Interleukin-24 (IL24) Is Suppressed by PAX3-FOXO1 and Is a Novel Therapy for Rhabdomyosarcoma Free

Rhabdomyosarcoma (RMS) is primarily observed in children and adolescents and accounts for approximately 5% of all pediatric cancers, with incidence rates of 4.5/10⁶ (1–3). Embryonal RMS (ERMS) and alveolar RMS (ARMS) are the two major classes of RMS in children and adolescents and differ with respect to their histology, genetics, treatment, and prognosis (3, 4). ERMS accounts for over 60% of RMS patients and is associated with loss of heterozygosity at the 11p15 locus (4). ERMS patients have a favorable initial prognosis; however, the overall survival of patients with metastatic ERMS is only 40% (5). Cytogenetic analysis of tumors from ARMS patients has identified that 2;13 and 1;13 chromosomal translocations generating PAX3-FOXO1 and PAX7-FOXO1 fusion genes, respectively, are highly prevalent (55% and 22%, respectively; ref. 6). The PAX3-FOXO1 fusion gene is the critical prognostic marker for ARMS patients with metastatic disease, with an estimated overall 4-year survival rate of 8% compared with 75% survival rate of patients with PAX7-FOXO1–expressing tumors (7–9).

Results of PAX3-FOXO1 knockdown or overexpression studies in RMS and other cell lines demonstrate the functional importance of this fusion gene in maintaining the aggressive cancer cell phenotype, and this is due, in part, to the pro-oncogenic PAX3-FOXO1–regulated genes (10–12). Treatments for RMS patients include radiotherapy, surgery, and chemotherapy with cytotoxic drugs and/or drug combinations of vincristine, dactinomycin, cyclophosphamide, irinotecan, ifosfamide, etoposide, doxorubicin, and others. A serious problem exists for RMS patients who survive current cytotoxic drug therapies, because these individuals as adults have an increased risk for several diseases (13).

The orphan nuclear receptor NR4A1 is overexpressed in colon, pancreatic, breast (estrogen receptor–positive and negative), and lung tumors. In breast, colon, and lung tumor patients, high expression of NR4A1 predicts decreased survival (14–19). Knockdown or overexpression of NR4A1 in cancer cells shows that this receptor regulates multiple cell-specific responses including cell proliferation, survival, cell-cycle progression, migration, and invasion in lung, melanoma, lymphoma, pancreatic, colon, cervical, ovarian, and gastric cancer cell lines (11, 14, 17, 19–27).

Studies in this laboratory have reported that NR4A1 is also overexpressed in RMS tumors compared with normal muscle tissue (27), and NR4A1 regulated many of the same genes/pathways observed in other solid tumors (16, 19, 20, 27–32). The NR4A1 ligand 1,1-bis(3-indolyl)-1-(p-hydroxyphenyl)methane (DIM-C-pPhOH), which acts as a receptor antagonist, inhibits growth, survival, and migration of RMS cells and also inhibits the NR4A1-regulated PAX3-FOXO1 oncogene and RMS tumor growth in a xenograft mouse model (27, 32). The present study was initiated after analysis of RNA-seq data showed that knockdown of NR4A1 or PAX3-FOXO1 or treatment with the NR4A1 antagonist DIM-C-pPhOH resulted in a 2- to 27.9-fold induction of the tumor suppressor–like factor interleukin-24 (IL24). In this study, we show that the oncogenic activity of PAX3-FOXO1 is due, in part, to suppression of IL24, and the anticancer activities observed after PAX3-FOXO1 knockdown/suppression are due primarily to induction of IL24. Overexpression of IL24 inhibited ARMS cell growth, survival, and migration, and IL24 (adenoviral) dramatically inhibited tumor growth in vivo, suggesting that the clinically approved IL24 adenoviral expression vector potentially (33) represents a promising new approach for ARMS therapy.

Rh30 human RMS cancer cells were obtained from the ATCC and authenticated in 2014 and were maintained at 37°C in the presence of 5% CO₂ in RPMI-1640 medium and supplemented with 10% fetal bovine serum and 5% antibiotic. Rh18 and Rh41 cells were received from Texas Tech University Health Sciences Center-Children's Oncology Group in 2015. Rh41 and Rh18 cell lines were maintained in Iscove's Modified Dulbecco's Medium (IMDM) supplemented with 20% fetal bovine serum, 1× ITS (5 μg/mL insulin, 5 μg/mL transferrin, 5 ng/mL selenous acid), and 5% antibiotic. IMDM was purchased from Thermo Fisher Scientific. RPMI-1640 was purchased from Sigma-Aldrich, and Lipofectamine 2000 for siRNA transfection was purchased from Invitrogen. Apoptotic, necrotic, and Healthy Cells Quantification Kit was purchased from Biotium. Cells were subsequently viewed using a filter set for FITC, rhodamine, and DAPI on an Advanced Microscopy EVOS fl, fluorescence microscope. RGB-4103 GelRed nucleic acid stain was used in place of Ethedium Bromide from Phenix Research Products. SB203580 was purchased from Cell Signaling Technology. The human IL24 cDNA clone in a pCMV-6 vector was purchased from Origene; the HDAC5-flag expression plasmid was obtained from Addgene. The C-DIM compounds were prepared as previously described (19), and a summary of the antibodies is provided in Supplementary Table S1.

Patient sample data of total RNA for sample sets were acquired from the NCBI geo database. Array samples were previously analyzed for quality control, quantile normalized, and multiple probes for each gene were averaged by the submitter. Each sample represented a normalized signal intensity and is plotted as relative expression. Additional hybridization and sample processing information can be found at https://www.ncbi.nlm.nih.gov/geo/ for NCBI GSE data sets and https://www.ebi.ac.uk/arrayexpress/ for the E-TABM-1202 data set. Relative expression values were listed in JMP Statistical Software, and a box plot was generated, from which a t test was performed; significance was determined as a P value less than 0.01 or 0.5, shown by an asterisk. RMS cells were treated with DIM-C-pPhOH for 48 hours or transfected with siPAX3-FOXO1 or siNR4A1 for 72 hours, after which RNA was extracted and sent to the Texas A&M AgriLife Sequencing Core for preparation, sequencing, and analysis. IL24 cDNA was transfected into RMS cells using Lipofectamine 2000 delivery at a concentration of 50 μmol/L before endpoint analysis using Western blot, PCR, or Annexin V staining. IL24 and GFP adenoviral vectors were purchased from Vector Biolabs with a human adenovirus type 5 (dE1/E3) viral backbone under the CMV promoter.

Rh30, Rh18, and Rh41 cells were plated in 12-well plates at 1.0 × 10⁵ and allowed to attach for 24 hours before treatment with DIM-C-pPhOH, DIM-C-pPhCO₂Me, or transfected with siNR4A1, siPAX3-FOXO1, or siIL24 with DMSO (dimethyl sulfoxide) as empty vehicle or siCtl siRNA (with lipofectamine vehicle) as controls, respectively. Cells were then trypsinized and counted at 24 hours using a Coulter Z1 cell counter.

RMS cells were seeded at 1.0 × 10⁵ in 2-well Lab-Tek chambered B#1.0 Borosilicate coverglass slides from Thermo Scientific and were allowed to attach for 24-hour transfection with IL24 overexpression vector, siPAX3-FOXO1 (100 μmol/L), or siIL24 (100 μmol/L) with a control of siCtl (with lipofectamine vehicle) for 72 hours, and Annexin V staining was determined as described (28). Hoechst staining from the apoptotic and necrotic cells assay (Biotium) was used to visualize apoptotic cells. Images were taken using an EVOS fluorescence microscopy from Advance Microscopy. Rh30 cells were immunostained for IL24 essentially as described (31).

RMS cells were seeded for 24 hours in a 24-well plate, allowed to attach, and subsequently transfected with IL24 overexpression vector, siPAX3-FOXO1 (100 μmol/L), or siIL24 (100 μmol/L) with a control of siCtl (nonspecific oligonucleotide). The cells were trypsinized, counted, placed in 24-well 8.0-μm-pore ThinCerts from BD Biosciences, allowed to migrate for 24 hours, fixed with formaldehyde, and then stained with hematoxylin. Cells that migrated through the pores were then counted.

Rh30, Rh18, and Rh41 cells were seeded in 6-well plates at 1.0 × 10⁵ and allowed to attached for 24 hours before treatment with DIM-C-pPhOH, DIM-C-pPhCO₂Me, or transfected with siNR4A1, siPAX3-FOXO1, or siIL24 with DMSO as empty vehicle or siCtl siRNA (nonspecific scrambled oligonucleotide; with lipofectamine vehicle) as controls, respectively. Cells were treated with C-DIMs or DMSO for 48 hours or transfected with siNR4A1, siPAX3-FOXO1, or siIL24 (all at 100 μmol/L) or siCtl for 72 hours, and Western blots of whole-cell lysates were determined as previously described (28).

Real-time PCR assays using RMS cell lines transfected with oligonucleotides or treated with C-DIMs were carried out using the SYBR Green RT-PCR Kit (Bio-Rad Laboratories) according to the manufacturer's protocol, and the ChIP assay for determining changes in pol II was carried out as described (28). Oligonucleotides and primers used are summarized in Supplementary Table S1.

Female athymic nude mice (6–8 weeks old) were obtained (Charles River Laboratory) and maintained under specific pathogen-free conditions, housed in isolated vented cages, and allowed to acclimate for 1 week with standard chow diet. The animals were housed at Texas A&M University in accordance with the standards of the Guide for the Care and Use of Laboratory Animals and the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). The protocol of the animal study was approved by the Institutional Animal Care and Use Committee (IACUC), Texas A&M University. Rh30 cell lines (4 × 10⁶ cells) grown in RPMI media containing 10% FBS were detached, resuspended in 100 μL of phosphate-buffered saline with matrigel (BD Bioscience) (75:25), and implanted subcutaneously in the mice. When tumors reached about 200 mm³ size, the mice were randomized into control and treatment groups (4 animals each). Intratumoral injections of either in vivo grade Ad-IL24 or Ad-GFP (control) were administered every third day for and a daily dose of 2 × 10⁸ PFU suspended in 50 μL of PBS for a total of 5 days.

Tumor volumes and weights and body weight were determined; the tumor size was measured using Vernier calipers, and the tumor volume was estimated by the formula: tumor volume (mm³) = (L × W²) × ½, where L is the length and W is the width of the tumor.

Results for each treatment group were replicated (at least 3×) and expressed a means ± SE. Statistical comparisons of the treated groups versus a control for each treatment were determined using the Student t test.

Changes in gene expression in Rh30 cells after knockdown of NR4A1 and PAX3-FOXO1 or treatment with DIM-CpPhOH were determined by RNA-seq and comparisons with controls. Figure 1A illustrates the changes in gene expression in the three treatment groups and the overlap of common genes that were induced (6) or repressed (7). The most highly induced gene, IL24, was induced 27.9-fold after PAX3-FOXO1 knockdown, and therefore, we examined the relative expression of IL24 in ARMS tumors versus normal muscle (Fig. 1B) using publicly available databases (34). IL24 levels were significantly higher in normal muscle versus ARMS (Fig. 1B) and also higher in PAX3-FOXO1–negative versus PAX3-FOXO1–positive tumors (Fig. 1B). Furthermore, the expression of IL24 and NR4A1 is inversely related in ARMS and normal muscle tissue samples (Fig. 1C). IPA analysis of the changes in gene expression after PAX3-FOXO1 knockdown involved changes in multiple pathways as illustrated in Supplementary Fig. S1A.

RNA-seq analysis shows that both NR4A1 and its ligand (DIM-C-pPhOH), which acts as an NR4A1 antagonist, induce IL24 mRNA in Rh30 ARMS cells. Transfection of siNR4A1 or treatment with DIM-C-pPhOH induced IL24 protein levels in this cell line (Fig. 2A) and induction was also observed after treatment with DIM-C-pPhCO₂Me, another NR4A1 ligand that acts as a receptor antagonist (ref. 28; Fig. 2A). A similar approach was used in Rh18 (Fig. 2B) and Rh41 (Fig. 2C and Supplementary Fig. S1B) ARMS cells, and both siNR4A1 and NR4A1 ligands induced IL24 expression. We also observed that knockdown of PAX3-FOXO1 (siPF) induced IL24 protein expression (Fig. 2D), and these results are consistent with the RNA-seq data (Fig. 1A) and confirm that PAX3-FOXO1 suppresses IL24 expression in ARMS cells.

Figure 3A shows that NR4A1 knockdown or treatment with NR4A1 antagonists DIM-C-pPhOH and DIM-C-pPhCO₂Me induced IL24 gene expression in Rh30, Rh18, and Rh41 ARMS cells, and similar results were observed after knockdown of PAX3-FOXO1 (Fig. 3B). Using primers for two proximal regions of the IL24 promoter, knockdown of PAX3-FOXO1 resulted in recruitment of pol II (Fig. 3C), and this was consistent with induction of this gene. ChiPseq results suggest that PAX3-FOXO1 does not directly bind promoters (35), and induction of IL24 may be indirect and related to downregulation of a PAX3-FOXO1–regulated gene such as histone deacetylase 5 (HDAC5) (36, 37), which is known to suppress muscle-specific genes (38). Figure 3D shows that knockdown of PAX3-FOXO1 decreased HDAC5, and this was also observed in the RNA-seq results. Knockdown of HDAC5 by RNAi resulted in induction of IL24 (Fig. 3E) and overexpression of HDAC5 partially reversed induction of IL24 after PAX3-FOXO1 knockdown (Fig. 3F), suggesting that PAX3-FOXO1–mediated repression of IL24 is indirect and due to HDAC5.

The anticarcinogenic effect of IL24 in ARMS cells was investigated by overexpression studies, and Fig. 4A shows that overexpression of IL24 inhibited Rh30, Rh18, and Rh41 cell proliferation. IL24 overexpression induced activation of caspase-3, caspase-9, and PARP cleavage in Rh30 (Fig. 4B), Rh18 (Fig. 4C), and Rh41 (Fig. 4D) cells and also induced Annexin V staining in the three ARMS cell lines (Fig. 4E). IL24 also inhibited migration of ARMS cells (Fig. 4F), and these results were similar to those previously observed in ARMS cells transfected with siPAX3-FOXO1 (27, 32). Supplementary Fig. S2 shows by immunostaining of Rh30 cells after transfection with pCMV-6-IL24 that a large percentage of the cells expressed IL24.

Previous studies have characterized several other IL24 induced responses in diverse cancer cell lines, including increased phosphorylation of STAT1 and STAT3, activation or induction of stress/survival genes including Bax, PERK, p38, GADD45, and GADD34, and downregulation of the survival genes Bcl-2 and Bcl-XL (39–44). These responses were investigated by overexpression of IL24 in Rh30 (Fig. 5A), Rh18 (Fig. 5B), and Rh41 (Fig. 5C) cells, and all of the IL24-mediated effects previously reported in other cancer cell lines were also observed in ARMS cell lines. It has also been reported that IL24 expression downregulates Bcl-2 and Bcl-XL and induces Bax, GADD45, and GADD34, and these effects are p-38 dependent. Figure 5D shows that the p38 inhibitor SB203580 also inhibited these same IL24-induced gene products in Rh30 cells, whereas SB203580 alone did not alter expression of these genes (Fig. 5E).

Previous studies show that knockdown of PAX3-FOXO1 by RNAi inhibits ARMS cell growth, survival, and migration (32), and the role of IL24 in mediating these responses was investigated by transfecting cells with siPAX3-FOXO1 (which induces IL24) alone and in combination with IL24 knockdown by RNAi (siIL24). IL24 knockdown alone had minimal effects on RMS cell growth, apoptosis, and invasion (Supplementary Figs. S3A–S3C). Results summarized in Fig. 6A and B show that transfection of ARMS cells with siPAX3-FOXO1 inhibited cell growth and migration, whereas in ARMS cells cotransfected with siPAX3-FOXO1 plus siIL24, these inhibitory responses were significantly attenuated. Figure 6C confirms that siPAX3-FOXO1 resulted in knockdown of PAX3-FOXO1 and induction of IL24 in Rh30, Rh18, and Rh41 cells, and cotransfection with siIL24 reversed these responses. Using the same treatment protocol, we observed that transfection of siPAX3-FOXO1 induced caspase-3, PARP cleavage (Fig. 6D), and Annexin V staining (Fig. 6E), and in cells cotreated with siPAX3-FOXO1 plus siIL24 these same markers of apoptosis were significantly attenuated. The effects of siIL24 alone on ARMS cell proliferation, invasion, and induction of Annexin V staining were minimal, and we observed similar effects of siPF alone and in combination with siIL24 on caspase 3/PARP cleavage using a second set (2) of oligonucleotides (Supplementary Fig. S3).

We also investigated the effects of direct injection of adenoviral IL24 into ARMS tumors (from Rh30 cells) in athymic nude mice, and there was a dramatic decrease in tumor growth and weight (Fig. 7A and B), and IL24 mRNA and protein were highly expressed in tumors receiving the adenoviral IL24 (Fig. 7C). Thus, our functional studies on the in vivo and in vitro effects of IL24 were complementary. In summary, results of this study show that inactivation of PAX3-FOXO1 results in downregulation of HDAC5, resulting in depression of IL24 expression, which in turn exhibits tumor suppressor–like activity (Fig. 7D) and clinical potential for treating RMS patients.

NR4A1 regulates PAX3-FOXO1 gene expression in ARMS cells, and treatment with the NR4A1 antagonist DIM-C-pPhOH downregulated PAX3-FOXO1 expression, resulting in inhibition of ARMS cell and tumor growth. RNA-seq was used to investigate the common and divergent effects of NR4A1 and PAX3-FOXO1 knockdown and DIM-C-pPhOH treatment on expression of genes in Rh30 cells, and we observed induction of 6 and inhibition of 7 genes in common by all 3 treatments (Fig. 1A). The most dramatic response was observed for IL24 mRNA, which was repressed in ARMS tumors (compared with normal muscle) and inversely expressed with PAX3-FOXO1 in ARMS tumors (Fig. 1B and C), but was induced 2.0-, 2.0-, and 27.9-fold after treatment with DIM-C-pPhOH or knockdown or NR4A1 and PAX3-FOXO1, respectively, and these treatments also induced IL24 protein (Figs. 1 and 2).

IL24, which is also known as melanoma differentiation-associated gene 7, is a unique tumor suppressor cytokine that is a member of the IL24 subfamily. IL24 activates signaling through the type 1 and 2 IL24 receptors (IL20R), which consists of the IL20RA:IL20RB and IL22RA1:IL20RB heterodimeric receptors, respectively (45). The tumor suppressor–like activities of IL24 include inhibition of cancer cell growth and migration, induction of apoptosis, and inhibition of drug resistance, and these activities are observed in several different cancer cell lines through activation/repression of multiple genes/pathways (reviewed in refs. 39–44, 46).

PAX3-FOXO1 does not bind directly to gene promoters (35), suggesting a possible indirect mechanism for suppressing IL24. Results of our RNA-seq studies and previous reports identified HDAC5 as a PAX3-FOXO1–regulated gene (36, 37), and previous studies showed a role for HDAC5 as an inhibitor of muscle gene expression in RMS cells (38). Our results confirm expression of HDAC5 protein in Rh30 cells, and the decrease in HDAC5 expression after PAX3-FOXO1 knockdown and the linkage to IL24 induction was confirmed by the effects of HDAC5 knockdown which resulted in the induction of IL24 (Fig. 3). These results are consistent with the scheme illustrated in Fig. 7D. Examination of patient array data (Fig. 1B) coupled with the 27.9-fold induction of IL24 (RNA-seq) observed after knockdown of PAX3-FOXO1 suggested that the oncogenic activity of PAX3-FOXO1 may be due, in part, to suppression of IL24. Therefore, we examined the effects of IL24 overexpression in ARMS cells (Fig. 4) and also determined the relative contributions of IL24 to the tumor suppressor–like activity observed after knockdown of PAX3-FOXO1 (Fig. 6). Overexpression of IL24 in ARMS cells inhibited cell growth and migration, activated caspase-dependent PARP cleavage and Annexin V staining, confirming that the anticarcinogenic activities of IL24 observed in ARMS cells are similar to those reported in other solid tumors (39–44, 46). Overexpression of IL24 activates or suppresses multiple genes and pathways in cancer cells that contribute to its tumor suppressor–like activity. For example, IL24 induces activation (phosphorylation) of p38 (MAPK) and induces the growth arrest and DNA damage (GADD)-inducible genes GADD45 and GADD34, activates STAT1 and STAT3, increases the Bax/Bcl-2 ratio, and activates (phosphorylation) the stress gene protein kinase R-like endoplasmic reticulum kinase (PERK; refs. 39–48). We also observed these same IL24-dependent responses in ARMS cells and their inhibition by SB203580 confirmed that p38 activation also plays an important role in mediating the effects of IL24 in ARMS cells (Fig. 5).

Thus, the direct effects of IL24 in ARMS cells are similar to those observed in other cancer cell lines, and the contributions of IL24 to the tumor suppressor–like activities observed after PAX3-FOXO1 downregulation were further investigated by RNAi (Fig. 6). Knockdown of PAX3-FOXO1 by RNA interference inhibited growth, migration, and induced apoptosis in ARMS cells as previously reported (32); however, cotransfection with siIL24 significantly attenuated the siPAX3-FOXO1–mediated anticarcinogenic activity. Moreover, the effectiveness of IL24 against ARMS tumors was confirmed in in vivo studies where adenoviral IL24 administration potently decreased ARMS tumor growth and weight in an athymic nude mouse model (Fig. 7).

In summary, the results of both in vitro and in vivo studies demonstrate that PAX3-FOXO1 suppresses expression of IL24 via HDAC5, and knockdown of PAX3-FOXO1 or overexpression of IL24 demonstrates that this cytokine exhibits a broad spectrum of anticancer activities in ARMS that are similar to results observed in other solid tumors. The safety of adenoviral-delivered IL24 has been reported in a phase I clinical trial in patients with advanced tumors, and our results in ARMS cells and tumors suggest that IL24 therapy represents a novel treatment modality for ARMS patients. This approach may be particularly efficacious for this disease in light of the toxic side effects associated with current therapies for childhood cancers (13).

No potential conflicts of interest were disclosed.

Conception and design: A. Lacey, E. Hedrick, S. Safe

Development of methodology: A. Lacey, E. Hedrick, K. Mohankumar, M. Warren, S. Safe

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Lacey, E. Hedrick, K. Mohankumar, M. Warren, S. Safe

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A. Lacey, E. Hedrick, Y. Cheng, K. Mohankumar

Writing, review, and/or revision of the manuscript: A. Lacey, E. Hedrick, K. Mohankumar, S. Safe

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): K. Mohankumar, S. Safe

Study supervision: S. Safe

Other (performed experiments): K. Mohankumar

This study was funded by the NIH (P30-ES023512), the Sid Kyle Endowment, Texas AgriLife Research, and the Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation.

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.