Metixene is an incomplete autophagy inducer in preclinical models of metastatic cancer and brain metastases
Jawad Fares, … , Amy B. Heimberger, Maciej S. Lesniak
Jawad Fares, … , Amy B. Heimberger, Maciej S. Lesniak
Published October 17, 2023
Citation Information: J Clin Invest. 2023;133(24):e161142. https://doi.org/10.1172/JCI161142.
View: Text | PDF
Research Article Neuroscience Oncology

Metixene is an incomplete autophagy inducer in preclinical models of metastatic cancer and brain metastases

Abstract

A paucity of chemotherapeutic options for metastatic brain cancer limits patient survival and portends poor clinical outcomes. Using a CNS small-molecule inhibitor library of 320 agents known to be blood-brain barrier permeable and approved by the FDA, we interrogated breast cancer brain metastasis vulnerabilities to identify an effective agent. Metixene, an antiparkinsonian drug, was identified as a top therapeutic agent that was capable of decreasing cellular viability and inducing cell death across different metastatic breast cancer subtypes. This agent significantly reduced mammary tumor size in orthotopic xenograft assays and improved survival in an intracardiac model of multiorgan site metastases. Metixene further extended survival in mice bearing intracranial xenografts and in an intracarotid mouse model of multiple brain metastases. Functional analysis revealed that metixene induced incomplete autophagy through N-Myc downstream regulated 1 (NDRG1) phosphorylation, thereby leading to caspase-mediated apoptosis in both primary and brain-metastatic cells, regardless of cancer subtype or origin. CRISPR/Cas9 KO of NDRG1 led to autophagy completion and reversal of the metixene apoptotic effect. Metixene is a promising therapeutic agent against metastatic brain cancer, with minimal reported side effects in humans, which merits consideration for clinical translation.

Authors

Jawad Fares, Edgar Petrosyan, Deepak Kanojia, Crismita Dmello, Alex Cordero, Joseph T. Duffy, Ragini Yeeravalli, Mayurbhai H. Sahani, Peng Zhang, Aida Rashidi, Victor A. Arrieta, Ilya Ulasov, Atique U. Ahmed, Jason Miska, Irina V. Balyasnikova, C. David James, Adam M. Sonabend, Amy B. Heimberger, Maciej S. Lesniak

×

Figure 3

Metixene improves survival in preclinical models of metastatic brain cancer.

Options: View larger image (or click on image) Download as PowerPoint
Metixene improves survival in preclinical models of metastatic brain can...
(A) Experimental setup for stereotactic intracranial injection of BT-474Br cells into the brains of nude mice and subsequent treatment with metixene. (B) Kaplan-Meier curve showing survival of metixene-treated mice versus survival of control mice upon intracranial injection of BT-474Br cells. (C) Bioluminescence imaging of the 2 groups (control vs. treated) across time. (D) Immunohistochemical staining for cleaved caspase-3 in brain samples from control and metixene-treated mice bearing BT-474Br tumors (quantification of the stained area is shown in Supplemental Figure 4A). Scale bars: 250 μm. (E) Experimental setup for intracarotid artery injection of MDA-MB-231Br cells into nude mice and subsequent treatment with metixene. (F) Kaplan-Meier curve showing survival of metixene-treated mice versus controls upon intracarotid injection of MDA-MB-231Br cells. (G) Histological brain sections upon death of the control mice confirmed the growth of brain tumors as a result of intracarotid injection of MDA-MB-231Br cells. H&E staining of mouse brain shows metastatic tumors (black-framed box), the formation of multiple micrometastases (blue-framed box), and vascular co-option (red-framed box). Scale bars: 2.5 mm; 100 μm, and 500 μm. (H) Immunohistochemical staining for cleaved caspase-3 in brain samples from control and metixene-treated mice bearing MDA-MB-231Br tumors (quantification of the stained area is shown in Supplemental Figure 4B). Scale bars: 250 μm. Survival curves were compared using a log-rank test. The linear mixed model was fitted for C, and the differences between the 2 groups for each time point were calculated using the least-squares means method and adjusted by Šidák’s method. *P < 0.05 and ***P < 0.001.