NAPRT expression and epigenetic regulation in pediatric rhabdomyosarcoma as a potential biomarker for NAMPT inhibition

Abstract

Purpose: New treatments are needed to improve survival in children with rhabdomyosarcoma (RMS). NAD⁺ biosynthesis, regulated by the enzymes NAPRT and NAMPT, represents a metabolic vulnerability due to high NAD⁺ turnover in cancers. Although NAMPT inhibitors (NAMPTi) show preclinical promise, clinical translation has been limited by toxicity and the lack of predictive biomarkers. Here, we evaluated NAPRT expression in RMS and its potential as an actionable biomarker to guide NAMPTi therapy.

Experimental design: NAPRT promoter methylation, transcript levels, and protein expression were assessed in RMS cells, PDXs, and primary tumors (n=109) from the Children's Oncology Group. In vitro sensitivity to NAMPTi was tested in molecularly diverse and isogenic RMS cell lines, examining the role of NAPRT expression in mediating cytotoxicity and the ability of nicotinic acid (NA) to rescue viability. In vivo efficacy was assessed using NAPRT-isogenic orthotopic xenograft models.

Results: NAPRT promoter hypermethylation was found in a subset of RMS models and patient samples. Immunohistochemistry showed loss of NAPRT protein in 30-40% of tumors, defined as <1% tumor cell staining. Methylation modestly correlated with protein expression. NAPRT-silenced cells were highly sensitive to NAMPTi, driven by NAD⁺ depletion and not reversible with NA. In vivo, NAMPTi induced significant tumor regression, which was not abrogated with NA administration in NAPRT-silenced models.

Conclusions: NAPRT loss occurs in a subset of RMS, offering a potential strategy to expand the therapeutic window of NAMPTi. Further research is needed to understand NAPRT regulation and optimize biomarker assay strategies for use in future clinical trials.

Keywords: NAMPT; NAPRT; metabolism; rhabdomyosarcoma.

Figure 1.. RMS models display loss of NAPRT and sensitivity to OT-82 that is not rescued through the Preiss-Handler pathway.

A. Schematic of NAD⁺ biosynthesis pathways and “protective rescue” paradigm in tumors with loss of NAPRT. B. Plot illustrating the methylation of NAPRT across various cancer types. NAPRT is hypermethylated in RMS. C. Plot illustrating NAPRT protein expression across various cancer types. Loss of NAPRT protein expression is observed in RMS. D. Protein expression of NAPRT in four RMS models by western blot. E-G. Cell viability in NAPRT-silenced RD (E) and RH41 (F) cells treated with increasing concentrations of OT-82 for 6 days with or without 10 μM NA. G-H. Cell viability in NAPRT-expressing RH28 (G) and RH30 (H) treated with increasing concentrations of OT-82 for 6 days with or without 10 μM NA. I-L. Total NAD⁺ quantification post-24 hours in RD (I), RH41 (J), RH28 (K) and RH30 (L) treated with respective IC₅₀ concentrations with or without 10 μM NA. The data is plotted as mean with error bars indicating SEM.*p<0.05; ***p<0.001; ****p<0.0001; n.s., not significant. NAPRT, nicotinate phosphoribosyltransferase; NAMPT, nicotinamide phosphoribosyltransferase; NA, nicotinic acid; GAPDH, Glyceraldyde3phosphate dehydrogenase.

Figure 2.. Generation and functional validation of NAPRT isogenic models.

A-D. Endogenously NAPRT-silenced RH41 cells transfected to ectopically express NAPRT (RH41^NAPRT+)^. A. Protein expression of NAPRT, NAMPT, QPRT, and Vinculin in RH41 parental and RH41^NAPRT+. B-C. Cell viability in RH41 (NAPRT-silenced) (B) and RH41^NAPRT+ (C) treated with increasing concentrations of OT-82 for 4 days with or without 10 μM NA functionally validating the exogenous expression of NAPRT D. Total NAD⁺ quantification post-24 hours in RH41 (NAPRT-silenced) and RH41^NAPRT+ treated with respective IC₅₀ concentration with or without 10 μM NA. E-H. CRISPR/Cas9-mediated knockout (KO) of NAPRT in endogenously NAPRT-expressing RH30 cells (RH30 NAPRT+). E. Protein expression of NAPRT, NAMPT, QPRT, and Vinculin in RH30 parentals and RH30 NAPRT-. F-G. Cell viability in RH30 models treated with increasing concentrations of OT-82 for 4 days with or without 10 μM NA functionally validating the KO of NAPRT H. Total NAD⁺ quantification post-24 hours in RH30 models treated with IC₅₀ concentrations of OT-82 with or without 10 μM NA. The data is plotted as mean with error bars indicating SEM. ****p<0.0001; n.s., not significant.

Figure 3.. NAMPTi induces early to late-phase apoptosis that is reversible with NA in NAPRT-expressing RMS models.

A-B. Cell cycle arrest after 48 hours of treatment with 0.1% DMSO, 10 μM NA, 10 nM OT-82, or the combination of OT-82 and NA in RH30 (NAPRT+) (A) and RH41 (NAPRT-) (B). C. Protein expression of Cleaved PARP1, PARP1, NAPRT, and Beta-actin on RH30 (NAPRT+) and RH41 (NAPRT-) treated with DMSO, 10 nM OT82, or combination of OT-82 and NA for 48 hours. D-G. Annexin-V/PI analysis after 48 hours of treatment of 0.1% DMSO, 10 uM NA, 10 nM OT-82, or the combination of OT-82 and NA on RH30 (NAPRT+) (D-E) and RH41 (NAPRT-) (F-G). The data represents mean with error bars as SEM. ****p<0.0001; n.s.,= not significant.

Figure 4.. OT-82 leads to delayed tumor growth, prolonged survival and decreased NAD⁺ levels that is not rescued with NA in NAPRT-silenced xenograft models.

A-F. Tumor growth curves (A-C), Kaplan-Meier plots (D-F), and body weight plots (G-I) of FOX SCID mice bearing RH30 (NAPRT+) (A, D, G), RH41 (NAPRT-) (B, E, H), RH41^NAPRT+ (C, F, I) following treatment with 30% cyclodextrin vehicle, 25 mg/kg OT82, and 25 mg/kg NAMPTi in combination with 25 mg/kg NA. J-K. Total NAD⁺ levels in tumors harvested from RH41 (NAPRT-) (J) and RH41^NAPRT+ (K) models treated with 1 cycle of 30% cyclodextrin vehicle, 25 mg/kg OT82, and 25 mg/kg NAMPTi in combination with 25 mg/kg NA. L-M. Total NAD⁺ levels in contralateral normal gastrocnemius muscle harvested from RH41 (NAPRT-) (L) and RH41^NAPRT+ (M) treated with 1 cycle of 30% cyclodextrin vehicle, 25 mg/kg OT-82, and 25 mg/kg NAMPTi in combination with 25 mg/kg NA. For J-M, each treatment included 3–5 biological replicates, with 2 technical replicates per biological replicate. The data is plotted as mean with error bars indicating SEM. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; n.s., not significant.

Figure 5.. Loss of NAPRT expression in RMS patient samples.

A. Summary of NAPRT IHC score in pilot tissue microarray (TMA) (n = 33 individual patients) using semi-quantitative scoring system based on percentage of positive tumor cells. B-C. Summary of NAPRT IHC score in patient derived xenograft (PDX) TMA (n = 29 individual RMS PDX) by percentage positive tumor cells (B) and by histology (C). D-F. Summary of NAPRT IHC score in Children’s Oncology Group (COG) TMAs (n = 109 individual patients) by percent positive tumor cells (D), fusion status (E), and histology (F). G. Heatmap representing methylation beta values of 12 CpG probes corresponding to the NAPRT1 gene (x-axis) of 99 pediatric RMS patients from COG TMA, stratified by fusion status (y-axis). H. Heatmap of 22 pediatric RMS patients with corresponding IHC and methylation data, stratified by total protein loss (top; IHC score = 0) or the presence of NAPRT protein (bottom; IHC score ≥1). I. Inverse relationship between RNA expression (TPM) and methylation beta values at the transcriptional start site (TSS) 200 and 1500 of 31 RMS patients (r² = 0.38). J. Graph showing a negative correlation trend of beta values at TSS200 and TSS1500 regions with NAPRT IHC score for 22 patients in (H). K. Correlation between COG TMA NAPRT IHC score and RNA expression of four patients (r² = 0.99). L. Representative NAPRT IHC (top) and hematoxylin and eosin (H&E; bottom) images in patients with corresponding IHC and transcriptomic data. In (A, B, D), percentage positive score: (Score = % Positive Cells) 0 = <1%; 1+ = 1–25%; 2+ = 25–50%; 3+ = 50–75%; 4+ = 75–100%.