Search

Perivascular macrophages (pvMs) are associated with cerebral vasculature and mediate brain drainage and immune regulation. Here, using reporter mouse models, whole brain and section immunofluorescence, flow cytometry, and single cell RNA sequencing, besides the Lyve1+F4/80+CD206+CX3CR1+ pvMs, we identify a CX3CR1– pvM population that shares phagocytic functions and location. Furthermore, the brain parenchyma vasculature mostly hosts Lyve1+MHCII– pvMs with low to intermediate CD45 expression. Using the double Cx3cr1GFP x Cx3cr1-Cre;RosatdT reporter mice for finer mapping of the lineages, we establish that CD45lowCX3CR1– pvMs are derived from CX3CR1+ precursors and require PU.1 during their ontogeny. In parallel, results from the Cxcr4-CreErt2;Rosa26tdT lineage tracing model support a bone marrow-independent replenishment of all Lyve1+ pvMs in the adult mouse brain. Lastly, flow cytometry and 3D immunofluorescence analysis uncover increased percentage of pvMs following photothrombotic induced stroke. Our results thus show that the parenchymal pvM population is more heterogenous than previously described, and includes a CD45low and CX3CR1– pvM population.

Glioblastoma patients have limited treatment options. Cancer stem-like cells (CSLC) contribute to glioblastoma invasiveness and repopulation; hence, they represent promising targets for novel therapies. BAL101553 is a prodrug of BAL27862, a novel microtubule-destabilizing agent inhibiting tumor cell proliferation through activation of the spindle assembly checkpoint, which is currently in phase I/II clinical development. Broad anticancer activity has been demonstrated against human cancer models, including tumors refractory to conventional treatments. We have shown that overexpression of microtubule + end-binding 1-protein (EB1) correlates with glioblastoma progression and poor survival. Here, we show that BAL27862 inhibits the growth of two glioblastoma CSLCs. As EB1 is overexpressed in the CSLC line GBM6, which displays a high tumorigenicity and infiltrative pattern of migration in vivo, we investigated drug activity on GBM6 according to EB1 expression. BAL27862 inhibited migration and colony formation at subcytotoxic concentrations in EB1-expressing control cells (GBM6-sh0) but only at cytotoxic concentrations in EB1-downregulated (GBM-shE1) cells. Three administrations of BAL101553 were sufficient to provoke an EB1-dependent survival benefit in tumor-bearing mice. Patterns of invasion and quantification of tumor cells in brain demonstrated that GBM6-sh0 cells were more invasive than GBM6-shEB1 cells, and that the antiproliferative and anti-invasive effects of BAL101553 were more potent in mice bearing control tumors than in EB1-downregulated tumors. This was associated with inhibition of stem cell properties in the GBM6-sh0 model. Finally, BAL27862 triggered astrocytic differentiation of GBM6 in an EB1-dependent manner. These results support the potential of BAL101553 for glioblastoma treatment, with EB1 expression as a predictive biomarker of response. Mol Cancer Ther; 15(11); 2740–9. ©2016 AACR.

Chemical structure of BAL27862 and BAL101553

Materials and methods: EB1 Western blot analysis GBM stem cells were lysed (lysis buffer: Tris 50 mM pH 8.0, NaCl 250 mM, Triton-X100 1%, SDS 0.1% with a cocktail of proteases and phosphatases inhibitors added freshly (all from Sigma-Aldrich). Thirty µg of total protein lysate were loaded onto a 12% SDS-PAGE gel. Nitrocellulose membrane (Bio-Rad laboratories, Marnes la Coquette, France) was blocked with 5% milk (powder) in Phosphate Buffer Saline (PBS) (Life technologies)-Tween (Sigma-Aldrich) pH 7.4 for 1 h and then incubated in PBS-Tween 0.1%-5% milk solution with mouse anti-EB1 antibody (clone 5, BD Biosciences, Le pont de Claix, France) (1/1000) and mouse α-tubulin (clone DM1A, Sigma Aldrich). After washing, membranes were incubated with anti-mouse peroxydase-conjugated secondary antibodies (Jackson Immunoresearch, Baltimore, USA) for 1 h. The bound antibodies were then detected using chemiluminescence detection kit (Millipore, Saint Quentin en Yvelines, France). Signals were recorded with G:BOX (Syngene/Ozyme, Saint Quentin en Yvelines, France ) and quantification was done with Image J software. Immunofluorescence analysis Indirect immunofluorescence was performed as previously described (15). Briefly, cells were fixed for 5 min with cold methanol (-20{degree sign}C), incubated with the anti-EB1 antibody 1/100 for 1 h and anti-mouse antibody Alexa 568 nm (Molecular Probes) 1/200; and FITC-coupled anti-α-tubulin antibody (clone DM1A; Sigma-Aldrich) 1/200 for 1 h at room temperature in the presence of DAPI for nuclear staining. For GFAP staining, cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton-X100 for 10 minutes. Primary antibody against GFAP (rabbit IgG, 10€‰Î¼g/ml, Dako) was incubated overnight at 4€‰{degree sign}C. Secondary antibody, Texas-red Goat anti-rabbit IgG (H+L) was purchased from Jackson Immunoresearch (Newmarket, UK), and was incubated for 1 h. Cells were analyzed using a Leica DM-IRBE microscope at a magnification of x100. Real-time quantitative PCR RNA samples were processed using a LightCycler 480 instrument (Roche Applied Science) and a LightCycler 480 SYBR Green I Master Mix (Roche Applied Science). Briefly, total DNA-free RNA (1 μg) was reverse-transcribed into cDNA using 1 μg of random hexamers and Superscript II reverse transcriptase as recommended by the manufacturer (Invitrogen Life Technologies, Cergy Pontoise, France). Measurements were performed in triplicate for each sample, and relative expression ratios of target gene transcripts (GFAP, TUBB3, CNP) and reference gene transcripts (18S, GAPDH, and ACTB) were calculated using qPCR efficiencies and cycle threshold (Ct) deviations of tumor and normal adult brain samples (control: Agilent Technologies) (25). RNA expression levels in GBM CSLCs were subsequently expressed as percentages compared to normal adult human brain samples corresponding to 100% of expression. Forward and reverse primers used for each gene are listed in Supplemental Table 1. Supplementary Table 1: Sequences of primers used in RT-qPCR Supplementary Figure 1: Chemical structure of BAL27862 and BAL101553 Supplementary Figure 2: Modulation of EB1 expression in GBM6 cells using siRNA or shRNA against EB1. (A) Western blot analysis of EB1 in GBM6 shEB1 clones (GBM6 shEB1, GBM6 shEB1#2), in GBM6 GFP shEB1 clone and in control clones (GBM6 wt, GBM6 sh0 and GBM6 GFP sh0). Ratios EB1/tubulin, relative to controls, from at least three independent experiments are presented under the blots. (B) Analysis of EB1 expression by Western blot of GBM6 wt or treated with siRNA against EB1 (GBM6 siEB1) or siRNA control (GBM6 si0). (C) Immunofluorescence staining of tubulin (green) and EB1 (red) in GBM6 wt, GBM6 si0 and GBM6 siEB1. Bar = 10 μm.

Sequences of primers used in RT-qPCR

International audience; Glioblastoma (GBM) is the most deadly type of malignant brain tumor, despite extensive molecular analyses of GBM cells. In recent years, the tumor microenvironment (TME) has been recognized as an important player and therapeutic target in GBM. However, there is a need for a full and integrated understanding of the different cellular and molecular components involved in the GBM TME and their interactions for the development of more efficient therapies. In this review, we provide a comprehensive report of the GBM TME, which assembles the contributions of physicians and translational researchers working on brain tumor pathology and therapy in France. We propose a holistic view of the subject by delineating the specific features of the GBM TME at the cellular, molecular, and therapeutic levels.

Simple Summary The Microtubule-associated protein Tau is expressed in different cancers; however, its role and prognostic value are still debated. In the present work, we evaluated the role of Tau in glioblastoma by down-regulating its expression in glioblastoma cells. We showed that Tau: (1) is required for tumor progression in nude mice; (2) is necessary for glioblastoma 3D cell organization, growth, and migration; and (3) regulates the PI3K/AKT signaling pathway. Abstract The Microtubule-Associated Protein Tau is expressed in several cancers, including low-grade gliomas and glioblastomas. We have previously shown that Tau is crucial for the 2D motility of several glioblastoma cell lines, including U87-MG cells. Using an RNA interference (shRNA), we tested if Tau contributed to glioblastoma in vivo tumorigenicity and analyzed its function in a 3D model of multicellular spheroids (MCS). Tau depletion significantly increased median mouse survival in an orthotopic glioblastoma xenograft model. This was accompanied by the inhibition of MCS growth and cell evasion, as well as decreased MCS compactness, implying N-cadherin mislocalization. Intracellular Signaling Array analysis revealed a defective activation of the PI3K/AKT pathway in Tau-depleted cells. Such a defect in PI3K/AKT signaling was responsible for reduced MCS growth and cell evasion, as demonstrated by the inhibition of the pathway in control MCS using LY294002 or Perifosine, which did not significantly affect Tau-depleted MCS. Finally, analysis of the glioblastoma TCGA dataset showed a positive correlation between the amount of phosphorylated Akt-Ser473 and the expression of MAPT RNA encoding Tau, underlining the relevance of our findings in glioblastoma disease. We suggest a role for Tau in glioblastoma by controlling 3D cell organization and functions via the PI3K/AKT signaling axis.

Perivascular macrophages (pvM) are closely associated with cerebral vasculature and play an essential role in drainage of the brain and regulation of the immune response. Here, using reporter mouse models and immunofluorescence on sections and whole brain, flow cytometry and single cell sequencing, we identify a Lyve1+ brain perivascular population lacking classical macrophage markers such as CD45 and Cx3cr1. We named the new non-conventional CD45 negative perivascular macrophages pvM2. These cells have a similar location, morphology and phagocytic function as conventional pvM. The pvM2 are not derived from hematopoietic stem cells, as they are negative in the VavtdT lineage tracing model. They increase in number after photothrombotic induced stroke established by flow cytometry and 3D immunofluorescence analysis. Since CD45 negative cells were typically excluded from macrophage studies, the presence of pvM2 has been previously missed and their role is of importance to assess in the brain disease models.

Perivascular macrophages (pvMs) are associated with cerebral vasculature and mediate brain drainage and immune regulation. Here, using reporter mouse models, whole brain and section immunofluorescence, flow cytometry, and single cell RNA sequencing, besides the Lyve1

International audience; Fifteen years after the establishment of the Stupp protocol as the standard of care to treat glioblastomas, no major clinical advances have been achieved and increasing patient's overall survival remains a challenge. Nevertheless, crucial molecular and cellular findings revealed the intra-tumoral and inter-tumoral complexities of these incurable brain tumors, and the essential role played by cells of the microenvironment in the lack of treatment efficacy. Taking this knowledge into account, fulfilling gaps between preclinical models and clinical samples is necessary to improve the successful rate of clinical trials. Since the beginning of the characterization of brain tumors initiated by Bailey and Cushing in the 1920s, several glioblastoma models have been developed and improved. In this review, we focused on the most widely used 3D human glioblastoma models, including spheroids, tumorospheres, organotypic slices, explants, tumoroids and glioblastoma-derived from cerebral organoids. We discuss their history, development and especially their usefulness.

International audience; Glioblastoma (GBM) are aggressive brain tumors with limited treatment options. Cancer stem-like cells (CSLCs) contribute to GBM invasiveness, representing promising targets. BAL101553, a prodrug of BAL27862, is a novel small molecule tubulin-binding agent, promoting tumor cell death through spindle assembly checkpoint activation, which is currently in Phase 1/2a in advanced solid tumor patients including GBM. This study aimed to evaluate long-term daily oral BAL101553 treatment of mice orthotopically grafted with GBM CSLCs (GBM6) according to EB1 expression-level, and to decipher its mechanism of action on GBM stem cells. Oral treatment with BAL101553 for 100 days provoked a large EB1 expression level-dependent survival benefit, together with a decrease in tumor growth and brain invasion. Formation of vascular structures by the fluorescent GBM6-GFP-sh0 cells, mimicking endothelial vascular networks, was observed in the brains of control grafted mice. Following BAL101553 treatment, vessels were no longer detectable, suggesting inhibition of the endothelial trans-differentiation of GBM stem cells. In vitro, BAL27862 treatment resulted in a switch to the endothelial-like phenotype of GBM6 towards an astrocytic phenotype. Moreover, the drug inhibited secretion of VEGF, thus preventing normal endothelial cell migration activated by CSLCs. The decrease in VEGF secretion was confirmed in a human GBM explant following drug treatment. Altogether, our data first confirm the potential of EB1 expression as a response-predictive biomarker of BAL101553 in GBM we previously published and add new insights in BAL101553 long-term action by counteracting CSLCs mediated tumor angiogenesis. Our results strongly support BAL101553 clinical studies in GBM patients.

BACKGROUND Glioblastoma (GBM) patients have limited treatment options. Cancer stem-like cells (CSLCs) contribute to GBM invasiveness, representing promising targets. BAL101553 is a novel small molecule tubulin-binding agent, promoting tumor cell death through spindle assembly checkpoint activation, which is currently in Phase 1/2a in advanced solid tumor patients including GBM. We have shown that overexpression of microtubule + end-binding 1-protein (EB1) correlates with GBM progression and poor survival. This study aimed to evaluate the effect of short-term IV and long-term daily oral BAL101553 treatment on mice orthotopically grafted with GBM CSLCs (GBM6) according to EB1 expression-level, and to decipher its mechanism of action on CSLCs. MATERIAL AND METHODS The A2B5+ GBM6 cell line was established from CSLCs isolated from a GBM patient. GBM6 cells display a mesenchymal phenotype with a high tumorigenicity and infiltrative pattern of migration. As EB1 is overexpressed in GBM6, we investigated drug activity on GBM6 according to EB1 expression, by using EB1-expressing cells (GBM6-GFP-sh0) and EB1-downregulated (GBM-GFP-shEB1) cells. BAL27862 is the active moiety of the prodrug BAL101553, which was developed to improve solubility. Two sets of animal experiments were performed with 4 groups randomized for GBM6-GFP-sh0- and GBM6-GFP-shEB1-bearing mice, receiving either BAL101553 (25 mg/kg) or vehicle intravenously on day 30, 33, and 36 after glioma implantation, or BAL101553 (30 mg/kg) or vehicle orally from day 35–135 post-grafting (5 days dosing/week). Survival and tumor invasion were analyzed. In vitro, BAL27862 activity was tested by using a cytotoxicity assay, and endothelial cell capillary-like tube formation assay. RESULTS Both treatment schedules provided an EB1-expression level-dependent survival benefit; with daily oral treatment this was almost one year longer than observed in vehicle-treated mice. Moreover, the anti-proliferative and anti-invasive effects of BAL101553 were more potent in mice bearing EB1-expressing tumors as compared to EB1-downregulated tumors. This was associated with inhibition of stem cell properties. Formation of vascular structures by the fluorescent GBM6-GFP-sh0 cells was observed in the brains of grafted mice. Drug treatment counteracted the formation of brain tumor vessels by inhibiting GBM6 trans-differentiation into tumor-derived endothelial cells, as well as VEGF secretion; thus, preventing normal endothelial cell migration activated by CSLCs. CONCLUSION BAL101553/BAL27862 triggers astrocytic differentiation and counteracts tumor angiogenesis by acting on CSLCs; thus, potentially adding to the direct anti-tumor cell effect. A high level of EB1 expression in CSLCs potentiates the drug’s effects, suggesting the potential of EB1 expression as a response-predictive biomarker for BAL101553 in GBM.

A2B5+ cells isolated from human glioblastomas exhibit cancer stem cell properties. The A2B5 epitope belongs to the sialoganglioside family and is synthetized by the ST8 alpha-N-acetyl-neuraminidase &alpha, 2,8-sialyltransferase 3 (ST8SIA3) enzyme. Glycolipids represent attractive targets for solid tumors, therefore, the aim of this study was to decipher A2B5 function in glioblastomas. To this end, we developed cell lines expressing various levels of A2B5 either by genetically manipulating ST8SIA3 or by using neuraminidase. The overexpression of ST8SIA3 in low-A2B5-expressing cells resulted in a dramatic increase of A2B5 immunoreactivity. ST8SIA3 overexpression increased cell proliferation, migration, and clonogenicity in vitro and tumor growth when cells were intracranially grafted. Conversely, lentiviral ST8SIA3 inactivation in low-A2B5-expressing cells resulted in reduced proliferation, migration, and clonogenicity in vitro and extended mouse survival. Furthermore, in the shST8SIA3 cells, we found an active apoptotic phenotype. In high-A2B5-expressing cancer stem cells, lentiviral delivery of shST8SIA3 stopped cell growth. Neuraminidase treatment, which modifies the A2B5 epitope, impaired cell survival, proliferation, self-renewal, and migration. Our findings prove the crucial role of the A2B5 epitope in the promotion of proliferation, migration, clonogenicity, and tumorigenesis, pointing at A2B5 as an attractive therapeutic target for glioblastomas.

In glioblastomas, apoptosis inhibitor proteins (IAPs) are involved in apoptotic and nonapoptotic processes. We previously showed that IAP inhibition induced a loss of stemness and glioblastoma stem cells differentiation by activating nuclear factor-κB under normoxic conditions. Hypoxia has been shown to modulate drug efficacy. Here, we investigated how IAPs participate in glioblastoma stem-like cell maintenance and fate under hypoxia. We showed that in a hypoxic environment, IAPs inhibition by GDC-0152, a small-molecule IAPs inhibitor, triggered stem-like cell apoptosis and decreased proliferation in four human glioblastoma cell lines. We set up a three-dimensional glioblastoma spheroid model in which time-of-flight secondary ion mass spectrometry analyses revealed a decrease in oxygen levels between the periphery and core. We observed low proliferative and apoptotic cells located close to the hypoxic core of the spheres and glial fibrillary acidic protein+ cells at their periphery. These oxygen-dependent GDC-0152 antitumoral effects have been confirmed on human glioblastoma explants. Notably, serine–threonine kinase activation analysis revealed that under hypoxic conditions, IAP inhibition activated ataxia telangiectasia and Rad3-related protein signaling. Our findings provide new insights into the dual mechanism of action of IAP inhibitors that depends on oxygen level and are relevant to their therapeutic application in tumors. Stem Cells 2019;37:731–742

In glioblastomas, apoptosis inhibitor proteins (IAPs) are involved in apoptotic and non-apoptotic processes. Here we used GDC-0152, a small molecule IAP inhibitor, to explore how IAPs participate in glioblastoma stem-like cell maintenance and fate under both hypoxic and normoxic environments. In hypoxia, IAPs inhibition triggered stem-like cells apoptosis and decreased proliferation in four human glioblastoma cell lines, whereas in normoxia it induced a loss of stemness and differentiation. In addition, we characterized a 3D glioblastoma spheroid model. By using MALDI images we validated that GDC-0152 penetrates in the entire sphere. TOF-SIMS analyses revealed an oxygen gradient correlated with spatial cellular heterogeneity with proliferative and apoptotic cells located close to the hypoxic core and GFAP+cells at the periphery. Notably, Serine-Threonine Kinases activation analysis revealed that oxygen level affects signaling pathways activated by GDC-0152. In hypoxia, IAPs inhibition activated ATR whereas in normoxia it activated NF-κB. Our data brings new mechanistic insights revealing the dual role of IAPs inhibitors like GDC-0152 that are relevant to their therapeutic application in tumors like glioblastomas.

Glioblastoma (GBM) is characterized by highly aggressive growth and invasive behavior. Due to the highly lethal nature of GBM, new therapies are urgently needed and repositioning of existing drugs is a promising approach. We have previously shown the activity of Proscillaridin A (ProA), a cardiac glycoside inhibitor of the Na(+)/K(+) ATPase (NKA) pump, against proliferation and migration of GBM cell lines. ProA inhibited tumor growth in vivo and increased mice survival after orthotopic grafting of GBM cells. This study aims to decipher the mechanism of action of ProA in GBM tumor and stem-like cells. ProA displayed cytotoxic activity on tumor and stem-like cells grown in 2D and 3D culture, but not on healthy cells as astrocytes or oligodendrocytes. Even at sub-cytotoxic concentration, ProA impaired cell migration and disturbed EB1 accumulation at microtubule (MT) plus-ends and MT dynamics instability. ProA activates GSK3β downstream of NKA inhibition, leading to EB1 phosphorylation on S155 and T166, EB1 comet length shortening and MT dynamics alteration, and finally inhibition of cell migration and cytotoxicity. Similar results were observed with digoxin. Therefore, we disclosed here a novel pathway by which ProA and digoxin modulate MT-governed functions in GBM tumor and stem-like cells. Altogether, our results support ProA and digoxin as potent candidates for drug repositioning in GBM.