Prostaglandin E1 Inhibits GLI2 Amplification-Associated Activation of the Hedgehog Pathway and Drug Refractory Tumor Growth

Abstract

Aberrant activation of the Hedgehog (HH) signaling pathway underlines the initiation and progression of a multitude of cancers. The effectiveness of the leading drugs vismodegib (GDC-0449) and sonidegib (LDE225), both Smoothened (SMO) antagonists, is compromised by acquisition of mutations that alter pathway components, notably secondary mutations in SMO and amplification of GLI2, a transcriptional mediator at the end of the pathway. Pharmacologic blockade of GLI2 activity could ultimately overcome these diversified refractory mechanisms, which would also be effective in a broader spectrum of primary tumors than current SMO antagonists. To this end, we conducted a high-content screening directly analyzing the ciliary translocation of GLI2, a key event for GLI2 activation in HH signal transduction. Several prostaglandin compounds were shown to inhibit accumulation of GLI2 within the primary cilium (PC). In particular, prostaglandin E1 (PGE1), an FDA-approved drug, is a potent GLI2 antagonist that overcame resistance mechanisms of both SMO mutagenesis and GLI2 amplification. Consistent with a role in HH pathway regulation, EP4 receptor localized to the PC. Mechanistically, PGE1 inhibited HH signaling through the EP4 receptor, enhancing cAMP-PKA activity, which promoted phosphorylation and degradation of GLI2 via the ubiquitination pathway. PGE1 also effectively inhibited the growth of drug refractory human medulloblastoma xenografts. Together, these results identify PGE1 and other prostaglandins as potential templates for complementary therapeutic development to circumvent resistance to current generation SMO antagonists in use in the clinic. SIGNIFICANCE: These findings show that PGE1 exhibits pan-inhibition against multiple drug refractory activities for Hedgehog-targeted therapies and elicits significant antitumor effects in xenograft models of drug refractory human medulloblastoma mimicking GLI2 amplification.