Human RD (ATCC) is an embryonal RMS cell line derived from a 7-year-old Caucasian female previously treated with cyclophosphamide and radiation. RD is treatment refractory (26), has MYC amplification (27), mutations in NRAS and TP53 (28), and is sensitive to VCR (29). Rd76-9 is a murine derived embryonal RMS cell line, isolated from a methylcholanthrene-induced mouse RMS tumor in a female C57BL/6 mouse (30) and was provided by Dr. Tim Cripe (Nationwide Children’s Hospital, Columbus, OH). The small molecule compound RCM1 (2-[2-oxo-2-(thiophen-2-yl) ethyl]sulfanyl -4,6-di(thiophen-2-yl)pyridine-3-carbonitrile) was synthesized by Vitas-M Laboratory (95% purity). Vincristine Sulfate (VCR, NDC 61703-309-16) was purchased through in-house pharmacy at Cincinnati Children’s Hospital.

Bisphenol A glycerolate diacrylate, 4,4’-Trimethylenedipiperidine, 6-amino-1-hexanol, Oleic acid, polyethylenimine (PEI, Mn ~600) poly(ethylene glycol) (PEG, Mn ~2000), and Folic acid (FA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Lecithin (from Soybean) was purchased from Tokyo Chemical Industry Co., Ltd. (TCI). DyLight 650 or 800 NHS ester, carbodiimide hydrochloride (EDC), N- Hydroxysuccinimide (NHS), SnakeSkin™ Dialysis Tubing, and 10K MWCO were purchased from ThermoFisher Scientific (Waltham, MA, USA).

The amphiphilic poly-beta amino ester (aPBAE) nanoparticle backbone was synthesized via a Michael Addition. Briefly, the Bisphenol A glycerolate diacrylate was first mixed with 4,4’-Trimethylenedipiperidine in DMSO at 50°C for 24 hours. The mixture was then added to 6-amino-1-hexanol with the temperature increased to 90°C and held for another 24 hours. The polymer backbone was capped by a FA modified PEI. The FA and PEG modifications were processed via EDC/NHS coupling, as described previously (31, 32). To encapsulate RCM1, nanoparticle components were mixed with RCM1 in DMSO, then moved to an aqueous condition to allow DMSO to diffuse and the nanoparticles to assemble for 4 hours, followed by dialysis for 48 hours to remove DMSO and impurities. UV/Vis spectroscopy has been widely used to determine drug loading capacity using various solvents, including DMSO (33–35), so RCM1-encapsulated nanoparticles were characterized by using UV/Vis spectroscopy to determine the amounts of RCM1 according to the standard curve for estimation of encapsulation concentration.

C56Bl/6J mice were purchased from the Jackson laboratory. To generate the subcutaneous syngeneic murine model, 1x10⁶ Rd76-9 rhabdomyosarcoma cells were re-suspended in equal volumes of PBS : Matrigel (Corning) and were injected into the flanks of 8-12 weeks old C57Bl/6J mice. Animals were treated with VCR or saline (control) every 7 days via intraperitoneal injections. Animals were treated with RCM1 encapsulated nanoparticles or empty nanoparticles (control) every 48 hours intravenously via the eye vein. Tumors were measured using calipers, and volumes were calculated in cubic millimeters using 1 2 ( L × ( W 2 ) ) , where L is the largest diameter and W is the diameter perpendicular to L. Serum was collected from animals treated with empty nanoparticles, treated with 0.25mg/kg VCR alone, 8µg RCM1 nanoparticles alone or treated with combination of 0.25mg/kg VCR and 8µg RCM1 nanoparticles. Each animal study had 3-7 mice per group.

3x10⁵ Tumor cells per well were seeded in 6-well plates and allowed to grow for 24 hours or 48 hours. After 24 hours, tumor cells were then treated with RCM1, VCR or RCM1 + VCR for 24 hours. Trypan blue was used to exclude dead cells and viable cells were counted using a Hemocytometer. Experiments were performed in triplicate.

3x10⁵ Tumor cells per well were seeded on 24mm square cover slips in 6-well plates and allowed to grow for 24 hours. Tumor cells were then treated with RCM1, VCR or RCM1 + VCR for 24 hours. Tumor cells were then fixed and stained as previously described (36). To visualize the nucleus, Hoechst 33342 (ThermoFisher Scientific) was used as a counter stain. For quantification, 5 random fields were acquired per sample and quantified using ImageJ. To perform immunostaining of tumor tissue, paraffin embedded Rd76-9 subcutaneous tumor sections were stained as described previously (37). 5 Random fields per sample were acquired and quantified using ImageJ. Antibodies used for immunostaining were anti-Ki-67 (Invitrogen, MA5-14520), anti-phospho-histone H3 (Santa Cruz, sc-374669 (C-2)), anti-CD31 (R&D, AF3628), anti-Cleaved-Caspase 3 (R&D, MAB835), anti-CHAC1 (Novus Biologicals, OTI1E2).

For stable knockdown of CHAC1, Rd76-9 cells were transduced with GIPZ Chac1 shRNA (V3LMM_499697, Horizon). GIPZ non-silencing lentiviral shRNA (RHS4346, Horizon) was used as control. Lysis Buffer (Qiagen) and β-Mercaptoethanol were used to lyse cells in vitro. Taqman gene expression assays for Chac1 and Beta-Actin were purchased from ThermoFisher. qRT-PCR was performed as previously described (38, 39). Protein extracts were prepared as described (40, 41). Antibodies used for western blots were anti-CHAC1 (Sigma, AV42623), and anti-Vinculin (Cell Signaling, E1E9V).

Dissected Rd76-9 tumors from control (treated with empty nanoparticle and saline), 0.25mg/kg VCR, 8µg RCM1 and 0.25mg/kg VCR + 8µg RCM1 groups were snap frozen in liquid nitrogen 16 days post tumor inoculation. RNA was extracted from bulk tumor samples and were sent for sequencing. Quality of RNA was determined using Fragment Analyzer with an average RQN for all samples of 9.7. RNA libraries were prepared for all samples using Illumina TruSeq Stranded mRNA Prep to generate poly-A enriched, non-stranded RNA libraries. Sequencing was performed using NoveSeq6000 with an estimated 30 million read per sample. Reads were aligned to the GRCm38 mouse genome and quantified using an index transcriptome version of GRCm38 using Kallisto using standard settings (42). Raw counts were normalized using DESeq2 (43). Differential gene expression between conditions was performed using DEseq2 which uses a negative binomial model for each gene. The Wald test was used for hypothesis testing when comparing the two groups. All p-values attained were corrected for multiple testing using the Benjamini and Hochberg method which is the default method in DEseq2. In the standard DESeq2 algorithm, alpha for false-discovery rate is set to 0.1 by default. Heatmap was generated using the pheatmap R package and the volcano plot was generated using the EnhancedVolcano R package (https://github.com/kevinblighe/EnhancedVolcano). Venn diagrams for differentially expressed genes were created using AltAnalyze (44).

The Student t test, one-way ANOVA, one-way ANOVA RM and two-way ANOVA RM were used to determine statistical significance. P-values <0.05 were considered significant. Values were shown as mean SD. All statistical analyses were obtained using GraphPad Prism.

All animal studies were approved by Cincinnati Children’s Research Foundation Institutional Animal Care and Use Committee and covered under our animal protocol (IACUC2022-0041). The Cincinnati Children’s Research Foundation Institutional Animal Care and Use Committee is an AAALAC and NIH accredited institution (NIH Insurance #8310801).

To determine the half maximal inhibitory concentrations (IC₅₀) of VCR and RCM1, dose response curves were generated for each single agent. After 24 hours, both VCR and RCM1 reduced the number of Rd76-9 murine RMS cells in a dose-dependent manner ( Figure 1A ). Using the dose response curves for VCR and RCM1, we selected IC₅₀ concentrations of 2.1nM for VCR and 8.7µM for RCM1 to use in combination treatment. Using these concentrations, the dual treatment with low doses of VCR and RCM1 synergized to reduce the number of Rd76-9 tumor cells after 24 hours compared to the single agents and control ( Figure 1B , S1A ). To further characterize this novel dual-drug therapy, we performed immunostaining of tumor cells using various markers. Since both RCM1 and VCR are cell-cycle specific anti-tumor agents, we first explored the effects of the combination therapy on tumor cell proliferation. Cell proliferation was assessed using immunostaining for Ki-67, a general proliferation marker, and phospho-histone H3 (PH3), a mitosis specific marker. After 24 hours, the combination therapy synergized to reduce the number of Ki-67-positive cells compared to each single agent ( Figure 1C ). Also, the two agents synergized to reduce the number of mitotic cells after 24 hours ( Figure 1D ). To characterize the effects of this combination therapy on cell death, we performed immunostaining for Cleaved-Caspase 3, a well-known marker for apoptosis. After 24 hours, this combination therapy synergized to increase the number of apoptotic RMS cells, suggesting that the dual-drug therapy with low doses of RCM1 and VCR induces tumor cell death with higher efficiency than single agents alone ( Figure 1E ). These results demonstrate that RCM1 in combination with VCR synergize to inhibit proliferation and increase apoptosis of murine RMS tumor cells in vitro.

Since the dual treatment with low doses of RCM1 and VCR was effective in murine RMS cells, we investigated the effects of this combination therapy using a human derived RMS cell line, RD. To determine an optimal treatment dose, we generated the dose response curves for each drug and selected 3.5nM for VCR and 3.0µM for RCM1 to use in combination therapy ( Figure 2A ). Consistent with the mouse data, this dual therapy with low doses of VCR and RCM1 synergized to reduce the number of human RD tumor cells more efficiently compared to single agents ( Figure 2B , S1B ). Next, we assessed tumor cell proliferation by performing immunostaining for Ki-67 and PH3. Similar to the murine RMS cells, the combination therapy synergized to reduce the number of Ki-67-positive proliferating tumor cells ( Figure 2C ) and PH3-positive mitotic tumor cells compared to single agents ( Figure 2D ). Importantly, combination of low doses of VCR and RCM1 synergized to increase the number of apoptotic tumor cells as shown by immunostaining for Cleaved-Caspase 3 ( Figure 2E ). Taken together, these data suggest that the RCM1 and VCR combination therapy can be a promising effective therapy for RMS in vivo.

To determine the half maximal inhibitory concentrations (IC₅₀) of VCR and RCM1 as single agents in vivo, Rd76-9 murine RMS cells were injected subcutaneously into the flanks of C57Bl/6J mice to create a syngeneic RMS model. Mice were treated with VCR at days 7 and 14 after tumor cell inoculation ( Supplementary Figure S2A ). Treatment with VCR reduced tumor volume in a dose-dependent manner ( Supplementary Figure S2B ). Using tumor volumes measured at day 16, we generated a dose response curve to determine the IC₅₀ concentration for VCR (0.25mg/kg) that later would be used for the combination therapy with RCM1 ( Supplementary Figure S2C ). Since VCR can be toxic (5, 8, 9), the animals were weighed at harvest to compare body weights. There were no differences in body weights between groups, suggesting that these doses of VCR are non-toxic ( Supplementary Figure S2D ). Treatment with VCR reduced tumor cell proliferation and mitosis in a dose-dependent manner as demonstrated by immunostaining of RMS tumor sections for Ki-67 and PH3 ( Supplementary Figures S2E, F ). We next analyzed the tumor-associated angiogenesis using immunostaining for CD31, a marker of endothelial cells. There was no difference in endothelial coverage between VCR treated and control groups ( Supplementary Figure S2G ). The effect of VCR on apoptosis of RMS cells was not significant as shown by immunostaining for Cleaved-Caspase 3 ( Supplementary Figure S2H ).

Even though RCM1 has been shown previously to be a promising anti-tumor agent (19), the pre-clinical and clinical delivery of RCM1 has some limitations since RCM1 is a highly hydrophobic compound. Nanoparticles have been shown to encapsulate highly hydrophobic anti-tumor agents and are used as a vehicle for drug delivery in patients (45–49). Since nanoparticles can be administered intravenously (i.v.), we tested whether hydrophobic RCM1 compound can be delivered to RMS tumors using nanoparticles. An amphiphilic poly-beta amino ester (aPBAE) backbone was used to generate nanoparticles. The amphiphilic polymers can self-assemble with cargo in an aqueous condition to form a hydrophobic core and a hydrophilic surface, which can keep the nanoparticle stable (50, 51). Next, lecithin was added to improve nanoparticle stabilization with PEG ligands (52, 53) ( Figure 3A , upper panel). Finally, the folic acid molecules were incorporated into nanoparticles to increase their specificity towards tumor cells that overexpressed folate receptor (31, 54, 55) ( Figure 3A , bottom panel). Nanoparticles were labeled with DyLight 800 to visualize their recruitment into RMS tumors in mice using IVIS. Compared to nanoparticles without folic acid (NP), the nanoparticles with folic acid (NP^FA) demonstrated the most efficient localization to the RMS tumors ( Figure 3B ), with higher average radiant efficiency ( Figure 3C ). Based on high tumor specificity, we used the NP^FA nanoparticles thereafter. The sizes of NP^FA were measured and used to calculate the hydrodynamic average diameter of NP^FA as being 160.67nm ( Figure 3D ). The surface charge of NP^FA was 38.13mV ( Figure 3E ), which is a suitable surface charge for tumor cell targeting (32, 56, 57). Next, RCM1 was encapsulated into NP^FA nanoparticles. The UV/Vis spectrophotometry was used to determine the presence of RCM1 in the nanoparticles based on the RCM1 absorbance at 310nm and 395nm ( Figure 3F ). These UV/Vis spectra were used to generate a standard concentration curve and to calculate the amount of RCM1 encapsulated in the nanoparticles ( Figure 3G ). To test the anti-tumor efficacy of RCM1-NP^FA, the tumor-bearing mice were treated with either control Empty-NP^FA or RCM1-NP^FA delivered i.v. every other day for 5 treatments ( Figure 3H ). Compared to Empty-NP^FA, RCM1-containing nanoparticles reduced tumor volume in a dose-dependent manner ( Figures 3I, J ). A dose response curve was generated to determine the IC₅₀ concentration for RCM1 in the nanoparticles, which was 8µg per injection ( Supplementary Figure S3A ). The animals were weighed at harvest. No differences in the body weights were found between mice treated with RCM1-NP^FA and control Empty-NP^FA ( Supplementary Figure S3B ). Treatment with RCM1-NP^FA reduced tumor cell proliferation in a dose-dependent manner as shown by immunostaining for Ki-67 and PH3 ( Supplementary Figures S3C, D ). Unlike treatment with VCR in vivo, the highest dose of RCM1-NP^FA reduced tumor-associated angiogenesis which was quantified as CD31⁺-vessel coverage within RMS tumors in mice ( Supplementary Figure S3E ). RCM1-NP^FA treatment did not induce tumor cell apoptosis at any concentration used ( Supplementary Figure S3F ). Collectively, these data demonstrate that both VCR and RCM1-NP^FA reduce tumor burden as single agents.

Using the IC₅₀ concentrations determined for both VCR (0.25mg/kg) and RCM1 (8µg/injection) in vivo ( Supplementary Figure S2C and Supplementary Figure S3A respectively), we next investigated the combinatorial effects of these drugs. The Rd76-9 RMS cells were inoculated subcutaneously into the flanks of mice and allowed tumor to grow for 7 days. On day 7, the tumor-bearing mice were treated with the first dose of both VCR and RCM1. VCR was administered once a week thereafter, and RCM1-NP^FA were administered every other day ( Figure 4A ). The combination of both drugs had shown the best efficacy in reducing tumor volumes ( Figure 4B ). There were no differences in body weights between control and all experimental groups of tumor-bearing mice ( Supplementary Figure S4A ). Blood was collected and a liver metabolic panel was analyzed from the serum of these mice. No differences were seen in the concentrations of albumin, total protein, ALP, AST, ALT, GGT and bilirubin ( Supplementary Figures S4B-D ). Taken together with the body weights, these results suggest that the combination therapy is non-toxic.

To characterize the effects of combination therapy on RMS tumors, the tumor sections were immunostained for Ki-67, PH3, CD31 and Cleaved-Caspase 3. Consistent with the in vitro data, the combination therapy synergized to reduce the number of proliferating cells in RMS tumors ( Figures 4C, D ). The combination therapy also reduced tumor-associated angiogenesis ( Figure 4E ). When assessing the cell death through Cleaved-Caspase 3 staining, we found that neither single agent at IC₅₀ concentrations were able to increase tumor cell death. However, the combination therapy with non-toxic doses of VCR and RCM1-NP^FA synergized to increase tumor cell apoptosis ( Figure 4F ). Our results suggest that the nanoparticle delivery of RCM1 in combination with the low doses of VCR synergizes to improve efficacy of both drugs.

To determine the molecular mechanism of synergistic effect after dual treatment with RCM1-NP^FA and VCR, the RNA-seq was performed to compare gene signatures of RMS tumors dissected from experimental and control groups of mice. RNA-seq analysis identified 59 differentially expressed genes in combination treated mice compared to vehicle ( Figure 5A ). Of which, ChaC Glutathione Specific Gamma-Glutamylcyclotransferase 1 (Chac1) was identified as one of the most downregulated genes in combination treated rhabdomyosarcoma tumors compared to vehicle-treated tumors ( Figures 5A, B ). Additionally, out of the 59 total differentially expressed genes in combination treated tumors, Chac1 was one of the two genes that is uniquely downregulated only in combination treated tumors, but not in either single drug treated tumors ( Figure 5C ). To verify the RNAseq data, we performed qRT-PCR using bulk RNA from dissected tumors. Chac1 mRNA was significantly decreased only in the tumors treated with combination therapy ( Figure 5D ). To further support our RNA-seq data, we examined the protein levels of CHAC1 after combination therapy in RMS tumors in vivo. Consistent with our RNA-seq and qRT-PCR data, the number of tumor cells expressing CHAC1 was reduced only after VCR+RCM1-NP^FA combination treatment, but not after either single agent ( Figure 5E ). Using TCGA datasets, we have also demonstrated that the high expression of CHAC1 was associated with a worse prognosis in sarcoma patients ( Figure 5F ), suggesting that CHAC1 can play a role in RMS tumorigenesis.

To determine the role of CHAC1 in RMS tumor cells, we generated a shRNA lentivirus against Chac1 (shChac1) and transduced Rd76-9 cells to inhibit Chac1 expression in vitro. CHAC1 protein levels in shChac1 RMS tumor cells were decreased compared to non-targeting control, confirming efficient Chac1 depletion ( Figure 6A ). Depletion of CHAC1 decreased the numbers of viable Rd76-9 tumor cells in culture compared to control tumor cells ( Figures 6B, C ). Immunostaining of Rd76-9 cells for CHAC1, Ki-67, PH3 and Cleaved-Caspase 3 protein demonstrated that the depletion of CHAC1 decreased the numbers of Ki-67⁺ and PH3⁺ proliferating tumor cells ( Figures 6D-F ) as well as increased the number of tumor cells undergoing apoptosis ( Figure 6G ). Taken together, the knockdown of CHAC1 in RMS tumor cells decreased cell proliferation and increased apoptosis, recapitulating the effects of the combination therapy with VCR and RCM1.