Coordinated Targeting of S6K1/2 and AXL Disrupts Pyrimidine Biosynthesis in PTEN-Deficient Glioblastoma

Abstract

Intrinsic resistance to targeted therapeutics in PTEN-deficient glioblastoma (GBM) is mediated by redundant signaling networks that sustain critical metabolic functions. Here, we demonstrate that coordinated inhibition of the ribosomal protein S6 kinase 1 (S6K1) and the receptor tyrosine kinase AXL using LY-2584702 and BMS-777607 can overcome network redundancy to reduce GBM tumor growth. This combination of S6K1 and AXL inhibition suppressed glucose flux to pyrimidine biosynthesis. Genetic inactivation studies to map the signaling network indicated that both S6K1 and S6K2 transmit growth signals in PTEN-deficient GBM. Kinome-wide ATP binding analysis in inhibitor-treated cells revealed that LY-2584702 directly inhibited S6K1, and substrate phosphorylation studies showed that BMS-777607 inactivation of upstream AXL collaborated to reduce S6K2-mediated signal transduction. Thus, combination targeting of S6K1 and AXL provides a kinase-directed therapeutic approach that circumvents signal transduction redundancy to interrupt metabolic function and reduce growth of PTEN-deficient GBM.

Significance: Therapy for glioblastoma would be advanced by incorporating molecularly targeted kinase-directed agents, similar to standard of care strategies in other tumor types. Here, we identify a kinase targeting approach to inhibit the metabolism and growth of glioblastoma.

Figure 1

Combination S6K1 and AXL targeting reduces PTEN-deficient glioblastoma tumors in vivo. A, Combination S6K1 and AXL inhibitors, LY-2584702 and BMS-777607, respectively, reduced subcutaneous tumor volumes. Nude mice were implanted with 1 million U87MG cells and then treated twice daily (BID) with inhibitors by oral gavage on a 5 + 2 schedule. Animals were euthanized once tumors measured 2,000 mm³. B, Brain lysates from NSG mice treated with 30 mg/kg of LY-2584702, 30 mg/kg of BMS-777607, a combination of LY + BMS, or DMSO for 3 days BID. Reduced rpS6 phosphorylation was detected in intact brains collected an hour after final dose. C, In a time course of combination treatment, rpS6 phosphorylation was reduced in B6 mouse intact brain lysates as early as 30 minutes after final treatment. LY-2584702 and BMS-777607 reduced rpS6 phosphorylation in the brain, even in the absence of tumor. D, The S6K1 inhibitor, LY-2584702, in combination with the AXL inhibitor, BMS-777607, reduced tumor growth in orthotopically implanted U87MG-GFP-Luciferase cells. Growth of intracranial tumors was determined from bioluminescence imaging before treatment began and again after a month of BID 5 + 2 oral gavage inhibitors (12.5 mg/kg) or vehicle control (n = 5). E, Tumor growth of D shown as the change in luciferase counts from pretreatment to 32 days after treatment initiation. P value = 0.015 (t test). F, Histological staining of phospho-rpS6 (S235/6) of tumor sections from whole excised brains of mice treated with DMSO or LY + BMS combination. The staining intensity of these samples were graded 1–3 in a blinded fashion. Scale bar, 200 μm.

Figure 2

PTEN-deficient patient-derived sphere growth is inhibited by combination S6K1 and AXL inhibition. A, JHH136 patient-derived GBM were treated for 72 hours with 10 μmol/L LY-2584702 and/ or 10 μmol/L BMS-777607. Phase contrast images show reduced sphere size with inhibitor combination compared with single inhibitor treatment and vehicle control (n = 3). Scale bar, 100 μm. B, Quantification of spheres in three independent cultures as in A are shown. P values from t tests indicate combination efficacy compared with single agent controls. C, Mayo59 treated with 10 μmol/L LY-2584702 and/or 10 μmol/L BMS-777607 for 7 days. Combination treatment reduced sphere size and induced an altered morphology (n = 3). Scale bar, 200 μm. D, Quantification of images in C shows significant reduction in sphere size in combination-treated spheres. E, Symptom-free survival of mice bearing intracranial tumors of Mayo59 PDX treated with vehicle or a combination of LY-2584702 and BMS-777607 together. Survival curve shows that animals gained a sustained delay in disease from combination treatment (n = 10 per group). F, Cox proportional hazard ratio analysis for mice in E. Log-rank score for combination treatment is below 0.002.

Figure 3

S6K1 and AXL inhibitors counteract pyrimidine biosynthesis in PTEN-deficient GBM. A, Steady state metabolite abundance in LN229 GBM transfected with siPTEN and then treated with vehicle control or combination S6K1 (LY-2584702, 10 μmol/L) and AXL (BMS-777607, 10 μmol/L) inhibitors for 5 hours (n = 4). B, Detail of nucleotide and their precursor metabolites from A. C, log₂ fold change of [U]-¹³C glucose labeled metabolites in U87MG GBM pretreated for 3 hours with 10 μmol/L LY-2584702 and 10 μmol/L BMS-777607 vs. vehicle control at 60 and 300 minutes after addition of ¹³C-glucose (n = 4). Statistically significant (>1.5 fold) metabolites are highlighted. D, JHH136 spheres were treated with inhibitors (10 μmol/L each) for 72 hours for western blot analysis. LY-2584702, S6K1 inhibitor, reduces phosphorylation of rpS6 and CAD leading to sustained impairment of pyrimidine synthesis and cell growth. Treatment with BMS-777607 increases H2A.X phosphorylation at Serine 139, indicating an increase in double-stranded DNA breaks. E, Mayo59 spheres were treated as in D. S6K1 inhibition reduces phosphorylation of rpS6 and CAD, impairing pyrimidine synthesis. DNA damage is evident in combination S6K1 and AXL inhibition.

Figure 4

Targeting requirements in the S6K1/S6K2 network. A and B, S6 phosphorylation was reduced only upon genetic silencing of both S6K1 and S6K2 in LN229 GBM cells. C, Combined inactivation of S6K1 and S6K2 via sgRNA and siRNA was required to silence S6K signaling in PTEN-deficient GBM U87MG-GFP-Luc. D, DepMap correlation shows dependence on S6K1 and S6K2 when PTEN is mutated. E, PTEN inactivation caused an induction of S6K signaling seen as an increase in the phosphorylation of rpS6. This increase can only be abrogated by combination inactivation of both S6K1 (exon 5) and S6K2 (exon 9). F, sgNT and sgS6K1 GBM LN229 cells were treated with 10 μmol/L LY-2584702 or 10 μmol/L PF-4708671 for 3 hours. Inhibitor suppression of phospho-rpS6 was modestly improved by knockout of S6K1. G, sgNT and sgS6K2 GBM LN229 cells were treated as in F. Inhibitor suppression of phospho-rpS6 was significantly enhanced by S6K2 knockout. H, sgNT and sgS6K2 (exon 9) LN229 cells transfected with siNT or siPTEN for 72 hours were incubated with S6K1 inhibitors for 3 hours. PTEN inactivation induced phospho-rpS6, which was blocked by the combination of sgS6K2 and an S6K1 inhibitor.

Figure 5

S6K2 and AXL have an effector—regulator relationship. A, GSEA identified an AXL response signature in RNA sequence analysis of sgS6K2 LN229 GBM cells. B and C, sgNT or sgS6K2 (exon 9) GBM U87MG-GFP-Luc (B) or LN229 (C) cells were stimulated with 300 nmol/L Phosphatidylserine (PtdSer) and/or 400 ng/mL Gas6 for 45 minutes as indicated. AXL autophosphorylation of pY702 was increased in the absence of S6K2. D, KiNativ analysis represented as a dendrogram (CORAL kinome tree) in which kinase inhibition in combination-treated shPTEN LN229 cells compared with vehicle is expressed as increased node size, branch intensity, and node intensity. Combination treatment with 10 μmol/L LY-2584702 and 10 μmol/L BMS-777607 for 3 hours specifically targeted S6K1, AXL, and MET. E, Model showing S6K2 as both a signaling effector and a feedback regulator of AXL. F, siPTEN LN229 sgNT or sgS6K2 cells were treated for 3 hours with 10 μmol/L LY-2584702 and/or 10 μmol/L BMS-777607 as indicated. S6K2 genetic deletion increased AXL phosphorylation. Combination of LY-2584702 with BMS-777607 was required to reduce both rpS6 and AXL phosphorylation. G, Mayo59 gliomaspheres were transfected with siNT or independent siS6K2 duplexes prior to treatment with 10 μmol/L LY-2584702 and/or 10 μmol/L BMS-777607 for 3 hours as indicated. Treatment with LY-2584702 reduced rpS6 phosphorylation and S6K2 genetic inactivation resulted in increased AXL, consistent with the requirement for combination with BMS-777607 in reducing GBM growth.