EWS-FLI1-regulated Serine Synthesis and Exogenous Serine are Necessary for Ewing Sarcoma Cellular Proliferation and Tumor Growth

Abstract

Despite a growing body of knowledge about the genomic landscape of Ewing sarcoma, translation of basic discoveries into targeted therapies and significant clinical gains has remained elusive. Recent insights have revealed that the oncogenic transcription factor EWS-FLI1 can impact Ewing sarcoma cellular metabolism, regulating expression of 3-phosphoglycerate dehydrogenase (PHGDH), the first enzyme in de novo serine synthesis. Here, we have examined the importance of serine metabolism in Ewing sarcoma tumorigenesis and evaluated the therapeutic potential of targeting serine metabolism in preclinical models of Ewing sarcoma. We show that PHGDH knockdown resulted in decreased Ewing sarcoma cell proliferation, especially under serine limitation, and significantly inhibited xenograft tumorigenesis in preclinical orthotopic models of Ewing sarcoma. In addition, the PHGDH inhibitor NCT-503 caused a dose-dependent decrease in cellular proliferation. Moreover, we report a novel drug combination in which nicotinamide phosphoribosyltransferase (NAMPT) inhibition, which blocks production of the PHGDH substrate NAD⁺, synergized with NCT-503 to abolish Ewing sarcoma cell proliferation and tumor growth. Furthermore, we show that serine deprivation inhibited Ewing sarcoma cell proliferation and tumorigenesis, indicating that Ewing sarcoma cells depend on exogenous serine in addition to de novo serine synthesis. Our findings suggest that serine metabolism is critical for Ewing sarcoma tumorigenesis, and that targeting metabolic dependencies should be further investigated as a potential therapeutic strategy for Ewing sarcoma. In addition, the combination strategy presented herein may have broader clinical applications in other PHGDH-overexpressing cancers as well.

Figure 1.

EWS-FLI1 knockdown affects PHGDH expression and serine synthesis. A, Immunoblot analysis of EWS-FLI1 and PHGDH expression in the indicated cell lines expressing the specified FLI1 siRNAs or a non-targeting control siRNA (siNT). Actin was used as a loading control. B, Schematic of the incorporation of glucose-derived ¹³C into 3-PG and serine. C, Detection of labeled 3-PG and serine in the indicated cell lines following transfection with the specified siRNAs. Data represent the mean + SD of 3 independent experiments. * p<0.05, ** p<0.005 compared to siNT.

Figure 2.

PHGDH knockdown inhibits ES proliferation and tumor growth. A, Immunoblot analysis of PHGDH expression in the indicated cell lines expressing the specified lentiviral control or PHGDH shRNAs. Actin was used as a loading control. B, Cellular proliferation was monitored in an Incucyte FLR. The indicated cell lines were assayed in serine/glycine-free RPMI 1640 media supplemented with 10% dialyzed FBS (No Ser). Data represent the mean ± SD of a representative experiment. C, Average tumor volume over time of mice orthotopically injected with TC32 or TC71 stable cell lines bearing the indicated shRNAs. Data represent the mean ± SEM (8 mice/group). ** p<0.005 vs. shControl. D, Immunoblot analysis of PHGDH in endpoint tumor lysates derived from the specified cell lines expressing the indicated control or PHGDH shRNAs. Actin was used as a loading control.

Figure 3.

The PHGDH inhibitor NCT-503 inhibits sarcoma cellular proliferation. A, Schematic of the serine synthesis pathway and NCT-503 inhibition of PHGDH. B, Cellular proliferation in response to NCT-503 or an inactive control was monitored in an Incucyte FLR. The indicated cell lines were assayed in RPMI 1640 media supplemented with 10% dialyzed FBS, with or without serine and glycine (No Ser/Gly). Data represent the mean ± SD of a representative experiment.

Figure 4.

Combined inhibition of PHGDH and NAMPT reduces ES proliferation and tumor growth. A, Schematic of the serine synthesis pathway and NCT-503/GNE-618 inhibition of PHGDH. B, Cellular proliferation in response to NCT-503 and/or GNE-618 was monitored in an Incucyte FLR. The indicated cell lines were assayed in RPMI 1640 media supplemented with 10% dialyzed FBS. Data represent the mean ± SD of a representative experiment. C, Average tumor volume over time of mice orthotopically injected with TC32 cells. Mice received the specified treatments, as indicated. Data represent the mean ± SEM (10 mice/group). * p<0.05, ** p<0.005.

Figure 5.

Exogenous serine is required for optimal ES proliferation and tumor growth. A, Cellular proliferation was monitored in an Incucyte FLR. Cell lines were assayed in RPMI 1640 media supplemented with 10% dialyzed FBS, with (Control) or without serine and glycine (No Ser/Gly). Data represent the mean ± SD of a representative experiment. B, Average tumor volume over time of the indicated orthotopically injected cell lines in mice fed a control diet or a serine and glycine deficient diet. Data represent the mean ± SEM (8 mice/group). * p<0.05, ** p<0.005 vs. Control Diet.