHH-MB cells. Heatmap of gene expression values of active transcripts (mean FPKM ≥ 1 in either condition) in SmoWT treated with GDC0449 (0.1 μmol/L; right) or CBL0137 (1 μmol/L; left) versus DMSO for 24 hours. Rows show Z scores calculated for each condition. The rank positions of Gli1 and Gli2 are shown. B, Venn diagram of significantly downregulated genes (log2 FC < −0.6, FDR < 0.05) in GDC0449 (GDC_Down)- or CBL0137 (CBL_Down)-treated SmoWT cells. C, Correlation analysis of gene expression changes of the commonly downregulated genes (log2 FC < −0.6, FDR < 0.05) by GDC0449 or CBL0137 treatment. R and P values are shown. D, GSEA showing enrichment of Hh target genes in DMSO-treated SmoWT cells versus CBL0137-treated ones. Gene set “SmoWT_GDC_Down” is composed of significantly downregulated genes (log2 FC < −0.6, FDR < 0.05) in GDC0449-treated SmoWT cells. E, qRT-PCR analysis of Hh-targeted genes Gli1, Gli2, Hhip, and Gpr153, in mouse SHH-MB lines treated with CBL0137 (1 μmol/L), GDC0449 (0.1 μmol/L), or DMSO for 24 hours. F, Western blot of Gli1 in mouse SHH-MB lines treated with CBL0137 (1 μmol/L) or DMSO for 24 hours. G, qRT-PCR analysis of Gli1 and Gli2 in SmoWT subcutaneous allografts treated with CBL0137 or vehicle. H, Illustration of ChIP-qPCR primer design for Gli1 and Gli2. I and J, ChIP-qPCR analyses of Ssrp1 binding at promoters of Gli1 and Gli2 in SMOi-sensitive or -resistant mouse SHH-MB cells treated overnight with CBL0137 (1 μmol/L) or DMSO. All qRT-PCR and ChIP-qPCR assays were performed in triplicate and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. ns, nonsignificant.

Close modal

On the basis of above bioinformatic analyses, Gli1, Gli2, Hhip, and Gpr153, which were previously reported as Hh target genes in medulloblastoma (22, 47, 48), were selected for further validation. As expected, these genes were sensitive to GDC0449 in SMOi-sensitive SmoWT cells but not in SMOi-resistant SmoD477G or SmoWT-sgSufu cells. In contrast, all of them were significantly downregulated by CBL0137 (Fig. 4E). Western blot analyses further confirmed the downregulation of Gli1 protein by CBL0137 in those lines (Fig. 4F). Moreover, qRT-PCR of vehicle- or drug-treated subcutaneous allografts of SmoWT demonstrated CBL0137 and GDC0449 also exhibited similar inhibitory effects on transcription of Gli1 and Gli2 in vivo (Fig. 4G). Next, we performed ChIP-qPCR analyses with anti-Ssrp1 antibody to further study how CBL0137 inhibited transcription of Gli1 and Gli2. Ten sets of ChIP-qPCR primers with amplicons of approximately 250 bp were designed for each gene, covering their promoter regions between 2,000 bp upstream and 500 bp downstream the TSS (Fig. 4H). Our results showed that Ssrp1 bound significantly to the promoter regions of Gli1 and Gli2, whereas CBL0137 treatment resulted in substantial depletion of Ssrp1 from these regions (Fig. 4I and J). Together, these data demonstrated that CBL0137 could suppress transcriptional output of aberrant Hh pathway in both SMOi-sensitive and SMOi-resistant mouse SHH-MB cells, and it exhibited anti-Hh activity through disrupting the binding of FACT to Gli1/2 promoters and blocking their transcription.

CBL0137 suppresses growth of multiple Hh-driven human tumor models in vitro and in vivo

To explore the therapeutic potential of FACT inhibition in Hh-driven human cancers, we first performed public human tumor dataset analyses of SHH-MB and SHH-ATRT (49, 50). We found that only SSRP1 but not SUPT16H was significantly upregulated in SHH-MB compared with normal brain across multiple analyzed tumor datasets (Fig. 5A; Supplementary Fig. S7A and C). Higher expression of SSRP1, but not SUPT16H, was significantly associated with worse prognosis of SHH-MB (Fig. 5B; Supplementary Fig. S7B). In ATRT tumor dataset analysis, SHH-ATRT, which was characterized by supervised clustering with previously reported marker genes for ATRT subtyping (SHH-subtype: GLI1/2 and MYCN, MYC-subtype: MYC, TYR-subtype: TYR; ref. 50), exhibited significant upregulation of both SSRP1 and SUPT16H compared with normal brain (Fig. 5C; Supplementary Fig. S7D). Notably, SSRP1 and SUPT16H exhibited similar aberrant expression patterns across various medulloblastoma and ATRT subtypes, suggesting a general oncogenic role of FACT complex in medulloblastoma and ATRT (Supplementary Fig. S7A; Supplementary Fig. S7C–S7E).

Figure 5.

CBL0137 inhibits growth of human Hh-driven tumor models in vitro and in vivo. A, Expression levels of SSRP1 in human SHH-MB from GEO datasets (Pomeroy dataset contains 11 normal samples and 52 SHH-MB samples; U133p2_SHH69 contains 59 samples from Pfister-223 dataset and 10 samples from dataset GSE37418; CB9 contains nine normal samples from dataset GSE7307). B, Kaplan–Meier analysis of overall survival of patients with SHH-MB from dataset GSE85217 with high (n = 80) or low (n = 92) SSRP1 expression levels. C, ATRT subtypes were classified by cluster analysis with characteristic markers from GSE70678 using online software (https://software.broadinstitute.org/morpheus/; left). Expression levels of SSRP1 in three ATRT subtypes and normal cerebellum from GEO datasets (nine normal samples from GSE7307, 49 ATRT samples from GSE70678; right). D, Cell viability of Hh-driven human tumor cells (ATRT03, SHH-MB_PDX1, and SHH-MB_PDX2) treated with CBL0137, CDK7i (THZ1), BETi (JQ1), SMOi (GDC0449), or DMSO at indicated concentrations for 72 hours. The assay was performed in triplicate. E, Nude mice with flank xenografts of SHH-MB_PDX1 were intravenously injected with CBL0137 (60 mg/kg-BW) or vehicle (cyclodextrin) every 4 days, and tumor volume was measured by caliper after drug treatment (right). Left, image of harvested tumors. F, The representative H&E staining images of flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle. G, The representative IHC staining images of flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle (Ki67+ for proliferation and cleaved caspase-3/CC3+ for apoptosis; left). The numbers of Ki67+ or CC3+ cells in each group (counted in five views and expressed as cells/mm2) are shown on the right. H, Nude mice with intracranial xenografts of SHH-MB_PDX1 were intravenously injected with CBL0137 (60 mg/kg-BW) or vehicle (cyclodextrin) every 4 days, and their survival was monitored until all mice died. Kaplan–Meier survival curve and median survival are shown. P value for survival analysis was determined by log-rank test (Mantel–Cox test). The data are presented as the means ± SD in A, C, D, G, and means ± SEM in E. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test in A, C, D, and G. ns, nonsignificant.

Figure 5.

CBL0137 inhibits growth of human Hh-driven tumor models in vitro and in vivo. A, Expression levels of SSRP1 in human SHH-MB from GEO datasets (Pomeroy dataset contains 11 normal samples and 52 SHH-MB samples; U133p2_SHH69 contains 59 samples from Pfister-223 dataset and 10 samples from dataset GSE37418; CB9 contains nine normal samples from dataset GSE7307). B, Kaplan–Meier analysis of overall survival of patients with SHH-MB from dataset GSE85217 with high (n = 80) or low (n = 92) SSRP1 expression levels. C, ATRT subtypes were classified by cluster analysis with characteristic markers from GSE70678 using online software (https://software.broadinstitute.org/morpheus/; left). Expression levels of SSRP1 in three ATRT subtypes and normal cerebellum from GEO datasets (nine normal samples from GSE7307, 49 ATRT samples from GSE70678; right). D, Cell viability of Hh-driven human tumor cells (ATRT03, SHH-MB_PDX1, and SHH-MB_PDX2) treated with CBL0137, CDK7i (THZ1), BETi (JQ1), SMOi (GDC0449), or DMSO at indicated concentrations for 72 hours. The assay was performed in triplicate. E, Nude mice with flank xenografts of SHH-MB_PDX1 were intravenously injected with CBL0137 (60 mg/kg-BW) or vehicle (cyclodextrin) every 4 days, and tumor volume was measured by caliper after drug treatment (right). Left, image of harvested tumors. F, The representative H&E staining images of flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle. G, The representative IHC staining images of flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle (Ki67+ for proliferation and cleaved caspase-3/CC3+ for apoptosis; left). The numbers of Ki67+ or CC3+ cells in each group (counted in five views and expressed as cells/mm2) are shown on the right. H, Nude mice with intracranial xenografts of SHH-MB_PDX1 were intravenously injected with CBL0137 (60 mg/kg-BW) or vehicle (cyclodextrin) every 4 days, and their survival was monitored until all mice died. Kaplan–Meier survival curve and median survival are shown. P value for survival analysis was determined by log-rank test (Mantel–Cox test). The data are presented as the means ± SD in A, C, D, G, and means ± SEM in E. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test in A, C, D, and G. ns, nonsignificant.

Close modal

Next, we utilized multiple Hh-driven human tumor models to verify the therapeutic potential of CBL0137 in vitro and in vivo, including two patient-derived SHH-MB models (SHH-MB_PDX1 and SHH-MB_PDX2) and ATRT03, a previously reported patient-derived SHH-ATRT model (22). SHH-MB_PDX1 and SHH-MB_PDX2 were characterized to exhibit robust expression of Hh signature genes (GLI1/2 and MYCN) and represented two distinct types of primary SMOi resistance: amplification of GLI2 and MYCN for SHH-MB_PDX1 and loss of SUFU expression for SHH-MB_PDX2 (Supplementary Fig. S7F and S7G). Tumor cells derived from above-mentioned three Hh-driven human tumor models were subjected to in vitro growth inhibition test and the results showed that all of them were resistant to GDC0449 but remained responsive to CBL0137, THZ1, and JQ1 (Fig. 5D). Moreover, we performed in vivo CBL0137 therapy against subcutaneous xenografting model of SHH-MB_PDX1. Small pieces of tumor tissues from passaged xenografts were injected into ventral flanks of nude mice and treated with CBL0137 by intravenous injection at 60 mg/kg-Western blot every 4 days. As shown in Fig. 5E and Supplementary Fig. S7H, CBL0137 treatment dramatically inhibited subcutaneous tumor growth without obviously affecting mice body weight. As in mouse SHH-MB allograft models, H&E staining, IHC staining (KI67+ and CC3+) and qRT-PCR measurement (KI67 and BCL2) revealed that CBL0137 reduced proliferation and induced apoptosis of SHH-MB_PDX1 in vivo (Fig. 5F and G; Supplementary Fig. S7J). SSRP1 and SUPT16H were also downregulated by CBL0137 (Supplementary Fig. S5G). Furthermore, we tested CBL0137 therapy on orthotopic xenografting model of SHH-MB_PDX1. The results showed that CBL0137 could significantly improve overall survival of intracranial xenografted mice without significantly altering their body weight, proving the therapeutic potential of CBL0137 in treating intracranial Hh-driven human cancers (Fig. 5H; Supplementary Fig. S7I).

Next, we asked whether CBL0137 treatment could also target aberrant Hh pathway in Hh-driven human cancers. Our qRT-PCR analyses revealed that GLI1/2 transcripts were downregulated in all three Hh-driven human tumor models treated with CBL0137, THZ1, and JQ1 but not GDC0449 in vitro (Fig. 6A). Western blot data also confirmed the reduction of GLI1 protein in response to CBL0137 treatment (Fig. 6B). Moreover, we found GLI1/2 transcripts were also significantly downregulated by CBL0137 when treating subcutaneous SHH-MB_PDX1 model in vivo (Fig. 6C). Furthermore, ChIP-qPCR analyses revealed that CBL0137 treatment resulted in significant reduction of FACT binding to the promoter regions of GLI1/2 in all tested Hh-driven human tumor models (Fig. 6DG), indicating a conserved anti-Hh mechanism of CBL0137 when treating human and mouse Hh-driven cancers.

Figure 6.

CBL0137 inhibits aberrant Hh pathway in human Hh-driven tumor models. A, qRT-PCR analyses of GLI1 and GLI2 in Hh-driven human tumor cells (ATRT03, SHH-MB_PDX1, and SHH-MB_PDX2) treated with CBL0137 (1 μmol/L), JQ1 (1 μmol/L), THZ1 (0.1 μmol/L), GDC0449 (1 μmol/L), or DMSO for 24 hours. B, Western blot of GLI1 in Hh-driven human tumor cells treated with CBL0137 (1 μmol/L) or DMSO treatment for 24 hours. C, qRT-PCR analysis of GLI1 and GLI2 in flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle. D–G, ChIP-qPCR analyses of SSRP1 binding at promoters of GLI1 and GLI2 in Hh-driven human tumor cells treated overnight with CBL0137 (1 μmol/L) or DMSO. All qRT-PCR and ChIP-qPCR assays were performed in triplicate, and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. ns, nonsignificant.

Figure 6.

CBL0137 inhibits aberrant Hh pathway in human Hh-driven tumor models. A, qRT-PCR analyses of GLI1 and GLI2 in Hh-driven human tumor cells (ATRT03, SHH-MB_PDX1, and SHH-MB_PDX2) treated with CBL0137 (1 μmol/L), JQ1 (1 μmol/L), THZ1 (0.1 μmol/L), GDC0449 (1 μmol/L), or DMSO for 24 hours. B, Western blot of GLI1 in Hh-driven human tumor cells treated with CBL0137 (1 μmol/L) or DMSO treatment for 24 hours. C, qRT-PCR analysis of GLI1 and GLI2 in flank xenografts of SHH-MB_PDX1 treated with CBL0137 or vehicle. D–G, ChIP-qPCR analyses of SSRP1 binding at promoters of GLI1 and GLI2 in Hh-driven human tumor cells treated overnight with CBL0137 (1 μmol/L) or DMSO. All qRT-PCR and ChIP-qPCR assays were performed in triplicate, and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test. ns, nonsignificant.

Close modal

CBL0137 works synergistically with other anti-Hh therapeutic agents on inhibiting Hh-driven cancer cells

It has been well recognized that single-agent targeted therapy could easily run into obstacles of primary or acquired resistance, which usually can be effectively overcome by combinatory therapy (51). Therefore, we systemically evaluate the efficacy of combinatory anti-Hh therapies using FACTi together with SMOi, BETi or CDK7i in both mouse and human Hh-driven tumor models in vitro. The inhibitory effects on tumor cell viability were measured with increasing doses of each anti-Hh agent individually or in combination and CI or BLISS synergy scores were calculated for each combination. CI value less than 1.0 and BLISS score more than 0 were considered synergistic. To further illustrate the combinatory therapeutic effects on aberrant Hh pathway, transcript levels of Gli1/2 (in mouse SHH-MB cells) or GLI1/2 (in Hh-driven human tumor cells) under synergistic conditions were also measured. Our results showed that CBL0137 worked synergistically with GDC0449 on inhibiting growth and Gli1/2 transcription of SMOi-sensitive SmoWT cells (Supplementary Fig. S8A). Moreover, CBL0137 worked synergistically with JQ1 or THZ1 on suppressing growth and Gli1/2 transcription of both SmoWT and SMOi-resistant SmoWT-sgSufu cells (Supplementary Fig. S8B–S8E). Similar results were obtained when the three SMOi-resistant Hh-driven human tumor models were tested, including ATRT03, SHH-MB_PDX1, and SHH-MB_PDX2 (Fig. 7A–F). CBL0137 worked synergistically with JQ1 or THZ1 on suppressing growth and GLI1/2 transcription in all of them (Fig. 7A–F). Together, these results revealed the synergistic therapeutic potential of FACTi with other anti-Hh agents for treating Hh-driven cancer cells and overcoming SMOi resistance.

Figure 7.

CBL0137 inhibits in vitro growth of human Hh-driven tumor cells in synergy with BET or CDK7 inhibitor. Cell viability of Hh-driven human tumor cells (A and B, ATRT03; C and D, SHH-MB_PDX1; E and F, SHH-MB_PDX2) treated with combinations of CBL0137 (CBL) and JQ1 or THZ1 at indicated concentrations for 72 hours (left). qRT-PCR analysis of GLI1 and GLI2 in Hh-driven human tumor cells (A and B, ATRT03; C and D, SHH-MB_PDX1; E and F, SHH-MB_PDX2) treated with indicated drug combinations for 24 hours (right). The CI and BLISS scores for each combination condition are listed, and the values of synergistic conditions (CI < 1.0 or BLISS score > 0) are shown in green. All assays were performed in triplicate and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test.

Figure 7.

CBL0137 inhibits in vitro growth of human Hh-driven tumor cells in synergy with BET or CDK7 inhibitor. Cell viability of Hh-driven human tumor cells (A and B, ATRT03; C and D, SHH-MB_PDX1; E and F, SHH-MB_PDX2) treated with combinations of CBL0137 (CBL) and JQ1 or THZ1 at indicated concentrations for 72 hours (left). qRT-PCR analysis of GLI1 and GLI2 in Hh-driven human tumor cells (A and B, ATRT03; C and D, SHH-MB_PDX1; E and F, SHH-MB_PDX2) treated with indicated drug combinations for 24 hours (right). The CI and BLISS scores for each combination condition are listed, and the values of synergistic conditions (CI < 1.0 or BLISS score > 0) are shown in green. All assays were performed in triplicate and the data are presented as the means ± SD. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired two-tailed Student t test.

Close modal

Despite the big success of SMOi drug development and clinical application, primary and acquired SMOi resistance have been increasingly recognized as crucial obstacles for treating Hh-driven cancers (12, 52). We and others have recently reported that epigenetic/transcriptional-targeted drugs, such as CDK7 inhibitor, BET inhibitor, or HDAC inhibitor, could effectively overcoming SMOi resistance through targeting the aberrant Hh pathway at GLI transactivation process, opening a plethora of new therapeutic opportunities (20–23).

In this study, we identified SSRP1, which encodes a subunit of the heterodimer FACT complex, as a novel Hh-induced oncogene that plays an important role in regulating transcriptional output of the aberrant Hh pathway in Hh-driven cancers (Figs. 1, 4, and 6). The FACT inhibitor drug CBL0137 was able to effectively treat various SMOi-sensitive and SMOi-resistant Hh-driven tumor models both in vitro and in vivo, including a human intracranial orthotopic SHH-MB PDX model (Figs. 2, 3, and 5). We further showed that CBL0137 worked synergistically with SMO inhibitor, BET inhibitor or CDK7 inhibitor on suppressing these Hh-driven tumor models in vitro (Fig. 7; Supplementary Fig. S8). Together, our study illustrates FACT inhibition as novel effective epigenetic/transcriptional-targeted therapeutic strategy for treating Hh-driven cancers and overcoming SMOi resistance. Because small-molecule drugs of all above-mentioned anti-Hh therapies are either tested in human clinical trials or FDA approved, our work provides preclinical support for initiating clinical trials of CBL0137 alone or in combination for treating Hh-driven cancers. More importantly, many of these drugs have been proven to exhibit efficient brain-penetrating capacity, thus supporting their application against intracranial Hh-activated cancer types such as SHH-MB and SHH-ATRT.

Because the primary goal of this study is to identify novel anti-Hh therapeutic strategy, we focused on dissecting the inhibitory effects of CBL0137 on transcriptional output of aberrant Hh pathway when investigating its therapeutic mechanisms against Hh-driven cancers. The qRT-PCR and Western blot analyses showed that CBL0137 treatment resulted in robust transcriptional blockade of GLI1/2, the core transcriptional factors of Hh pathway, in all tested mouse and human Hh-driven tumor models, including multiple SMOi-resistant ones. RNA-seq analyses of GDC0449- or CBL0137-treated SmoWT cells confirmed that CBL0137 could significantly suppress Hh-activated oncogenic transcriptional program, which was enriched in metabolism and translation-related biological functions or pathways (Supplementary Fig. S6A–S6C). Furthermore, our ChIP-qPCR analyses of DMSO or CBL0137-treated cancer cells demonstrated that CBL0137 inhibited GLI1/2 transcription through disrupting the binding of FACT complex to their promoter regions. Notably, recent studies reported that CBL0137 could also cause chromatin damage and trapping of FACT complex, resulting in its redistribution and subsequent disruption of spatial genome organization and enhancer function in cancer cells (34, 35, 53). Whether CBL0137 could induce similar global destructive effects on FACT distribution or chromatin structure of Hh-activated cancer cells and how would these contribute to its anti-Hh activity requires further investigation.

FACT complex is composed of SSRP1 and its partner SUPT16H. Both FACT subunit coding genes have been demonstrated to be upregulated and associated with more aggressiveness, stemness, and poor prognosis in various cancer types (28, 45). In line with previous studies, we also found patients with SHH-subtype medulloblastoma with higher SSRP1 level exhibited significantly worse prognosis (Fig. 5B), and previously reported FACT-associated oncogenic signatures, such as DNA replication, repair, and transcription, were enriched in significantly upregulated genes of SSRP1-high SHH-MB (Supplementary Fig. S9A–S9C). In contrast, SSRP1-high SHH-MB did not exhibit significant upregulation of Hh pathway gene signature (Supplementary Fig. S9D). More importantly, RNA-seq analysis of CBL0137-treated SmoWT cells revealed that the significantly upregulated genes of SSRP1-high SHH-MB could be effectively suppressed by CBL0137 treatment, suggesting that FACT inhibition targeted not only the aberrant Hh pathway but also some other cancer-dependent biological processes and pathways when treating Hh-driven cancers (Supplementary Fig. S9E). Because majority of CBL0137-downregulated genes (1,066/1,289 = 82.7%) did not overlap with GDC0449-downregulated genes (Fig. 4B), further investigation of Hh-independent target genes and related therapeutic mechanisms of CBL0137 would potentially help better understand FACT-targeted therapy against Hh-driven cancers and provide more insight for designing future clinical trials. For example, it would be intriguing to test whether CBL0137 would work synergistically with radiotherapy, chemotherapy, or PARP inhibitor drugs for treating Hh-driven cancers through blocking DNA repair and enhancing DNA damage as reported before in other cancer types (29, 33).

In summary, our study proposes FACT inhibition as another effective epigenetic/transcriptional-targeted therapeutic strategy for suppressing aberrant Hh pathway and overcoming SMOi resistance. More importantly, we identify a clinical-trial-applicable FACT inhibitor drug CBL0137 as an effective anti-Hh therapeutic agent, providing preclinical support for initiating its clinical trials for treating Hh-driven cancers, including the intracranial ones. New CBL0137 phase I trials for child patients would particularly be necessary as all the current ones do not recruit patients less than 14 years old. On the other hand, as most of the candidate genes in our initial screening have not been studied in Hh-driven cancers yet, we believed there should still be other epigenetic/transcriptional modulators that could serve as effective anti-Hh therapeutic targets for overcoming SMOi resistance. Particularly, we noticed there were multiple RNA splicing modulators in our candidate genes (Supplementary Fig. S1B). As a recent study reported recurrent noncoding U1 snoRNA mutations could drive aberrant alternative splicing that activated Hh pathway in adult and adolescent SHH-MB (54), it would be interesting to further examine whether other splicing-related mechanisms could contribute to aberrant Hh activity and serve as potential therapeutic targets for Hh-driven cancers.

J. Mo reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. F. Liu reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. X. Sun reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. H. Huang reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. K. Tan reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. R. Li reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. W. Jiang reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. Y. Sui reports personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. X. Chen reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. J. Ma reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. K. Zhao reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. Y. Tang reports grants and personal fees from Chinese government during the conduct of the study and personal fees from Chinese government outside the submitted work. No disclosures were reported by the other authors.

J. Mo: Conceptualization, data curation, supervision, funding acquisition, validation, investigation, visualization, writing–original draft, project administration, writing–review and editing. F. Liu: Conceptualization, data curation, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. X. Sun: Data curation, software, validation, investigation, visualization, methodology. H. Huang: Data curation, software, investigation, visualization. K. Tan: Investigation. X. Zhao: Software, investigation. R. Li: Software, investigation. W. Jiang: Investigation. Y. Sui: Conceptualization, Resources, investigation. X. Chen: Conceptualization, resources, investigation. K. Shen: Conceptualization, software, visualization. L. Zhang: Conceptualization, software, supervision, visualization, project administration, writing–review and editing. J. Ma: Conceptualization, resources, software, supervision, visualization, project administration, writing–review and editing. K. Zhao: Conceptualization, resources, software, supervision, investigation, visualization, project administration, writing–review and editing. Y. Tang: Conceptualization, resources, software, supervision, funding acquisition, investigation, visualization, project administration, writing–review and editing.

This work was supported by Chinese Universities Scientific Fund, Innovative Research Team of High-Level Local Universities in Shanghai (SSMU-ZDCX20180800), the Recruitment Program of Global Experts of China (to Y. Tang), the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning (to Y. Tang), Shanghai Rising-Star Program (to Y. Tang), and National Natural Science Foundation of China (81572761, 81772655 and 81972646 to Y. Tang; 81572692 to K. Zhao; 81772797 to K. Shen; 31871332 to L. Zhang; 82002978 to J. Mo), National Key Research and Development Program of China (2018YFC1004602 to L. Zhang), Postdoctoral Science Foundation of China (2019M651527 to J. Mo), Shanghai Science and Technology Committee (17411951800 and 19411952100 to J. Ma), Shanghai Municipal Education Commission Gaofeng Clinical Medicine Grant Support (20172007 to X. Chen), and Guangci Distinguished Young Scholars Training Program (GCQN-2017-A18 to X. Chen). The authors thank Yoon-Jae Cho (Oregon Health & Science University), Rosalind A. Segal (Dana-Farber Cancer Institute), Charles M. Rudin (Memorial Sloan-Kettering Cancer Center), Fei Lan (Fudan University School of Medicine), Mingliang Zhang (Shanghai Jiao Tong University School of Medicine), and Jing Xue (Shanghai Renji Hospital) for reagents and/or helpful suggestions. The authors would like to thank Yingdong Zhang on his technical support on the HPC. This work was supported by the HPC platform of ShanghaiTech University.

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.

1.
Varjosalo
M
,
Taipale
J
. 
Hedgehog: functions and mechanisms
.
Genes Dev
2008
;
22
:
2454
72
.
2.
Alcedo
J
,
Ayzenzon
M
,
Von Ohlen
T
,
Noll
M
,
Hooper
JE
. 
The drosophila smoothened gene encodes a seven-pass membrane protein, a putative receptor for the hedgehog signal
.
Cell
1996
;
86
:
221
32
.
3.
Rohatgi
R
,
Milenkovic
L
,
Scott
MP
. 
Patched1 regulates hedgehog signaling at the primary cilium
.
Science
2007
;
317
:
372
6
.
4.
Kim
J
,
Kato
M
,
Beachy
PA
. 
Gli2 trafficking links Hedgehog-dependent activation of Smoothened in the primary cilium to transcriptional activation in the nucleus
.
Proc Natl Acad Sci U S A
2009
;
106
:
21666
71
.
5.
Briscoe
J
,
Therond
PP
. 
The mechanisms of hedgehog signalling and its roles in development and disease
.
Nat Rev Mol Cell Biol
2013
;
14
:
416
29
.
6.
Evangelista
M
,
Tian
H
,
de Sauvage
FJ
. 
The hedgehog signaling pathway in cancer
.
Clin Cancer Res
2006
;
12
:
5924
8
.
7.
Cortes
JE
,
Gutzmer
R
,
Kieran
MW
,
Solomon
JA
. 
Hedgehog signaling inhibitors in solid and hematological cancers
.
Cancer Treat Rev
2019
;
76
:
41
50
.
8.
Hanna
A
,
Shevde
LA
. 
Erratum to: ‘hedgehog signaling: modulation of cancer properties and tumor microenvironment’
.
Mol Cancer
2016
;
15
:
35
.
9.
Yang
L
,
Xie
G
,
Fan
Q
,
Xie
J
. 
Activation of the hedgehog-signaling pathway in human cancer and the clinical implications
.
Oncogene
2010
;
29
:
469
81
.
10.
LoRusso
PM
,
Rudin
CM
,
Reddy
JC
,
Tibes
R
,
Weiss
GJ
,
Borad
MJ
, et al
Phase I trial of hedgehog pathway inhibitor vismodegib (GDC-0449) in patients with refractory, locally advanced or metastatic solid tumors
.
Clin Cancer Res
2011
;
17
:
2502
11
.
11.
Kieran
MW
,
Chisholm
J
,
Casanova
M
,
Brandes
AA
,
Aerts
I
,
Bouffet
E
, et al
Phase I study of oral sonidegib (LDE225) in pediatric brain and solid tumors and a phase II study in children and adults with relapsed medulloblastoma
.
Neuro Oncol
2017
;
19
:
1542
52
.
12.
Yauch
RL
,
Dijkgraaf
GJ
,
Alicke
B
,
Januario
T
,
Ahn
CP
,
Holcomb
T
, et al
Smoothened mutation confers resistance to a hedgehog pathway inhibitor in medulloblastoma
.
Science
2009
;
326
:
572
4
.
13.
Sharpe
HJ
,
Pau
G
,
Dijkgraaf
GJ
,
Basset-Seguin
N
,
Modrusan
Z
,
Januario
T
, et al
Genomic analysis of smoothened inhibitor resistance in basal cell carcinoma
.
Cancer Cell
2015
;
27
:
327
41
.
14.
Kool
M
,
Jones
DT
,
Jager
N
,
Northcott
PA
,
Pugh
TJ
,
Hovestadt
V
, et al
Genome sequencing of SHH medulloblastoma predicts genotype-related response to smoothened inhibition
.
Cancer Cell
2014
;
25
:
393
405
.
15.
Lee
Y
,
Kawagoe
R
,
Sasai
K
,
Li
Y
,
Russell
HR
,
Curran
T
, et al
Loss of suppressor-of-fused function promotes tumorigenesis
.
Oncogene
2007
;
26
:
6442
7
.
16.
Jagani
Z
,
Mora-Blanco
EL
,
Sansam
CG
,
McKenna
ES
,
Wilson
B
,
Chen
D
, et al
Loss of the tumor suppressor Snf5 leads to aberrant activation of the Hedgehog-Gli pathway
.
Nat Med
2010
;
16
:
1429
33
.
17.
Zhao
X
,
Ponomaryov
T
,
Ornell
KJ
,
Zhou
P
,
Dabral
SK
,
Pak
E
, et al
RAS/MAPK activation drives resistance to smo inhibition, metastasis, and tumor evolution in shh pathway-dependent tumors
.
Cancer Res
2015
;
75
:
3623
35
.
18.
Lin
L
,
Huang
M
,
Shi
X
,
Mayakonda
A
,
Hu
K
,
Jiang
YY
, et al
Super-enhancer-associated MEIS1 promotes transcriptional dysregulation in Ewing sarcoma in co-operation with EWS-FLI1
.
Nucleic Acids Res
2019
;
47
:
1255
67
.
19.
Wu
B
,
Pan
X
,
Chen
X
,
Chen
M
,
Shi
K
,
Xu
J
, et al
Epigenetic drug library screening identified an LSD1 inhibitor to target UTX-deficient cells for differentiation therapy
.
Signal Transduct Target Ther
2019
;
4
:
11
.
20.
Tang
Y
,
Gholamin
S
,
Schubert
S
,
Willardson
MI
,
Lee
A
,
Bandopadhayay
P
, et al
Epigenetic targeting of Hedgehog pathway transcriptional output through BET bromodomain inhibition
.
Nat Med
2014
;
20
:
732
40
.
21.
Gruber
W
,
Peer
E
,
Elmer
DP
,
Sternberg
C
,
Tesanovic
S
,
Del Burgo
P
, et al
Targeting class I histone deacetylases by the novel small molecule inhibitor 4SC-202 blocks oncogenic hedgehog-GLI signaling and overcomes smoothened inhibitor resistance
.
Int J Cancer
2018
;
142
:
968
75
.
22.
Liu
F
,
Jiang
W
,
Sui
Y
,
Meng
W
,
Hou
L
,
Li
T
, et al
CDK7 inhibition suppresses aberrant hedgehog pathway and overcomes resistance to smoothened antagonists
.
Proc Natl Acad Sci U S A
2019
;
116
:
12986
95
.
23.
Pak
E
,
MacKenzie
EL
,
Zhao
X
,
Pazyra-Murphy
MF
,
Park
PMC
,
Wu
L
, et al
A large-scale drug screen identifies selective inhibitors of class I HDACs as a potential therapeutic option for SHH medulloblastoma
.
Neuro Oncol
2019
;
21
:
1150
63
.
24.
Orphanides
G
,
Wu
WH
,
Lane
WS
,
Hampsey
M
,
Reinberg
D
. 
The chromatin-specific transcription elongation factor FACT comprises human SPT16 and SSRP1 proteins
.
Nature
1999
;
400
:
284
8
.
25.
Belotserkovskaya
R
,
Saunders
A
,
Lis
JT
,
Reinberg
D
. 
Transcription through chromatin: understanding a complex FACT
.
Biochim Biophys Acta
2004
;
1677
:
87
99
.
26.
Tan
BC
,
Chien
CT
,
Hirose
S
,
Lee
SC
. 
Functional cooperation between FACT and MCM helicase facilitates initiation of chromatin DNA replication
.
EMBO J
2006
;
25
:
3975
85
.
27.
Saunders
A
,
Werner
J
,
Andrulis
ED
,
Nakayama
T
,
Hirose
S
,
Reinberg
D
, et al
Tracking FACT and the RNA polymerase II elongation complex through chromatin in vivo
.
Science
2003
;
301
:
1094
6
.
28.
Garcia
H
,
Fleyshman
D
,
Kolesnikova
K
,
Safina
A
,
Commane
M
,
Paszkiewicz
G
, et al
Expression of FACT in mammalian tissues suggests its role in maintaining of undifferentiated state of cells
.
Oncotarget
2011
;
2
:
783
96
.
29.
De
S
,
Lindner
DJ
,
Coleman
CJ
,
Wildey
G
,
Dowlati
A
,
Stark
GR
. 
The FACT inhibitor CBL0137 Synergizes with cisplatin in small-cell lung cancer by increasing NOTCH1 expression and targeting tumor-initiating cells
.
Cancer Res
2018
;
78
:
2396
406
.
30.
Kim
M
,
Neznanov
N
,
Wilfong
CD
,
Fleyshman
DI
,
Purmal
AA
,
Haderski
G
, et al
Preclinical validation of a single-treatment infusion modality that can eradicate extremity melanomas
.
Cancer Res
2016
;
76
:
6620
30
.
31.
Somers
K
,
Kosciolek
A
,
Bongers
A
,
El-Ayoubi
A
,
Karsa
M
,
Mayoh
C
, et al
Potent antileukemic activity of curaxin CBL0137 against MLL-rearranged leukemia
.
Int J Cancer
2020
;
146
:
1902
16
.
32.
Barone
TA
,
Burkhart
CA
,
Safina
A
,
Haderski
G
,
Gurova
KV
,
Purmal
AA
, et al
Anticancer drug candidate CBL0137, which inhibits histone chaperone FACT, is efficacious in preclinical orthotopic models of temozolomide-responsive and -resistant glioblastoma
.
Neuro Oncol
2017
;
19
:
186
96
.
33.
Carter
DR
,
Murray
J
,
Cheung
BB
,
Gamble
L
,
Koach
J
,
Tsang
J
, et al
Therapeutic targeting of the MYC signal by inhibition of histone chaperone FACT in neuroblastoma
.
Sci Transl Med
2015
;
7
:
312ra176
.
34.
Chang
HW
,
Valieva
ME
,
Safina
A
,
Chereji
RV
,
Wang
J
,
Kulaeva
OI
, et al
Mechanism of FACT removal from transcribed genes by anticancer drugs curaxins
.
Sci Adv
2018
;
4
:
eaav2131
.
35.
Gasparian
AV
,
Burkhart
CA
,
Purmal
AA
,
Brodsky
L
,
Pal
M
,
Saranadasa
M
, et al
Curaxins: anticancer compounds that simultaneously suppress NF-kappaB and activate p53 by targeting FACT
.
Sci Transl Med
2011
;
3
:
95ra74
.
36.
Jin
MZ
,
Xia
BR
,
Xu
Y
,
Jin
WL
. 
Curaxin CBL0137 exerts anticancer activity via diverse mechanisms
.
Front Oncol
2018
;
8
:
598
.
37.
Joung
J
,
Konermann
S
,
Gootenberg
JS
,
Abudayyeh
OO
,
Platt
RJ
,
Brigham
MD
, et al
Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening
.
Nat Protoc
2017
;
12
:
828
63
.
38.
Bliss
CI
. 
The toxicity of poisons applied jointly
.
Ann Appl Biol
1939
;
26
:
585
615
.
39.
Mo
JL
,
Liu
Q
,
Kou
ZW
,
Wu
KW
,
Yang
P
,
Chen
XH
, et al
MicroRNA-365 modulates astrocyte conversion into neuron in adult rat brain after stroke by targeting Pax6
.
Glia
2018
;
66
:
1346
62
.
40.
Cho
YJ
,
Tsherniak
A
,
Tamayo
P
,
Santagata
S
,
Ligon
A
,
Greulich
H
, et al
Integrative genomic analysis of medulloblastoma identifies a molecular subgroup that drives poor clinical outcome
.
J Clin Oncol
2011
;
29
:
1424
30
.
41.
Northcott
PA
,
Buchhalter
I
,
Morrissy
AS
,
Hovestadt
V
,
Weischenfeldt
J
,
Ehrenberger
T
, et al
The whole-genome landscape of medulloblastoma subtypes
.
Nature
2017
;
547
:
311
7
.
42.
Robinson
G
,
Parker
M
,
Kranenburg
TA
,
Lu
C
,
Chen
X
,
Ding
L
, et al
Novel mutations target distinct subgroups of medulloblastoma
.
Nature
2012
;
488
:
43
48
.
43.
Schuller
U
,
Heine
VM
,
Mao
J
,
Kho
AT
,
Dillon
AK
,
Han
YG
, et al
Acquisition of granule neuron precursor identity is a critical determinant of progenitor cell competence to form Shh-induced medulloblastoma
.
Cancer Cell
2008
;
14
:
123
34
.
44.
Buonamici
S
,
Williams
J
,
Morrissey
M
,
Wang
A
,
Guo
R
,
Vattay
A
, et al
Interfering with resistance to smoothened antagonists by inhibition of the PI3K pathway in medulloblastoma
.
Sci Transl Med
2010
;
2
:
51ra70
.
45.
Kim
J
,
Aftab
BT
,
Tang
JY
,
Kim
D
,
Lee
AH
,
Rezaee
M
, et al
Itraconazole and arsenic trioxide inhibit hedgehog pathway activation and tumor growth associated with acquired resistance to smoothened antagonists
.
Cancer Cell
2013
;
23
:
23
34
.
46.
Garcia
H
,
Miecznikowski
JC
,
Safina
A
,
Commane
M
,
Ruusulehto
A
,
Kilpinen
S
, et al
Facilitates chromatin transcription complex is an “accelerator” of tumor transformation and potential marker and target of aggressive cancers
.
Cell Rep
2013
;
4
:
159
73
.
47.
Shahi
MH
,
Afzal
M
,
Sinha
S
,
Eberhart
CG
,
Rey
JA
,
Fan
X
, et al
Human hedgehog interacting protein expression and promoter methylation in medulloblastoma cell lines and primary tumor samples
.
J Neurooncol
2011
;
103
:
287
96
.
48.
Lee
EY
,
Ji
H
,
Ouyang
Z
,
Zhou
B
,
Ma
W
,
Vokes
SA
, et al
Hedgehog pathway-regulated gene networks in cerebellum development and tumorigenesis
.
Proc Natl Acad Sci U S A
2010
;
107
:
9736
41
.
49.
Chun
HE
,
Johann
PD
,
Milne
K
,
Zapatka
M
,
Buellesbach
A
,
Ishaque
N
, et al
Identification and analyses of extra-cranial and cranial rhabdoid tumor molecular subgroups reveal tumors with cytotoxic T cell infiltration
.
Cell Rep
2019
;
29
:
2338
2354
.
50.
Johann
PD
,
Erkek
S
,
Zapatka
M
,
Kerl
K
,
Buchhalter
I
,
Hovestadt
V
, et al
Atypical teratoid/rhabdoid tumors are comprised of three epigenetic subgroups with distinct enhancer landscapes
.
Cancer Cell
2016
;
29
:
379
93
.
51.
Wang
L
,
Wang
H
,
Song
D
,
Xu
M
. 
Liebmen m new strategies for targeting drug combinations to overcome mutation-driven drug resistance
.
Semin Cancer Biol
2017
;
42
:
44
51
.
52.
Metcalfe
C
,
de Sauvage
FJ
. 
Hedgehog fights back: mechanisms of acquired resistance against Smoothened antagonists
.
Cancer Res
2011
;
71
:
5057
61
.
53.
Kantidze
OL
,
Luzhin
AV
,
Nizovtseva
EV
,
Safina
A
,
Valieva
ME
,
Golov
AK
, et al
The anti-cancer drugs curaxins target spatial genome organization
.
Nat Commun
2019
;
10
:
1441
.