Exploring vulnerabilities in malignant glioblastoma

Glioblastoma (GBM), the most common and malignant primary brain tumor in adults, is among the most difficult cancers to treat with a median survival of only 15 months. GBMs are highly complex tumors with several unique features explaining the lack of effective therapies: infiltrative growth of the tumor cells prevents complete surgical removal of the tumor, the blood-brain barrier (BBB) effectively inhibits drug delivery to the tumor site, and identification of subpopulations of glioma stem cells (GSCs) that are an important source of cellular heterogeneity and therapeutic resistance. Novel therapeutic approaches for treatment of these devastating tumors are urgently needed. In this study, we investigated the molecular mechanisms underlying tumor initiation, progression, and therapy resistance of malignant human GBM. We aimed at identifying vulnerabilities that could potentially provide novel therapeutic targets for treatment of GBM. We utilized patient-derived GSC cultures and patient-derived xenograft tumors as models to study GBM. In the first study, we demonstrated that mammary-derived growth inhibitor (MDGI), also known as heart-type fatty acid binding protein 3 (H-FABP/FABP3), was not only highly expressed but also played a significant role in GBM invasion. We identified a novel function for MDGI in maintaining the lysosomal membrane integrity. Unexpectedly, GBM cells were extremely vulnerable to silencing of MDGI expression. We demonstrated that MDGI silencing caused lysosomal membrane permeabilization (LMP), which is an alternative cell death pathway leading to irreversible apoptosis. LMP can be induced by pharmacological agents such as antihistamines. Interestingly, we demonstrated that treatment of patient-derived xenograft tumors with antihistamine clemastine effectively eradicated the invasive tumor cells and prolonged animal survival in a preclinical study in vivo. In the most invasive patient-derived GBM model, treatment with clemastine led to a complete eradication of the tumor. Our results encourage testing clemastine in a clinical trial of patients with GBM. In the second part of this study, we provided important insight into GSC plasticity driving tumorigenesis and therapy resistance of GBM. We identified a molecular mechanism where CD109 physically interacts with glycoprotein 130 (GP130) to regulate the interleukin-6/signal transducer and activator of transcription 3 (IL-6/STAT3) signaling pathway. We further demonstrated that the CD109/STAT3 axis was essential for the maintenance of stemness and plasticity of GSCs. When CD109 was silenced, GSCs differentiated into astrocytic-like cells and were unable to dedifferentiate into the stem-like state. Moreover, the CD109/STAT3 axis was needed for the tumorigenicity of patient-derived xenograft models in vivo. Importantly, genetic targeting of CD109 and pharmacologic inhibition of STAT3 both sensitized the GSCs to chemotherapy. These results suggest that therapeutic targeting of CD109/STAT3 axis in combination with chemotherapy might potentially increase the effect of chemotherapy in patients with GBM and help to overcome the therapy resistance. This study provides important insight into novel disease mechanisms with potential therapeutic implications for GBM patients.