Antitumor activity of novel pyrazole-based small molecular inhibitors of the STAT3 pathway in patient derived high grade glioma cells

Abstract

Abnormal activation of signal transducer and activator of transcription 3 (STAT3) transcription factor has been observed in many human cancers with roles in tumor initiation, progression, drug resistance, angiogenesis and immunosuppression. STAT3 is constitutively activated in a variety of cancers including adult high grade gliomas (aHGGs) such as glioblastoma (GBM), and pediatric high grade gliomas (pHGG). Inhibiting STAT3 is a promising target-specific chemotherapeutic strategy for tumors with aberrant STAT3 signaling. Here we investigated the antitumor effects of novel pyrazole-based STAT3 pathway inhibitors named MNS1 (Mayo Neurosurgery 1) in both pediatric and adult HGG tumor cells. MNS1 compounds selectively decreased cell viability and proliferation in patient-derived HGG cells with minimal toxicity on normal human astrocytes. These inhibitors selectively blocked IL-6-induced STAT3 phosphorylation and nuclear localization of pSTAT3 without affecting other signaling molecules including Akt, STAT1, JAK2 or ERK1/2 phosphorylation. Functional analysis showed that MNS1 compounds induced apoptosis and decrease tumor migration. The anti-tumor effects extended into a murine pHGG (diffuse intrinsic pontine glioma) patient derived xenograft, and systemic toxicity was not evident during dose escalation in mice. These results support further development of STAT3 inhibitors for both pediatric and adult HGG.

Fig 1. Pyrazole-based MNS-1 inhibitors block STAT3 phosphorylation.

(a) Diagram of STAT3 pathway; activation is by cytokines, most commonly IL-6, and growth factor receptors, most commonly gp130. Phosphorylated STAT3 forms a reciprocal homodimer that translocates into the nucleus where it interacts with DNA and activates gene transcription. The activated form is typically inhibited by Protein inhibitor of activated STAT (PIAS) and Protein-tyrosine Phosphatase (PTPase). (b) Chemical structure of MNS1-Leu. (c) Chemical structure of MNS1-MV. (d) Western blot of pSTAT3 (Y705) in adult GBM and pHGG patient derived cell lines at baseline compared to non-tumor brain control (removed for epilepsy). (e) Western blot analysis of IL-6 (100 ng/mL) stimulation of STAT3 phosphorylation in dBT114 for 5–30 min. (f) Western blot analysis on the effect of IL-6-induced STAT3 phosphorylation in dBT114 in presence of MNS1-MV and MNS-Leu in various concentrations after a 10 min exposure to IL-6 stimulation.

Fig 2. MNS1 compounds selectively inhibit the viability of adult GBM and pediatric glioma tumor cells.

(a) Viability of control Normal Human Astrocytes treated with various concentrations of MNS1 compounds for 72 h show no toxicity up to 25 μM. (b) Dose-response curve of MNS1 analogs in MC-BT114. Dose-response curve in MC-BT114 (aGBM), MC-PED8 (H3K27M), SF7761 (H3K27M) tested with (c) MNS1-MV or (d) MNS1-Leu after 72 h treatment. Values are the means ± S.E.M (error bars) of triplicate experiments.

Fig 3. MNS1 compounds effectively inhibit STAT3 activation and nuclear translocation in both adult GBM and pediatric HGG cells.

(a) Luciferase activity driven by a STAT3-responsive transcriptional element on an expression vector showed inhibition of STAT3 against IL-6 in presence of MNS1-MV or MNS1-Leu. Values are the means ± S.E.M (error bars) of triplicate experiments. One-way ANOVA was used to calculate statistics. * p<0.05. Subcellular fractionation analysis on STAT3 translocation into the nucleus upon IL-6 stimulation (100 ng/ml) in dBT114 or dPED17 in presence of (b) 25 μM of MNS1-MV or (c) 10 μM of MNS1-Leu. N denotes for nucleus fraction. Cy denotes for cytosol fraction. All experiments were repeated independently three times. (d) Confocal imaging of pSTAT3 in dBT114 in presence of MNS1-MV. Treatment of MNS1 analogs did not affect activation of (e) JAK2 and Akt upon G-CSF stimulation or (f) STAT1 and ERK1/2 activation upon IL-6 stimulation.

Fig 4. MNS1 compounds induce apoptosis and inhibit migration in adult GBM cells.

(a) Flow cytometry analysis of annexin V/PI staining in dBT114 cells showed dose dependent decrease of cell viability, via apoptosis with treatment of MNS1-Leu for 24h. Values are the means ± S.E.M (error bars) of triplicate experiments. (b) Western blot analysis showed decreased expression of Bcl-2 with increased dose of MNS1-MV or MNS1-Leu. Total Erk1/2 was used as loading control. Images of (c) dBT116 (aHGG) migration assay and (d) quantification after MNS1-MV treatment. Images of (e) dBT116 migration assay and (f) quantification post MNS1-Leu treatment. Values are the means ± S.E.M (error bars) of triplicate experiments. 8 fields were imaged at 20X and counted in each experimental group.

Fig 5. MNS1-Leu slows pHGG growth in vivo and reduces STAT3 activation.

Dose escalation study with MNS1-Leu in PDX model of pHGG (DIPG17) starting at 10 mg/kg and increasing to 30 mg/kg over 21 days. (a) BLI imaging in PDX model of vehicle control vs. MNS1-Leu at day 21. (b) Caliper measurements of flank “DIPGXVII” xenograft tumors showed significant increase in tumor size when treated by vehicle, but not when treated by MNS1-Leu. (c) Similarly, bioluminescence of these tumors also showed significantly increased flux in those treated by vehicle, with no difference in those treated by MNS1-Leu. Values are the means ± S.E.M (error bars) of triplicate experiments. (d) Immunohistochemistry staining for pSTAT3 in these tumors demonstrated a decrease in pSTAT3 expression in those treated by MNS1-Leu compared to those treated by vehicle.